View Full Version : Guide to 500k+ scores through Huge map domination on Immortal difficulty


VirusMonster
Jan 28, 2008, 03:38 AM
Hello Civfanatics =) I am very happy for achieving the first and, between February 10th and March 25th of 2008, the only score over 500K in the main HOF tables. Between these dates, my submission also remained as the earliest finish date on Huge map size at Immortal difficulty. The game was played with the :bts: 3.13 patch and against the better :bts: AI, and here is the Hall of Fame link for the saves:

http://hof.civfanatics.net/civ4/game_info.php?entryID=10520

And more saves: (to view them, you need to install the 3.13.001 HOF mod, download here (http://hof.civfanatics.net/civ4/mod.php#bts_download))

169710
169601
169602
169603
169604

The strategy for this high scoring game can be summarized as a beautifully executed Praetorian rush resulting in a halt of economy, followed by a quick recovery and continuous warmongering. I hope this detailed writeup will be a solid guide for anyone trying to achive highest CivIV scores and earliest finish dates.

Along this comprehensive and pretty long writeup, you'll find my point of view to the unending debates of CivIV. Those educational discussions would be:


Dynamics of a powerful hybrid economy

detailed analysis of the hybrid economy composition from:

Food Economy :food: (FE) (with clear definition)
Specialist Economy :gp: (SE) (effective GP usage for an early domination victory)
Cottage Economy :commerce: (CE) (farms to cottages ratio in a hybrid economy)
TradeRoute Economy :traderoute: (TRE) (misleading FE effect explained)
ResourceTrade Economy :gold: (RTE) (warmonger's friend)
analysis of the following variables of economic growth at 1000BC, 500BC, 1AD, 500AD, and 1000AD:
empire size and population
civic upkeep
city maintenance
army expenses
ResourceTrade Economy
TradeRoute Economy

financial vs. organized trait comparison at 1000BC, 500BC, 1AD, 500AD, and 1000AD
detailed comparison between Espionage Economy :espionage: (EE) and Manual Research

technology cost explained
espionage base cost of technology stealing explained
city revolt base cost explained
espionage mission cost modifiers explained
espionage point spending modifier explained
arithmetic of the espionage mission cost modifiers explained
religious modifier arithmetic explained
spy detection analysis with links to game mechanic articles
mission success rate analysis with links to game mechanic articles
tips for low spy casualties
spy production costs for the average Medieval scenario
spy maintenance costs for the average Medieval scenario
total mission overhead
luck factor in EE
:commerce: to :science: conversion rates for EE and Manual Research
Bureaucracy effect to EE and Manual Research compared
effect of larger map sizes on Bureaucracy economy
Vassalage+Organized+EE synergy
summary comparison of conversion rates between EE and Manual Research
most effective way for generating :espionage:
pure EE vs. hybrid EE

the strongest recovery method: PeaceTreaty Economy (PTE) (AI gives tech for peace treaty)
comparison of warmonger civics:
Police State vs. Representation (for sustaining research during a soft war)
Bureaucracy vs. Vassalage (impact of map size and synergy with the Organized trait)
Theocracy vs. Pacifism (to fuel golden ages through GPs for increased war production)
FreeMarket vs. Mercantilism

most significant components of speedy recovery after economic crash
how to complete 3-4 Great Wonders despite early warmongering, a must read :-)
smart way of poprushing 2 units instead of 1 without getting an extra unhappiness penalty
when to shutdown research for the initial rush and final warmongering
which tiles to work after each warmongering phase


Diplomacy against the BTS AI:


forcing an expensive tech trade out of your peace treaties
getting that expensive tech trade the AI thinks you have nothing valuable to pay for
researching Guilds and Banking before the AIs do
sustaining the peace with your friends while warmongering
picking your next war
responding to AI demands and tech trade requests


Combat tactics against the better BTS 3.13 AI


how to capture the AI cities quickest with very high win/loss ratios
stack composition
how to promote the siege units
observations on healing
observations on quests
Great General usage


Tips for scoring high in the HOF tables


game and map settings
leader choice for Immortal and Deity difficulties
worker stealing revisited
subtleties of the perfect Praetorian rush
early game strategy
the ideal starting location
most significant tip for early finish dates
late game tips for delaying the domination limit cross



I will hit the points above as I write out the game chronologically. For example, you will have to wait until mid-game for comparison of various economies. Additional short observations are scattered along the writeup as well.

I also send my gratitudes from here to all contributors of the War Academy, especially the Game Mechanics section. They helped me to understand [civ4] in a much subtler way. Special thanks goes to OTAKUjbski for helping me organize this article.

VirusMonster
Jan 28, 2008, 03:38 AM
A few words on game and map settings:

The following are my preferences for aiming highest score and earliest finish date in HOF:


Pangea map type enables earlier domination victory dates compared to multicontinental map types, since transporting your army to the other continent and later sending reinforcements take significantly more time than conquering a single big continent. Furthermore, maps with more than one main continent usually end up splitting luxury resources, and you might have to wait a long, long time to connect that rare luxury resource in the other continent. Even Terra maps have this problem, hence map types with a single landmass are probably best for competitive HOF play at Immortal difficulty. On the other hand, for Deity difficulty, Pangea might not be the easiest map choice according to this post (http://forums.civfanatics.com/showpost.php?p=6361388&postcount=40) by ABigCivFan.
Huge land size chosen to allow for maximum population and land limits. While on smaller mapsizes you can get a higher score due to an earlier finish date, on larger map sizes, your base score from land and population will be higher. For highest scores, huge map size is probably the best choice.
Temperate climate chosen so I can get all health and happiness resources on the map. For example, tropical climate setting does not put any silver resource to the map. How do I know? Regenerated some starts and opened up in the world editor.
High Sea levels, so hopefully, less land will make domination attempt not more difficult than it already is. On the other hand, low sea levels would probably have made the final base score higher.
17 Enemy AIs, because it is more fun and the maximum number the HOF allows. It also gives a little less land to each AI at the start to settle, but has the danger of creating a very difficult early game. For example, in this game, I was sandwiched between Montezuma and Genghis Khan. I was fighting against 4 AI opponents simultaneously in the 2000-1000BC era. If you can survive the BC era and grow larger than your closest competitor, things get little bit easier, but diplomacy still remains a mess.
Cylindrical, I don't think it makes much difference compared to Torodial.
Standard resources, because HOF does not allow balanced. The Pangea map script was usually very generous as well :-)
Marathon speed to slowdown units becoming obsolete and Ancient era start
Locked modified assets
No barbarians, because I want my stolen workes to arrive home safely. I could probably escort the workers with warriors, but I did not want anymore problems than the 17 I already had. Most other Immortal and Deity HOF top scores also have barbarians turned off, and I didn't want to break the trend. A praetorian rush, however, can definitely handle normal barbarians and favors such setting, because barbarians will help your Praetorians promote to CityRaiderII before the initial rush.
Did I pick opponents? Well, having lost a very long game against a protective AI that outteched and owned me after I had achieved 40% of total land mass really discourages me to play against Protective AIs. Thus, I choose my 17 AI opponents to be not protective. I also avoided financial for some reason. Not sure why however, probably because of the whole propaganda about Financial being the best trait. I might put those saves up too, so you can understand my frusturation. It was a stalemate and I could not advance my land to the domination limit, and the harder I sent stacks against his cities, the more I lost :-) I am laughing now, but it was a very valuable lesson.
Random events are turned on for more fun :-)


Which leader did I choose?

The mighty Julius Caesar of of course! =) I tried both Persians leaders and Hannibal as well, but decided Romans would be the best choice. Victoria and MansaMusa were other candidates. Suleiman also looks like a very strong leader for Immortal and Deity difficulties.

I want to try Charlamange next time. I think everything about him is overpowered. I don't understand why people think he is so bad. His UB is the best in the game, and his UU is really strong against any unit of its era, creating great synergy with Trebuchets and the +1 movement bonus Engineering brings. Imperialistic helps for early game fast expanding due to better chopping and later has the potential to create an extremely strong army. In conclusion, without even counting the effects of the Protective trait, Charlamange is a very solid leader.

Worker stealing revisited

One downside of Charlamange, however, is that he starts with Hunting. Initial warrior will be replaced by a scout, and he can't steal workers as early as a leader starting without Hunting. Worker stealing is a great way to boost your economy in the ancient era. In the 500K game, I stole 3 workers, equaling 360 hammers on marathon speed and roughly 40 turns of population growth. That much hammers and food are the price for 5 Praetorians. 5 Praetorians usually compose half of the initial rush army.

A leader starting with Hunting must spend 6-10 turns on marathon speed for producing the first warrior and might be less successful in worker stealing, because AI capital's cultural borders would most likely grow before the initial warrior reaches a possible target. Once cultural borders grow, worker stealing becomes somewhat luck dependent and might come delayed, because worker's location cannot be predicted easily with a single warrior. Delaying or even failing your worker stealing plan would be some huge setback for early economy.

In that case, I suggest you locate an unimproved resource inside enemy cultural borders and camp your warrior next to it until the enemy worker arrives. Even then, some AIs, mostly the more aggresive ones, could guard their workers with archers.

Did I regenerate the start?

I even used HOF MapFinder utility to find the start I liked. I did, because I am competing for the best score in HOF and don't plan on torturing myself with an average start. The fact I regenerated the start does not mean I fail defeating average starts regularly, but it probably means I would not score as high and maybe fail seldomly. I do agree, however, that beating an average start gives a different kind of pleasure :-) And I really enjoy reading writeups where game was won despite an average start or by an underdog leader at higher difficulties and appreciate the author's skills.

Subtleties of the perfect Praetorian rush:

To master the Praetorian rush took much trial&error. Despite amazingly advantageous starting positions, I lost many games in the early game and learned from my mistakes. For example, I figured what the optimal city size should be before first settler production (4) and what the optimal number of cities should be before the first Praetorian rush (4-5 for a huge map).

Once, while delaying my praetorian rush and building monuments, granaries and barracks in my cities, I got war declared by around 10 AIs simultaneously, because my power was ridicilously low. Thus, I decided to delay building granaries until after the inital Praetorian rush. Built Monuments, followed by Barracks and poprushed extensively. Monument should be always poprushed first once your city grows size 2.

Another significant mistake was settling my 2nd city on commerce or hammer rich locations. To poprush the inital rush army fastest, I figured settling on food rich locations was usually much better. The enemy capitals would serve as commerce&hammer rich locations anyways :-) Knowing where to settle your cities is one of the most important skills in Civ4, and my advice is simple: Food, Food, and again Food.

I know it sounds simple, but learning to leave enough units to defend my cities took some time as well. I must have been producing an unsufficient number of units in my test games, because I was either not able to successfully rush the neighbouring AI or not able to defend my cities against some other AI during this initial rush period. Eventually in the 500K game, I survived warring simultaneously against 4AIs(!!) during the 2000-1000BC era.

Additionally, learning the new AI tricks took some time: AI is definitely tougher in BTS. Most significant improvement is the effective usage of AI poprushing when you are about to capture one of their cities. AI also won't middlessly attack your well defended city, but can camp outside some forest near your city. This situation can create a stalemate for a long period unless you want to take the risk and attack the stack despite the forest defense bonus.

And the economy! There are too many points to cover about a strong economy. Don't worry, I am not hiding any secrets, you will find all of them inside the writeup. Can you believe I lost a game after reaching 44% of total lands (51% is domination limit) mainly because of my overexpansion? Well, I have heard it many times, overexpansion kills any empire, but experiencing it firsthand after 10 hours of Civ4 gaming made me rethink my overall strategy. I did not lose that game financially, but rather technologically. You might sometimes want to delay your expansion.

Early game strategy:

My starting strategy was as follows: Regenarate the start and always settle on place until 2 gem resources appear in the 20 squares of the capital radius. Gold gives 1 more commerce, but gem is better, because 2 food and 1 hammers from a grassland gem instead of 3 hammers from a gold hill allow your city to grow faster.

Most likely, actually almost always, such start would also include 1 food resource in one of the 20 tiles around your capital.

Next, build 2 warriors and send the initial warrior to worker stealing, preferably from one of the non-aggressive AIs. Do not steal more than 3 workers; I have lost many games due to archer rush following a worker steal. Welcome the new BTS AI! =) Rushing with archers =) I have no barbarians turned on just so I don't have to escort workers back. With so many AIs on the map, the barbarians will not be able to find a spot to spawn anyways.

Initially, I tried researching IronWorking right after BronzeWorking and settled my 2nd, 3rd and 4th cities once I know where the Iron are. My most successful attempts were however games where I immediately settled the 2nd city on a food rich location.

In the 500k game, I researched Wheel after BronzeWorking, followed by IronWorking. Capital grew size 4 before working on the 2nd luxury resource. After BW->Roads->IW, I researched Mystisicm to build monuments at my food rich cities, followed by Agriculture and/or Animal Husbandry.

Pottery and Writing should also be discovered before the initial praetorian rush, because once the research halts, cottages and libraries must be built for a speedy recovery.

Here is the starting screenshot:

168435

The power of 2 gem or 2 gold starts is obvious. While your palace can only produce 8 commerce per turn, a capital with gems can produce +5 while a city with gold can produce +6 commerce. Having 2 gold resources would effectively increase your research rate for the expensive IW tech by %150.

2 Gems would increase your reseach by %125. Don't make the mistake of working the gold mine too early however, growth is just as important as research. Grow size 4 before you work on the gold mine. I am telling this with experience after having lost a 4 goldhills start =)

With gem, it usually does not matter, because you can find gems usually on grassland giving you 2 food, 1 food only if it is grasslandhill, but still better than the 0 food gold hill.

In the 500k game, I had a lucky 1 gem, 1 gold start. 1 gem, 1 gold start could be even better than a 2 gem start because of the extra happiness resource. Rome's radius also contained corn, and copper showed up on grassland after BW.

Here are some screenshots for my start, and the actual writeup begins:

3 workers stolen at 3250BC, 2 are working the mines, and the 3rd one is on the way:

168437

Chopping and producing the first settler with Rome size 4 at 2875BC (75th turn):

168438

Here are my city placement choices: Notice that all city locations marked 1, 2, and 3 have 2 food resources. Antium was settled at 2785BC, Cumae at 2680, and Neapolis at 2370. I was not sure about founding a 5th city, but had an extra settler and settled Ravenna at 2230.

168439

Locations of iron at 2400BC:

168440

Good that 2 of my already built cities will have iron in their 20 square radius. When you settle right next to a food resource before you have discovered IW, don't fear about settling on the Iron, since Iron and food resources never appear in adjacent tiles.

Observations on quests:

I got a badly timed quest at 2190BC, but can't complain =) Just as 1 turn was left to hook up the road to the already mined iron resource, the Greed quest shows up telling I should capture Istanbul for free units. Well, Istanbul was already on my way, since I was planning to expand toward the oceans.

Later in the game, I also got the Best Defense quest 2-3 turns before I had finished researching Economics. The Best Defense quest asked for completion of 10 Castles, well can't build castles anymore with Economics. If I completed the mission, I could choose from:
- the not extremely important CityGarrison promotion to all my melee units or
- the very nice diplomatic bonus of gaining +3 Attitude with all AIs I have met or
- the super +25 Espionage points bonus (roughly 6 free spy specialists) for the Great Wall
At that stage, my empire needed unit production badly, and I decided to skip this mission. I suspect poor quest timing was another reason for my refusal.

The ultimate quest for Praetorian rush would have been the Elite Swords mission. It asks for gathering around 10 Swordsman units (in Roman case 10 Praetorians), and as reward, you can choose all swordsman units produced in your capital receive the CityRaider I promotion. If you settle your first GG at the capital, then you can pump CRIII Praetorians very, very early in the game.

1920BC: Discovered Hunting and Pottery through tech trade

1700BC: Discovered Archery and Writing through tech trade

1670BC: War starts vs. Mehmed. Praetorians captured Istanbul by 1650BC, Edirne by 1620BC, and razed the last Ottoman city Ankara by 1600BC, creating a Great General during Edirne siege. I always settle the first general at capital to train CityRaiderII promoted Praeteorians right from the start.

1650BC: After capturing Istanbul, I receive 6 zero-experienced praetorians due to completion of the Greed quest. They create a small jump in my power graph.

Same turn, Asoka declares war on me. I am not really afraid of him, his empire is more than 10 tiles far in the south, and being usually the peaceful AI, he is unlikely to send a huge stack. Nevertheless, it is 1 more enemy to be taken care of.

1630BC: Pyramids built and looking at info screen: it belongs to the Mongols, they should be my first target after the Ottomans. I figured either Mehmed or Genghis would finish the Pyramids first, since both of them had Stone nearby.

1620BC: Genghis adopts Police State, big problems very soon :trouble:

1610BC: Bismarck declares war.

1560BC: Mongols declare war on me before I could on them :-) The battle begins.

1540BC: Now Montezuma declares war. I thought I lost the game at this stage; I am in war vs 4 AIs. 2 of them are aggressive leaders, and Bismark isn't usually friendly either. I would be very dissappointed that if this super start ended up with a quick loss :(

1530BC: Fortunately, Asoka makes peace with me. Unlike Bismark, Montezuma, and Genghis, he is usually willing to discuss peace.

1530-1510BC: Bismarck's stack parks in the forest outside my northern city, Neapolis. I need larger numbers in capturing the Mongols, and these 3 defending Praetorians at Neapolis would have been much help in capturing Karakorum. Similar situation is happening in my southern city where 3 Praetorians are guarding against Asoka and possibly Montezuma.

My outpost city, Ravenna, near Aztec lands has just reached population 2, and I am poprushing 2 archers (90 hammers from rush + 10 hammers = 50 per archer x 2) These 2 archers are going to replace the southern city, Antium, defenses, and hopefully, the 3 Praetorians there will be sent toward Mongols.

VirusMonster
Jan 28, 2008, 03:40 AM
1510BC: Next target is Pyramids, but Mongols won't fall easy. Look at his city on my way toward his capital. :trouble: Walls already?! 4 units inside the city and 4 more on the way? Shock promoted Axeman? :scared:

168441

My research has halted as well due to army expenses and the 2 new cities, Istanbul and Edirne.

1500BC: Montezuma finally captures Ravenna, and my empire stops losing gold at 0% research.

1480-70BC: Siege of tough Karakorum, city has walls up (+50%) and fortified (+25%) units.

8 Praetorians + 1 Axeman vs. 5 Axemen, 2 Archers and 1 Spearman at +75% defense

168445

Battle lasted 2 turns: It was so exciting, because I almost did not make it =) During the first turn of attack, I lost 2 Praetorians, followed by killing an axeman, followed by another 2 Praetorian loss. At this point, I had 4 Praetorians and 1 axeman left vs. 4 Axemen, 2 Archers, and 1 Spearman and was feeling very bad. How could I have wasted such a wonderful start at this Mongol city so stupidly? :gripe: Anyway, now my remaining Praetorians attack.

Bum :) Can't always be unlucky, I guess RNG is fair. I kill 2 archers, 1 spearman and one of the wounded axeman. On next turn, Karakorum has 3 Axemen all of which have been wounded during previous round. Fortunately, Genghis did not poprush any more axeman in previous round.

My wounded units were 4 Praetorians and 1 Axeman. However, since his units were fortified they all healed 20 HP (base healing inside cities for fortified units) Luckily, my units were around 3 exp before the battle and gained a 2nd promotion. A promotion boosts %50 lost hp immediately. CityRaider promotions also further negate the effects of +75% city defenses down to +55% and +30% respectively.

Now, my turn of attack with the following odds of battle:
73.6%(win), 27%(loss), 39%(win), 87%(win) and captured Karakorum :dance: I felt very lucky at this point :rockon: not only for controlling the Pyramids, but also for not having lost my entire army at the siege of Karakorum. 4 out of 9 survived, that is roughly 50%, which is great against a city that has walls up and 7 fortified units. I could have arrived with more units if I wasn't in war with Bismark and Montezuma as well.

Gaining control of the Pyramids, I switch to Police State civic and start producing Praetorians at +25% rate.

I continue eliminating the remaining Mongol cities one after another.

1420BC: My second general will be settled in Karakorum. Karakorum is going to be a strong production city, possibly my second military center. It has much hills and good food resources.

1360BC: Genghis gets his first general. He usually settles the generals, so I must be careful not to raze anymore of his cities.

1330BC: Peace with Bismarck. I grew weary of the stalemate of his stack parking in my forests, so I decided to attack him at all costs. His stack took some casualities, and peace agreed soon after.

Peter declared war on Montezuma btw, which should help me sign a peace treaty with Montezuma faster.

1300BC: I am paying 16 gold to units within my borders. 16 gold roughly corresponds to 25% research rate. When my army stack moves outside my cultural borders, I pay an additional 16. 32 gold corresponds to roughly 50% research rate. Talk about pre-1000BC warmongering, it is all about army maintenance.

Civic upkeep costs 10 gold (should be 20, but I am organized -50%), and city maintenance 27. I am running 0% research rate with no profits. Empire population size is 27, but it should grow to 50 pretty soon. Keep in mind I am not working any cottages at this point, and no library specialists are assigned. Everything was set on Praetorian production and poprushing. As some might call, full scale food economy.

Dynamics of a powerful hybrid economy

"Food economy" (FE) term can be misleading as its definiton is not entirely clear. To denote excessive use of poprushing for war and infrastructure production, the term "slavery economy", "poprushing economy", or "whip economy" could be used. Specialist economy, on the other hand, does not use poprushing as much and favors Caste System over Slavery, yet probably still works as many farms and food resources as a "slavery economy", because excess food is required to assign those specialists. Then, food economy becomes, according to my dictionary :-), the compositon of a strong "slavery economy" with a soft specialist economy. According to this definition, during my first Praetorian rush, I was running a full scale "slavery economy" with almost no research. Then, after assigning some scientists and later spy specialists, yet still depending largely on poprushing for infrastructure production, I switched to a real food economy.

A soft "specialist economy" under Slavery, in other words, a FE, has various advantages. Distributing early GP generation and Wonder production over a couple of cities and running Slavery instead of Caste System allows for both continuous high-output production and earlier GPs than a single GP farm would produce. The early GPs and their variety add significant flexibility to your gameplan.

Academies and infiltration missions are perfect research boosts in any game. On a huge map where you already control most of the earth due to domination victory plans, saving the less useful early GPs for 2-3 Golden Ages is probably the strongest idea. On marathon speed and while controlling the Mausoleum of Maussollos, 3 consequitive Golden Ages correspond to 72(!) turns, roughly 360 (!) years, of super high production and commerce yields. Holy Shrines also gain importance as the map size gets larger. On smaller map types, lightbulbing becomes more advantageous to capitalize on early military tech lead.

Throughout this 500k game, I worked mainly farms rather than cottages, because the speed I acquired new health and happiness resources clearly outperformed the speed my cities grew to the health and happiness cap. Additionally, since my almost continuous war campaign depended greatly on Slavery usage, farms had to be worked for faster city regrowth. Capturing Pyramids very early in the game and switching to Representation soon after also affected my decision to work a few extra farms to support those 2-3 specialists per city. Finally, since each population unit contributes to the final score, I wanted to maximize the empire population as early as possible.

Obviously during a less successful war campaign or a more peaceful game with lower health and happiness caps, working a higher number of cottages would have been the more logical choice. Cottages always compose some part of the total economy, because high happiness and health caps are not achieved instantly. Especially after you reach an economic halt due to warmorgering, it is natural to switch to work the cottage tiles instead of the farms. So when I use the term "FE", don't think my definition is strict; cottage economy could be and should be composing, depending on player's taste, 0 to 40% of the total economy in the early-mid game. Post Liberalism and/or Democracy, this precentage should increase even further, and since Slavery gets outdated after Emancipation, it would be wiser to call the overall economy mainly a "Cottage Economy" only first at that stage.

Let me also add that traderoute economy (TRE) has little to do with how populated your cities are, because there is no competition among empires for foreign traderoutes with high base profits. Consequently, the FE and TRE can be thought as seperate concepts.

On the other hand, a minor boost of FE to TRE occurs, because each population unit above 10 increases the base traderoute profit in your cities by 5%. This minor boost is negligible, because even if working all farms with a super high happiness cap could theoretically result in higher average city population throughout your empire, the traderoute effect of this excess population in FE compared to any other economy type is rather small. For example:

+5% per extra population * on average 5 more citizens in a FE than a CE = 25% higher base traderoute profits. With 3-4 traderoutes per city and base traderoute profits around 2.0-3.0 at middle game, excess traderoute profits per city in a FE correspond roughly to (2.0-3.0)*(3-4)*0.25=1.5-3.0 commerce.

In conclusion, the effect of FE to traderoute profits is negligible. More on TRE dynamics here.

------------------

After the conquest of Karakurum, my cities have started building granaries followed by libraries, while the capital kept producing CRII praetorians. All of my strong production cities had access to 2 food resources. For example, Antium at size 4 gave 4 food surplus per turn. It requires 90 food to grow size 5: 90/4=23 turns. The unhappiness penalty caused by poprushing lasts 30 turns on marathon speed; therefore, Antium could safely poprush every 30 turns, still stay at 4 population, and grow slightly. This growth dynamic is another reason why you can delay building granaries before the initial Praetorian rush, because obviously even if you built the Granaries and grew while still poprushing every 30 turns, your growth will be most likely limited by the happiness cap.

Cumae was my strongest city with 8 excess food per turn. It was the first city to finish the granary and worked on the library next.

1260BC: Mongols capture Edirne, it had only 1 Praetorian defending. I was not careful.

1230BC: I capture Mongol city of Beshbalik. It has a settled GreatGeneal =) but it would have been much better if the settled general was in Karakorum.

1210BC: I recapture Edirne. Mongols down to a single city. Now, look what is happening next turn =)

1200BC: I sign a peace treaty with Mongols and get =) guess what =) Some might argue that no AI would enter such a trade, but I guess the warmonger Genghis does not put enough importance on techs. Look at my war weariness toward him.

168450

He gives Alphabet and Masonry for a 10-turns peace treaty. :woohoo: My research was on a halt for centuries, I still have no single library built and now this extremely generous peace offer. How could I resist?

Before the treaty, I was the only civilization that had no knowledge of Alphabet and Masonry, thus, another empire might also have given these techs for free. Let's not put all fault on Genghis, he was merely giving away something that every other AI had to a tech-poor human player =)

Genghis Khan eventually retired to heavens at 990BC. I still pray for him. :religion:

Espionage economy(EE) vs. Manual research

Alphabet trade is probably a good point to add my two cents on the Espionage economy(EE) vs. Manual research issue. Alphabet enables training of spies, fantastic units in nullifying city defenses and stealing techs. Particularly in tech stealing missions, stationing spies for 5 turns gives 50% off, translating to some big reduction in research expenses. Some possible scenario for stealing techs would be the following:

First, notice that according to the Technology Research Explained (http://forums.civfanatics.com/showthread.php?t=146163) article, this technology cost depends on map size, game speed, game difficulty, number of teammates, and tech base cost. For example in this game, the technology cost for Optics was:

600 (tech base cost) * 3 (marathon speed multiplier) * 1.25 (immortal difficulty multiplier) * 1.5 (huge map size multiplier) = 3375 :science:

Secondly, espionage tech base cost is calculated through multiplying the technology cost in beakers by 1.50.

Then, the espionage base cost for Optics became: 3375 :science: * 1.5 = 5062 :espionage:

172108

On a side note, city revolt mission base cost is 500 :espionage: regardless of map size and game speed. Consequently, marathon speed and larger map sizes favor this mission type more. The same modifiers that apply to tech stealing missions also apply to city revolt missions.

Thirdly, many espionage mission cost modifiers affect this espionage base cost. Let's consider the following tech stealing mission scenario:

Your 5-turns stationed (-50%) spies steal from a nearby city (+20% distance penalty) with traderoute connection (-20%, careful (!): road connection is not enough) and with your state religion present, but AI having a different state religion (-15%, easy to achieve through Open Borders and missionaries). Assume 5 relatively cheap culture spread missions (http://forums.civfanatics.com/showthread.php?t=262243) were performed consecutively on this city, each reducing the city culture modifier by roughly %2.5, a total of -12.5%.

Further assume that your continuous espionage spending has doubled your espionage power (EP) relative to the target empire. You can look at the espionage graph to guess your relative EP. Then, the espionage point spending modifier becomes:

((2 * enemy EP) + your EP) / ((2 * your EP) + enemy EP) = 0.80

Fourthly, notice that the mission cost modifiers that show up when you move the mouse cursor on top of a possible espionage mission are not added, but multiplied. Multiplying all modifiers, espionage mission cost total modifier becomes:

50% * 120% * 80% * 85% * 87.5% * 80% = 28,56%

The mission will cost 70% cheaper; in other words, you will need to spend roughly only 1/3 of the espionage base cost.

If, in addition to your state religion being present in the target city, the holy city of your state religion also belongs to your empire, then the espionage religious modifier is subtracted by 25%. For this holy city modifier, it does not matter if you and the target empire share the same state religion. Then, for the scenario above, the espionage mission total cost modifier becomes:

50% * 120% * 80% * (85%-25%) * 87.5% * 80% = 20,16%

These bonuses don't depend much on the buildings you have built; therefore, Espionage Economy can be great in recovering where many human cities still lack infrastructure. These cost reductions look fine, but don't underestimate the hidden costs of an EE.

Let's remember that spies can get caught inside enemy territory. If your crucial military tech stealing plan gets delayed, you might have to fight a war with an obsolete army and lose the game. Thus, sending a sufficient number of spies to secure the espionage mission becomes very important for the success of an EE.

According to Bhruic's spy detection article (http://forums.civfanatics.com/showthread.php?t=247719) and Roland Johansen's summary post (http://forums.civfanatics.com/showpost.php?p=6062359&postcount=5), open borders agreement significantly reduces the chances of your spies getting caught. When one of your spies has just moved or is on the same tile as another one of your spies, then it has a 1.5% increased chance of being caught with open borders. Overall, with open borders, the chance of being caught each turn in enemy territory ranges between 0% and 7.5%, while with closed borders, this chance ranges from 0% to 18.75%.

With the 3.13 patch, spies cannot get caught while travelling on ships; thus, stealing from coastal cities using a naval transport would eliminate the risk of getting caught except when the mission fails.

You might be asking how many spies are required to secure a tech stealing mission. Let's assume you are stealing from an AI with whom you have signed the Open Borders agreement, and also, let's assume pessimistically the chance of your spies getting caught within enemy territory each turn is roughly 6%. For an optimistic scenario under pure EE, the chances of getting caught could be as good as 0.6%! A realistic, yet optimistic scenario under hybrid EE would be 2% spy detection chance per turn.

Let's continue with the pessimistic scenario and assume that it takes roughly 3 turns for your spies to reach the target city and 5 extra turns for the -50% espionage reduction to set in. Then, the chance of a single spy surviving for 8 turns becomes (1-0.06)^8=60%

Spies can also get caught during a mission. According to Roland Johansen's post on spy detection during missions (http://forums.civfanatics.com/showpost.php?p=6076611&postcount=30), the chance of being caught during a mission ranges between 0% and 75%. For tech stealing missions, this range extends to 75%*1.25=93.75%. Let's assume for our hybrid EE purposes that our spies have a 50% success rate for tech stealing missions. Significantly higher success rates are possible under pure EE. Then, 60%*50%=30% success rate for each spy sent; ceiling(100/3)=4. Under these suboptimal odds, on average, the 4th spy will succeed in stealing a tech from an AI with whom you have signed the Open Borders agreement.

Don't be fooled that 4 spies will be sufficient to secure the mission however. 30% success rate for each spy translates to 70% failure rate for each spy. Then, with 2 spies, failure rate becomes 49%, with 3 spies 34,3%, with 4 spies 24%, with 5 spies 16,8%, with 6 spies 11,7%, with 7 spies 8,2%, with 8 spies 5,7%, with 9 spies 4,0%, with 10 spies 2,8%, with 11 spies 1,9%... I hope my point is clear. While on average you should expect to lose not more than 3-4 spies on the tech stealing mission, you should still bring roughly 10-12 spies to secure the mission :-/ Good news is that 30% success rate was roughly the worst case scenario for a hybrid EE economy. In a focused hybrid EE or pure EE, mission success rates are significantly higher; therefore, you can get away with fewer number of spies.

According to the analysis above, the hybrid EE could be difficult to manage in the early Medieval era. In addition to all cities getting the courthouse spy specialists, a few castles and 100% espionage slider spending bursts might be required to gain espionage superiority against the target civilization. This espionage spending superiority would not only reduce the spy detection and mission failure chances significantly, but also reduce the mission cost further.

In conclusion, to keep your spy losses to a minimum, make sure you always:

steal from an enemy with whom you have signed Open Borders agreement
carry and station your spies with a naval transport
steal from an enemy against whom you have gained espionage spending superiority
take the shortest path to the target city to minimize turns spent inside enemy territory
send sufficient number of spies to avoid the worst case scenario
bring your spies to the target city through different paths or at different intervals to avoid spy stack penalty (assuming the number of turns spent in enemy territory remains the same)

Under a pessimistic scenario in a hybrid EE, I would suggest sending 10 spies to secure the mission, ie 2,8% mission failure chance. An optimistic pure EE scenario could get away with as few as 2 spies per mission. For a realistic hybrid EE scenario, let's take the average and assume 6 spies will be sufficient to secure the mission. Significantly more number of spies must be sent against an enemy with closed borders, but hopefully, you will gain a few trustable friends during the course of any game. =)

Now, let's add the production and maintenance costs for these 6 spies to the calculations. Each spy costs 80 hammers on marathon speed and can be rushed through a single population unit, 90 hammers for marathon speed again. Consequently, when 6 of your cities poprush spies simultaneously in the Medieval era, it is not a big deal for your economy. In this game, the lost population unit for a 11 size city regrew in roughly 12 turns with 5 food surplus and a granary. Say this citizen was working a village with +3 commerce, then the total lost commerce for 12 turns becomes: 12*6*3=216.

Furthermore, each spy incurs +1 gold for unit costs and an additional +1 gold for traveling outside cultural borders. Let's assume you are sending 6 spies for a tech stealing mission and spend 10 turns on the way: 2*6*10=120 gold.

For simplicity reasons, let's ignore the low cost of 5 consecutive spread culture missions, since their effect carries over to future missions. The culture spread mission base cost on a city with 1000 total culture would be: 1000*0.05*3=150 :espionage:. With modifiers, final cost of 5 missions would be roughly 200 :espionage:.

Then, assuming an average 1:1.65 commerce to beaker conversion ratio in the early game, the total effective price for the tech to be stolen increases by (216+120)*1.65=554 beakers. Note that this extra spending does not depend on the base cost of the tech to be stolen. For a 3000 beaker tech, this extra spy production and maintenance costs add a further 18,4% to the base cost. For a 10000 beaker tech, these extra spy production and maintenance costs only add a further %5,5 to the base cost. As your EE gets stronger, the spy production and maintenance fees compose less of the total espionage cost, and the EE becomes a more viable option. Additionally, the more you increase the espionage spending, the less you will pay for spy production and maintenance due to lower spy detection and mission failure chances.

To reduce this luck component involved in the mission overhead cost, a mainly pure EE should be favored over a hybrid EE. The switch from manual research to pure EE can be gradual. Code of Laws, Nationhood, Constitution, Communism, and Democracy are key techs on they way to a pure EE.

Including realistic spy maintenance costs (+12%) and also including the earlier-mentioned 1.50 multiplier for technology cost to espionage base cost conversion, the effective espionage cost reduction for a tech stealing mission becomes 112%*150%*28,56%=47.98%. This ratio converts a single commerce into roughly two science beakers, 2.08 to be exact. Additionally controlling the holy city of your state religion would have reduced the effective espionage cost down to 112%*150%*20,16%=33.86%. Then, the commerce to science conversion rate would be roughly 2.95.

Now, let's look at how effective manual research converts commerce into science beakers for a Medieval economy. In the early and middle phases of the game where the only scientific building multiplier comes from libraries (+25%) or maybe a few academies (+50%), building multiplier becomes roughly 1.30. When all prerequisites are met, this modifier gets multiplied by 1.20. Additonal modifiers are +20% for each optional prerequisite met, and precentage of living AIs already knowing this tech (max +30%). In conclusion, early-mid game manual research would translate on average a single commerce into science beakers at roughly 1.3*1.2*1.06=1,65 rate. The optimistic case of manual research would result roughly 1.3*1.4*1.3=2.36 rates.

Bureaucracy civic with +50% commerce in capital can complicate the calculation above to some extent. Since this +50% bonus applies to commerce and not science, both EE and Manual Research economies are affected almost equally.

On the other hand, academies (+50% science) are more common in most games than Scotland Yards (+100% espionage), because libraries can assign 2 scientists after Writing, while to assign the first and single spy specialist the expensive CoL must be researched. Especially on smaller empires, a capital with an academy running Bureaucracy can carry the weight of most of the manual research. Assuming 4/5th of the empire-wide commerce is generated at the capital, the average :commerce: to :science: conversion rate for smaller empires becomes: (0.8*1.75*1.5 + 0.2*1.25) *1.2 = 2,82. This high ratio of manual research in small empires clearly favors manual research over EE, except when you control the holy city of the religion present in the target city.

As the empire size grows bigger or on larger map types, the benefits of the super science capital running Bureaucracy diminish: (0.2*1.75*1.5 + 0.8*1.25) *1.2 = 1.83 Furthermore, Bureaucracy is not the best large&huge map size civic, since to dominate a larger map size, you'll need more than a single production center. For this reason, Vassalage might outperform Bureaucracy in large&huge maps assuming you are not stuck with a smallish empire.

EE also benefits from larger map sizes due to reduced "mission overhead to total mission cost" ratio. Since the exact value of mission overhead has a probabilistic function not directly dependent on map size, the increased tech costs on larger map sizes favor EE.

Finally, Vassalage+Organized synergy helps to cut down total expenses by 6-10%. Assuming early game 1:1 :gold: to :commerce: conversion rate, an Organized leader is better off with Vassalage and EE than Bureaucracy and manual research.

For the average case of both economies in the Medieval era, manual reseach allows :commerce: to :science: conversion rates of around 1.65, while espionage economy can afford 2.08 rates. Controlling the holy city of the religion present in the target city, along with other espionage mission cost modifiers, would clearly favor the effectiveness of EE (2.95 conversion rate) over manual research. These EE figures are not difficult to achieve except maybe on Deity difficulty. As the game progresses, more or less similarly advantageous infrastructure buildings and civics are introduced for both type of economies, and the EE continues on average to convert commerce to science more effectively than a manual-research economy.

In addition to converting commerce to science optimally, the EE also creates a minor, but nice synergy with the spy specialists after Code of Laws discovery. Once you build the courthouses and assign spy specialists, you get 4 (!) espionage beakers + 1 extra science beaker per each specialist.

4 * 2.08 + 1 * 1.65 = 9.97 effective science beakers.

This many beakers cannot be matched by any commerce tile in the early game, except by a riverside gold resource. With 10 spy specialists assigned and each courthouse contributing two additional espionage points, 60 espionage points can be generated per turn. Mid-game techs such as Construction cost around 2000 beakers with the settings for this game, ie marathon speed, immortal difficulty, and huge map size. Even if you continue manually researching a totally different tech path, you can accumulate enough espionage points to steal Construction from AI in approximately 2000/(60*2.08)=16 turns. If same amount of citizens were assigned as scientists instead of spy specialists, researching Construction would take roughly 2000/(10*3*1.65)= 44(!) turns, almost 3 times longer than the spy economy.

While the math behind the effectiveness of EE for larger map&empire sizes is obvious, I have not used any spies for tech stealing missions until much later in this game, but probably should have much earlier. Because tech lead has somewhat better trading&military potential, I am not advocating a full Espionage economy, but rather a hybrid economy in bursts where you first focus on manually researching some techs that lead to a tech monopoly and then on stealing some of the rest. Especially after assigning a significant number of spy specialists and allocating some of your commerce into 100% espionage spending bursts, there is little reason not to steal expensive techs from weak AIs. Since this type of hybrid economy will be somewhat hard to manage if some espionage points must be saved for city revolt missions, I suggest producing more than the usual number of bombard promoted siege units to reduce city defenses faster.

Pure espionage economy, ie fully shutting down manual research, could get well along with aggresive warmongering. On larger maps, however, warmongering will eventually be limited economically. At that point, a tech trade through peace treaty followed by a focused manual research is probably the best idea to backfill the many missing techs through tech trading. Alternatively, you can attempt to steal an expensive tech from the leading AI and backfill the rest from the other AIs through tech trading.

I hope you liked this discussion on espionage economy. Now, let's continue with the writeup.

VirusMonster
Jan 28, 2008, 03:41 AM
1190BC: Roman empire must heal the wounds of war, so I sign a peace treaty with Montezuma and enter a rather long peaceful period.

1000BC: Imperial treasury equals to 1000 gold due to the Mongolian conquest.

Still at 0% research with no profit and no libraries built. Number of cities=10, total population=39, so average city size is 4. Representation provides +3 happiness to 6 largest cities, increasing their happiness cap to 11. Consequently, almost all citizens are focused on working farms and food resources. At this stage of economic recovery, Organized trait clearly outperforms Financial, because working even a single cottage does not make much sense with this low population and lack of infrastructure.

Various AIs have already researched Currency (1), Aesthetic (1), Metal Casting (1), and Mathematics (5). I don't remember if I tried to bribe them into wars to delay their tech speed. Most likely, since my backwards empire had no techs to offer, and none of the AIs seem to have Currency, my bribe attempts were not successful.

The empire is paying 9 for units costs, 1 for units outside borders, 46 for city maintenance, and 14 for civic costs(I am running slavery and representation).. all data taken from the financial advisor screen.

Good news is libraries are about to be built! Keeping up the hope. I was very optimistic about my game at this stage despite being backwards. I guess the Mongolian Alphabet gift had a tremendeous effect on my attitude =)

This economic crash will recover amazingly quick within the next 50 turns so watch out

860BC: I've decided to convert my treasury advantage into a tech lead, and possibly into a wonder lead. After a few libraries have been built and 2 scientists have been assigned, I run 100% research losing 74 gold per turn. I also emphasize commerce in all cities.

In return, my empire gets approximately:

(6 scientists x 6 (3 science per specialist + 3 science for representation) x 1.25 (libraries give +25% research bonus)) + (3 GreatGeneral specialists x 3 science for Representation) = 54 :science: from specialists

(74 commerce from citizens working commerce tiles + 15 commerce from traderoutes) * 1.25 (library bonus) = 111 science through commerce, which helps me boost my research rate to around 54 + 111 = 165 :science: per turn. Great =) Aesthetics costs 1686 beakers and can be researched in approximately 10 turns.

Once I had Aesthetics before most of the AIs did, I went into a very successful techtrading spree that brought me tons of gold from AIs that already had Currency and tons of techs. In particular, I was interested in learning Mathematics so I could chop some aesthetic wonders =)

Here are the details of these trades:

Backfilling with Aesthetics

840BC: Discovered Mathematics and Polytheism through Aesthetics trade (the Parthenon can be built now)

I don't recall the exact trade, but most likely, I have reseached 1 or 2 turns into Mathematics so the AI can reduce the price on it. Then, I traded Aesthetics (1686 beakers) in exchange for Mathematics (1404 beakers). The AI must have given Polytheism as well.

830BC: Traded for Priesthood and Meditation (can chop Shwedagon Paya!!! =))

810BC: Traded for Currency (see screenshot), Organized Religion(+25% to wonder production), Monarchy, and Sailing

Note that Asoka did not want to trade Currency with Aesthetics initally. There is a huge beaker difference between the two techs, not only that, but also at higher difficulties, the AI sells his techonologies more expensively.

So what to do? Waste all of your gold reserves to reseach a few turns into the desired tech, because when you research some portion of the technology, the AI will charge you less for the trade. Here is the screenshot of trading Currency from the only AI knowing it or maybe Isabella had it too, but definitely not more than 2 leaders. Watch that I have only 177 gold left at this point and Asoka gets 175 of it as part of the trade =)

171353

Wonder production supremacy

790BC: I am chopping for Shwedagon Paya. Thanks to Isabella for spreading Buddhism to my cities and thank you for trading Organized Religion with me. Also, thanks to the other AIs for trading Mathematics with me for 50% chop hammer increase =) 834 :hammers: at Karakorum in 1 turn at 790BC!

168453

Shwedagon Paya will help me run Theocracy early on, and during Golden Ages, I can switch to Pacifism to boost my :gp: generation rate even further. Without the Spiritual trait, my continuous warmongering plan could probably accommodate switching into and out of Pacifism only during Golden Ages.

720BC: Thank you the jumbo elephants! Your efforts won't go waste, less battles there will be through the Statue of Zeus =)

According to Krikkitone's article on War Weariness Mechanics, on Immortal difficulty, AI's war weariness unhappiness in a city is 70% less compared to that of a human city with same war weariness. The Statue of Zeus increases this 30% WW multiplier to 60%. WW applies only on wars where you are not culturally dominant. Consequently, when you are defending in your own territory against an AI attack, the AI will be two times more willing to sign a peace treaty after some causalities. In conclusion, the Statue of Zeus is mainly a defense-oriented wonder.

680BC: I was trading Marble from DeGaulle, but it has been cancelled. I don't recall exactly, but I was trading Aesthetics excessively with all AIs to get their gold reserves. Most likely, after trading it with DeGaulle, I triggered an unwanted response. Now, Istanbul has 4 forest left that are going to be chopped without the 100% production bonus. To spread Buddhism to Istanbul, I poprushed a Buddhist missionary in a nearby city. Hopefully, Istanbul will still be able to complete the Parthenon on time.

630BC: Poprushing the Parthenon =) Along with the Roman Forum, the great person generation rate will be boosted by 75%.

168455

The Aesthetics triplet is complete:

Shwedagon Paya, Statue of Zeus, and Parthenon :woohoo:

I was hoping to save some forests for the Great Library in Istanbul, but oh well, all are chopped =) I guess I might as well avoid building the Great Library with no Marble around.

610BC: Discovering Code of Laws. Courthouses will be rushed very soon.

570BC: Discovered Literature, but don't have Marble, and no one is willing to trade. Skipping the Great Library or maybe not, because my Pyramids + Shwedagon Paya city can get a GreatEngineer =)

510BC: Karakorum pops my first GP, a GreatScientist, sent to the capital to construct an academy. Definitely skipping the Great Library now.

500BC: Let's have a look on has changed in 50 turns on Marathon speed :) Remember that at 1000BC Romans seemed to have a horrible economical situation.

500BC financial situation

Empire population at 64 with 10 cities. 7 Libraries have already been built, and 14 scientists can be supported at 6 beakers with Representation. 84 beakers from scientists + 9 beakers from the 3 Great Generals = 93 beakers, not counting the +25% Library bonus. Adding the soon-to-be-built Academy research bonus at the capital, the empire should make 130 beakers by scientists alone. However, as of 500BC, I operate 5 scientists only, since my cities are focused on growing to the population cap.

(5*6 beakers per scientist specialist)+(3*3 beakers per settled GreatGeneral)+(4 beakers for 1 SpySpecialist) *1.25 for libraries= around 52 :science: per turn through specialists

If I am getting 52 beakers through specialists, how could I be running my research at 60% (141 :science: generated per turn)? Total expenses are at 93, and total gold generated per turn at 100% goldrate displays 125 :gold:. How could the 125-93=28 gold surplus correspond to the missing 141-52=89 beakers per turn? Even if we assume that 28 gold gets the 1.25 library multiplier, it is still 35 beakers. Also count for the minor 7 gold deficit per turn at 60%. Still, where are the missing 89-(35+(7*1.25 library effect)=46 beakers?

Here is the deal. Let's look at the gold situation on the financial advisor screen:

168456

:gold: :gold: :gold: 37 gold gained per turn through resource trading. 37 gold makes up almost 30% of total gold generated at 100% goldrate. 40% of the total expenses(37/93) are being compansated through resource trading. Thus, we can reduce the goldrate by %30 and boost our research from 30% to 60%: a 100% increase in research slider rate, add the library effect and it is even more than 100%. 46 new :science: per turn are added through resource trading. ResourceTrading is so key in recovering from economic collapse after overexpansion. If you've gained so much land that your economy halts, then make best use of it and sell your newly acquired resources!

You might be asking how much commerce traderoutes are bringing in at this point. The empire spans 10 cities at this point, and all cities have 2 :traderoute:s each. Most cities get 2:commerce: per traderoute, and a few get 3, creating a total of 40-45 extra :commerce:. Traderoutes account for 30% of our reseach rate at equilibrium, giving a similar economic boost to that of ResourceTrading. The missing components to high research rate can now fully be explained.

And I am researching at 100% using imperial gold reserves. This rush is creating 200 :science: per turn and thus, Construction(1965 beakers) can be researched in around 10 turns. (104 from regular commerce + 52 from specialists + 42 traderoute commerce)

------------

Also, notice that the effects of the organized trait start to be felt. With Representation and Organized Religion, my civic upkeep is at 23. It would have been 46 without the organized trait, corresponding to %20 of our total commerce and around 15% of our total research. You might be asking to yourself whether Organized or Financial would have been better. With 64 citizens at 500BC, if one third of these citizens would be working on 2 commerce tiles, then it would break even with the organized trait. But during wartime and even during peace time, when you want to focus for production exclusively and use poprush extensively, you will experience a drop in the total commerce generated. Because to grow your population back to the happiness cap in the quickest way possible will require you work on the farms and not the cottages.

Furthermore, if you want to work the mines for production, you will need a 1 or 2 farms to support the mines. You can't cottage everything if you need the hammer production. If you are going to depend on the poprush, then you still have to live with very few actual hammers produced.

I think organized is a solid trait for warmongering and works fantastic with the Slavery civic. I will make further comparisons at 1AD, 500AD, and 1000AD.

I have only 1 Courthouse built at this stage so city maintenance is almost 100% with 50 gold. Also note that unit expenses still compose a large chunk of my total expenses, 18/93=19%.

-------------

460BC: Mausoleum of Maussollos has been built by the Ramesses... hmm definite future target =)

450BC: Construction complete, now I can produce Catapults..

410BC: Montezuma declares war on me. I am not sure if I bribed someone into the war before or after he declared war.

Montezuma's lands stand to the east of my empire. Asoka has marble, but we have been good friends, and I dislike backstabbing for a single resource. He acts as an absorbtion mechanism to all threats from the south as well. Thus, I wanted to war Montezuma anyways.

I also check the relations screen every so often, because I want to make sure every AI dislikes another civ more than they dislike me =) I would agree to most demands by the AIs except tech trade requests or tech demands for my fresh tech monopoly.

390BC: 3rd GreatGeneral born.. and settled in the capital for fresh CityRaiderIII units. 3exp from barracks + 4 exp from Generals + 2 from Vassalage + 2 from Theocracy = 11 exp.

I need to reseach Feudalism however. Shwedagon Paya already provides me with Theocracy.

360BC: Tlatelolco captured. It stood between Rome and the Aztec capital.

310BC: Metal Casting discovered. I will get an additional 25% production bonus with Forge as well as some badly needed happiness.

300BC: Tenochtitlan, the Aztec capital, has been captured.

My great economy can support sufficient espionage spending to perform city revolt missions. Furthermore, I have brought a few catapults to soften up the defenders, so my Praetorians can have an easier time.

Here is a great trick for poprusing as well. I have used it several times to poprush 2 units instead 1 without getting an extra unhappiness penalty.

The poprushing system works such that if you require more than 90 hammers(on marathon) to complete the units production, it will use 2 citizens. Poprushing 2 citizens actually produces 180 hammers. Yet, when you rush a unit without having put any hammer production into it, you need to poprush 1 more citizen due to the poprush penalty you get.

Here is the deal: A Catapult costs 100 hammers on Marathon speed, while a Praetorian costs 90, a swordsman 80. Put less than 10 hammer production to the catapult. Once you have 1-9 hammers spent on the production of the catapult, then you will poprush the Catapult still for 2 citizens, because it costs more than 90 hammer, but this time without the poprush penalty.

The excess hammers you get, around 81-89 are spent on your next production. Thus, you can poprush your praetorian or swordsman in addition to the catapult every 30 turns(on marathon again).

Growing your 11 size city on marathon takes 126 food, 63 with a Granary. If you have a food surplus of 4, then you can grow 2 population almost every 30 turns, enabling you to poprush again 2 citizens for 180 hammers.

This tactic works with a Horse Archer as well, since Horse Archers cost 100 hammers, same as catapults.

260BC: I am slowly invading the Aztec lands, but for sure. I wait for more troops to arrive and also wait for healing between city captures.

250BC: To attack Aztec lands with a stronger army, I adopt Theocracy.

210-200BC: Texcoco captured and recaptured.

170BC: Ouch, Isabella, my closest scoring competitor has just got war declared by 4 other AIs.

Since the end of my wars with the Mongols and since I become the score leader, the AIs have been very busy declaring war to each other. I have been offered several times to join sides with them, but I repeatedly refused joining any wars, keeping my resource trading profits high.

I also discover Civil Service this turn and start upgrading my Praetorians into Maceman.

160BC: I adopt Bureaucracy. Hopefully, my 4th Great General will also join the capital for a total of 3 settled GGs there, thus allowing me to produce CityRaider III troops without Feudalism.

140BC: The 4th Great General emerged. Theology discovered same turn as well, but I already had Shwedagon Paya enabling me Theocracy anyway.

Ravenna, the 5th city I settled at the start of the game, has been recaptured.

100BC: Tlaxcala captured.

90BC: A random wedding event asked me to risk my relations with Bismarck in return of improved relations with every other empire out there. Right after my approval, Bismarck declares war on me and captures one of my Aztec cities while my army is away.

80BC: My lovely Teothiuacan captured by Bismarck. =)

70BC: I make peace with Montezuma to focus my attention on Bismark. Possibly this peace treaty will result Montezuma becoming a vassal of somebody else, but oh well, he had 2 cities left anyway, not a huge treat anymore. I can take him and his master at the same time if I have to.

50BC: Tenochtitlan, Aztec ex-capital captured by Bismarck. My army is approching to beat Bismarck, but I was late to defend my cities =)

20BC: My stack beats 5 of Bismarck's units while only losing 1 catapult. Catapults have barrage II promotion; it is great for land battles. I bring some WarElephants to join my hybrid army.

Keep in mind this is a soft war, and I haven't switched to Police State yet. I am still running Representation for the research bonus. I want to keep my research as high as possible until the discovery of some key military techs, such as Engineering and Feudalism.

10BC-1AD: Bismarck is losing many WarElephants against my stack. Aztec cities flip sides a few times until I gain control of the two Aztec cities previously captured by Bismarck.

1AD: Now, let me recap the financial situation. Here is the financial advisor screenshot:

168457

Empire size=16 cities, empire population=105 citizens. 10 courthouses already built, 4 more on the way. I like to run some spy specialists, since they are the most economical way of creating espionage points for city revolt missions. The HOF special domestic advisor does not display the number of spy specialists correctly, so I can't count them easily, but I am getting 57 :science: per turn at 0% research rate. I can safely assume that all of that research is coming from specialists and the settled GGs under Representation.

I could have run more scientists to boost my reseach rate significantly, (max = 193 beakers from specialists), but warmongering allowed access to many new resources, and cities are still busy growing to the happiness limit as quickly as possible. My war against Germany also requires working the farms to sustain poprushing. I break even at 60% research rate, with 225 :science: per turn.

ResourceTrade foreign income is 46, paying for 1/3 of total expenses, 136 :gold:. Without the resource trading, to allocate 46 more gold, I would have to drop the reseach by around 46 * 1.25 (the library research bonus) = 57 beakers, 25% of the 225 initally produced. Without successful resource trading, research would drop roughly by 25%. If I composed my research mainly of scientists, I could definitely afford many more at this point in the game, then the resource trading would have contributed to a smaller fraction of total research.

Traderoute profits equal to 65 commerce and compensate roughly half of the total expenses. Their effect to the total economy remained rather constant and significant since 500BC.

--------

One more observation on civic upkeep. I am running Representation, Slavery, and Theocracy. Civic upkeep cost is 43, similar to that of the resource trading income. With Organized trait cutting civic upkeep by half, I am saving 46 gold, preventing a drop in research rate by approximately 25%. Civic upkeep makes up for 30% of total expenses, and inflation kicked in by 7%.

If I had financial trait at this stage of the game with 105 population and if 40% of my population worked on commerce tiles already generating 2 commerce, then it would break even with the organized trait. It is reasonable to assume that that at least half of your population will be working on commerce tiles by 1 AD, thus around 1 AD, financial trait benefits started to balance out with those of the organized.

On the other hand, keep in mind, I wouldn't be able to build 10 courthouses at this point if Julius Caeser wasn't Organized. My city maintence expenses would have been slightly higher. You can see that the total city maintenance is at 65, which roughly corresponds to %50 of my total expenses. Lower number of courthouses would have increased this ratio upto 60-70%.

Finally, the unit costs make around 13% of total expenses, a decline by 6% from the 500BC data. As your empire grows economically, it can naturally support a larger army, and the costs of a such army still should not make up a higher precentage of total expenses than in the past. Although I followed a strong warmongering strategy during the rest of the game, my unit expenses to total expenses ratio continued to stay the same.

VirusMonster
Jan 28, 2008, 03:41 AM
OK, let's fast forward to 500AD:

30AD: The third city I settled, Cumae, gives birth to my second GP, another Great Scientist. I make a suboptimal decision and send this GS to create an academy at Istanbul, my GreatPerson farm. Well, my cottage specialization city, Antium, had maximum science rating after Rome, and Rome already had an academy; therefore, the academy would have fitted Antium better.

I decided to not lightbulb Philosophy, because several AIs already had it. It turned out that my decision was optimal, since game lasted an additional 200 turns, and Philosophy would have been manually researched through the academy much sooner.

60AD: Montezuma becomes Vassal of Peter, I guess after the Germans and the Ramess of Egypt, I need to fight against both Peter and Montezuma.

70AD: I capture Munich and make peace with Bismarck to build more troops to advance further into Germans lands. The AI production bonuses halt my advance.

170AD: Bismarck declares war on Isabella =)

190AD: I discover Engineering, great for rushing units to battle with 3 movement points on roads. To increase war production, I am also building forges in most of my hammer rich cities. Forums, the Roman replacement for market, are only built at my GP farm for the obvious +25% GP generation bonus and at a few of the other commerce rich cities.

I am not commenting on the many wars the AIs have between each other. For example this turn, Alexander declared war on my soon-to-be-again enemy, Bismarck =) I keep my relations good with everyone and avoid unnecessary fighting. My current focus is invading Germany, and I keep it simple.

250AD: Declaring war on Bismarck.

300AD: I capture Essen. My troops advance very slowly, since I haven't produced sufficient number of catapults or spies to capture cities quickly enough. Marginal German city placement also delays my advances since I don't want to split my stack. My maceman stack could have been larger in numbers as well. Well, I hope the forges, courthouses, and forums were worth it =)

310AD: Finally discovered Feudalism, will switch to Vassalage pretty soon. I feel Bureaucracy is a small map civic, since to dominate a larger map size, you'll need more than a single production center.

320AD: Something very interesting happens which I did not notice until much later in the game. Peter which is on the far east of the map captures NewYork from the Americans. Well, Americans are on the far west side of the map =) How could that be? I thought it was a huge map. Just when I thought I finished Peter off later in the game, I realized he had 1 more city left, the NewYork.

330BC: Big event for me: Apostolic Palace decides for war against Bismarck =) Isabella and Peter send armies to finish Bismarck off.

Same turn, random event also gives 1000 beakers toward Divine Right, I was not planing to research it anyways, so not a big deal.

370AD: Great turn for me. I capture Dortmund and get a Great General, which I settle at Karakorum, my second military production city after the capital.

Same turn, my GP farm, Istanbul, gives birth to my 3rd GP, a Great Artist. What should I do with him?

420AD: Enough with focusing on research. I switch to Police State and Vassalage. I probably should have used my Great Artist at this point for a Golden Age and switched my civics after. My main aim is to produce a huge Trebuchet army with CityRaider II&III promotions. Also, I need sufficient number of spies to make my advances quicker. Some more macemen and pikemen would help as well. Basically, I need a real army =)

Most significant tip for early finish dates:

As another top conquest & domination scorer in HOF, The-Hawk, suggests for early finish dates: minimize builder tendencies and produce as many attack units as you can. I could not agree with him more. From 420AD forward, my entire empire is set on unit production, except minimal infrastructure such as granaries, barracks, courthouses, and libraries.

Siege promotions:

I also realized during my German siege that I have been promoting my siege units the wrong way all along. City Raider is a much better promotion than Barrage, because it significantly increases your siege units win/loss ratio.

The main reason for not capturing enemy cities quickly enough is lack of collateral damage. Because siege units are the first ones to enter the battle and probably compose most significant portion of your losses, if you can keep their losses to a minimum, then you will not have to wait for replacements for the 2nd target city on the way.

440AD: Berlin captured. It has some nice wonders built and 2 settled Great Generals. Chicken Itza, The Great Wall, Hanging Gardens and an Academy.

460AD: Frankfurt captured. Germans're done. Music discovered as well.

500AD: Financial advisor analysis

168460

Compared to 1 AD, the precentage of city maintenance to total expenses have not changed. It is still around 50%, 107/205.

55 :gold: come through resource trading. Total expenses are 205 :gold:, thus the precentage of foreign trade profit compensating for total expenses has dropped from 1/3rd to 1/4th since 1AD. I have been trading very aggressively through trade renegotiations, my total foreign trade profit increased, but my empire size grow at a larger rate, thus I got more expenses. I did not use the trade withdrawal trick, but I have used the HOF alert system to notify me whenever the AI had 1 extra gold per turn to trade, so I could cancel the previous trade and reinitiate it at a higher profit.

Civic upkeep for running Police State, Vassalage, Slavery and Theocracy equals 75 gold. Organized trait saves 75 gold, which would have increased total expenses by 37.5%. Compared to 1 AD, this precentage has incrased from 30% to 37.5%, but we are using the Vassalage civic at this point giving us 22 free units. The cost of 22 more units would have been roughly %11 of total expenses, -2% from the 1AD data. You can see there is some synergy in using Vassalage civic with the Organized trait. Army expenses are being cut by half under Vassalage by an Organized leader.

I control 12 courthouses and 4 more are being constructed. Empire spans 20 cities with 195 total empire population. Average city size is 10. Happiness and health cap are around 15, thus I can afford 5x20=100 more citizens. At this stage, I could want to work more commerce tiles instead of farms as well, because growth became less important. Inflation increased from 7% to 13% as well.

To make up for 75 gold under Financial trait, approximately 37.5% of the total 195 citizens would have to work on 2 or more commerce generating tiles. Compared to 1AD, there is a slight decrease in the number of citizens required to work 2 or more commerce tiles to break even with the Organized trait, thus you can see that Financial is getting slightly in the lead of Organized. Also keep in mind that with Financial, your cities will be generating more beakers at the same research precentage. Thus, around 500AD financial starts to get slightly better economically. I am not counting the effects of Forums and Grocers yet, because I have only 1 city with a forum built by 500AD.

And let's not forget the TradeRoute Economy. Mainly due to Snaaty's misleading FE article, many people mix it up for food economy, whereas :traderoute: dynamics have very little to do with food. For a relatively distant traderoute partner city, the base traderoute profit depends on the population of the AI city the trade is generated from. Divide the AI city size by 10 to get the base traderoute profit. For example, for an AI city of size 15, the base profit would be 1.5; for AI city size 20, base profit would be 2.0, etc... According to Krikkitone's trade explained article, less distant traderoute partner cities might get a lower traderoute base profit. Rest of the modifiers are explained in detail when you move the mouse cursor on the traderoute profit.

I have 2 :traderoute:s per city at this point due to Currency. My most populated city gets 2 routes at +4 :commerce: each. Next most populated city traderoutes bring +3 :commerce: each. My lower populated cities bring +2 :commerce:. Watch that the human player city size does not affect the total traderoute profits your empire is getting, it only affects how the foreign traderoutes are distributed across your cities.

It is said that total commerce generated through traderoutes can be seen at the info screen as imports/exports. It seems I have 301 imports at the Info screen, but the numbers do not add up. According to the Info Screen explained article, assuming average trade route profit for city is (2+3)=5 commerce and I have 21 cities, the total commerce generated should be around 21*5=105, but what I see at the Info screen is 307/105.

Most probably, the 3.13 patch fixed the issue of Import/Export confusion and the AI is making 3 times more profits out of my empire. I probably should have switched to Mercantilism in this case.

Compared to resource trading bringing in 55 gold per turn, traderoutes bring roughly twice the amount of gold to my cities. Consequently, :traderoute: economy pays for 50% (105/205) of total expenses.

500AD: Battle plan screenshot

168461

Watch the arrows I drew, they explain everything. Dark red is :devil: Peter, yellow is :queen: Hatshepsut . Green is :egypt: Ramesses, teal is :king: Louis. Purple is my friend Asoka, and, as mentioned earlier, acts as an absorbtion mechanism to all threats from the west and southwest.

Let's fast forward to 1000AD.

515AD: Hatshepsut declares war on Peter. Great news for me :-)

520AD: Discovering Guilds before the AI does.

530AD: Declaring war on Hatshepsut, the yellow empire on the screenshot.

540AD: Discovering Compass.

545-565AD: Captured Heliopolis, Thebes, and Pi-Ramesses, destroying Hatshepsut Empire.

570AD: Kolhapur near Karakorum assimilated culturally. :assimilate: :culture:

580AD: Banking discovered. I probably should have switched to Mercantilisim, but I thought I was making +300 traderoute profit according to Import/Export screen that makes no sense. On the contrary, the AIs made 3 times as much traderoute profits than I did. More on Mercantilisim vs. FreeMarket comparison later.

I was the first one to research Guilds btw, and I am also the first one to research Banking. Who said you can't win the race to Banking at Immortal?

585AD: DeGaulle wants to be my vassal although I am not in war with him. I refuse the first time around, but accept within a few turns.

While agreeing to be his master, I have acquired new enemies, Gandhi and his vassal Frederick. I don't fear them much, because Gandhi is never a good warmonger, and his empire is in the far southwest of the map. Ramesses also declares war on me; he was a neighbour of DeGaulle and was controlling the mighty Mausoleum of Maussollos.

605AD: I capture Akhetaten, Egyptian city on my way to the Mausoleum of Maussollos at Elephantine.

625AD: Ramesses is now a vassal of Peter.

695AD: My 4th GP, a Great Spy, is born in an unexpected city, Ning-hsia. Distributing your GP generation over a few cities in the early periods of the game and only in the later periods focusing your GP generation on one city seems optimal for getting those GPs earliest. Early GPs help significantly for an early victory date.

The most effective use of this Great Spy would have been to steal 5-6 techs from Justinian. When I replayed the game and infiltrated Justinian, the espionage point spending modifier became -31%(!). With the 9000 espionage points brought through the GreatSpy, I could steal Divine Right, Philosophy, Paper, Optics, and Education. Fantastic deal! Also, don't forget that future missions will continue being cheaper and the spy detection chances are significantly reduced. Imho, Great Spies during Mediavel era should always be used for infiltration.

But what did I do? :-/ Saved this Great Spy for a future Golden Age :-) Still better than constructing a long-term minded Scotland Yard; nevertheless, significantly less optimal than infiltration for an early domination victory.

720AD: Discovered Drama.

735AD: Peter declares war on me. Actually, I wanted to declare war on him around 10 turns ago, but because I accepted one of his demands or because he accepted to one of my demands or because his vassal Montezuma accepted to one of my demands, I could not declade war on him, thus I had shifted my army position away from Peter's lands; he backstabbed me. :backstab:

Peter and Montezuma capture 2 of my cities, but Monty is destroyed within a few turns at 760BC.

750AD: Asyut captured.

760AD: I am happy that Elephantine (belonged to my vassal De Gaulle) has been captured by the Russians. I will capture Mausoleum of Maussollos at Elephantine.

Xochicalo captured from the Aztecs.

790AD: Calixtlahuaca recaptured. Giza and Akhetaten captured. I am invading the Ramesses lands. He is a vassal of a weak AI called Peter.

800AD: Peter discovers Liberalism. Uhoh, big boy got too smart ;)

805AD: Byblos captured and Gunpowder discovered.

835AD: Memphis captured. Egyptians down to last city. Philosophy discovered on same turn.

840AD: Alexandria captured. Ramesses wiped out. Novgorod captured same turn from the Russians by my 2nd stack.

865AD: Moscow captured. Nice capital =)

870AD: Yaroslav captured.

875AD: Istanbul gives birth to my 5th GP, a Great Merchant, and I decide to save it for Sushi Cooperation if the game goes that far. I seriously considered sending it for a trade mission to one of Gandhi's coastal cities, but decided not to, becaue this Pangea map type would not allow for the intercontinental trade modifier to double the mission profit.

880AD: 2 GGs are born on the same turn. 1 for Peter 1 for me. Rostov captured.

895AD: Bad news. Giza joins my vassal DeGaulle culturally. Akhetaten joins Isaballa. I am afraid that the soon-to-be-captured Mausoleum of Maussollos city could flip as well, thus I plan to send some strong troops there and switch to universal suffrage to buy a theater. It has too few population and can't poprush the theater, and I don't want to switch out of Slavery to Caste System to assign artist specialists, because my war production depends highly on Slavery usage.

900AD: St. Petersburg captured.

905AD: Yakutsk captured. Yekaterinburg captured. Peter becames my vassal, and I learn about the NewYork city he captured ages ago in the faaaaar west of the map. Nationalism discovered same turn.

910AD: Divine Right and Paper discovered, possibly through a tech trade. I don't exactly remember.

940AD: Started the war against Louis in the southeastern part of the world.

955AD: I capture Avignon and discover Education same turn.

965AD: Elephantine finally captured =) Controlling the Mausoleum of Maussollos. I should have used my GreatArtist right away for my first GoldenAge, but I am not wasting my GP collection yet :) 3 GPs were waiting, doing nothing, for so many turns :-/ Definitely not the smartest way to use them :-)

985AD: Lyons captured. Avignon recaptured, and some defense left in the city to prevent future recaptures =)

Louis becomes Isabella's vassal, and I must fight Isabella now as well. My closest city to Isabella's border has a castle up; thus, I seriously doubt she can harm me.

995AD: 1 more GG is born, and Orleans, Louis's capital, is captured. It is a fantastic city with tons of wonders and 5, yes 5, settled GreatProphets.

168462

1000AD: Financial advisor analysis

168463

With 42 cities, I've hit the conqueror's plateau, and my city maintenance costs due to newly added cities do no longer increase through city distance maintenance. You can observe the immediate result that the city maintenance to total expenses ratio is reduced to 45% from 50% compared to 1AD and 500AD. For a more detailed conquerer's plateau analysis, refer to Gato Loco's city upkeep explained article.

Unit supply and unit costs have slightly increased compared to 500AD, since I have been mainly producing units only in my entire empire during the 500-1000AD period. However, because imperial population also doubled since 500AD, more military units can be supported for free, and more :gold: is generated to compansate for army expenses. Unit expenses at 1000AD account only for 5% of total expenses, an almost negligible amount. There is also the Vassalage component of army expenses in civic upkeep, but it is cut by half through Organized trait. According to Vassalage formula, this expense corresponds to 0.15 * empire population, 0.15 * 409 = 60 gold. Cut by half and it becomes 30, only 5% of total expenses; consequently, total army expenses make up approximately 5%+5%=10% of total expenses. For more of this beautiful math, refer to Roland Johansen's great articles on unit maintenance and civic upkeep :-)

Total foreign resource trade expenses have dropped, because I have eliminated some of the AIs and have gained new enemies. Consequently, the ratio of foreign trade to total expenses has dropped significantly from 25% at 500AD to 7% at 1000AD. Another bad news on economy is that inflation has skyrocketed from 13% to 25% in 500 years.

As mentioned earlier, my empire size was 42. Of these 42 cities, 31 have courthouses built. Total empire population is 407 citizens. Average city population did not seem to change much within 500 years, but the population of the highest populated cities has increased by 3-5 citizens. Forums and Grocers are built or are being built in the highly populated cities.

Now, let's compare Financial and Organized traits one last time.

A doubled civic upkeep for a non-Organized leader would result in 29% increase in total expenses. This ratio has decreased by 8% since 500AD where it was 37.5%, it is back down to the 1AD levels. Empire population size become 409, it has doubled since 500AD. We have 409 citizens that could be working many, many already 2 commerce generating tiles. To offset the extra 167 gold, a non-organized, but financial leader would have to allocate roughly 40% of empire population to work on cottages. This ratio does not seem to have changed much since the 500AD period, but more cities have forums, grocers, and libraries built. Consequently, the extra commerce can be converted to gold and science beakers at a more effective rate. It is clear that Financial trait is getting in the lead of Organized at this point in the game.

The GreatLighthouse was granting 2 extra traderoutes on my coastal cities. Imports/exports ratio had become 339/145. My GP farm was getting most trades with 4 traderoutes each bringing +6 commerce.

As I should have done at 500AD, I should still have switched to Mercantilism, because when I did so while replaying my game, 42 cities with an extra merchant specialist each were getting a total of 42x3=126:gold:, almost fully compensating the Decentralization economy. The extra +25% beaker bonus of the Decentralization economy due to heavy library infrastructure was more than sufficiently compensated through internal traderoute profits, roughly 42x2=84 commerce. Furhermore, 20 imports commerce under Mercantilism was still coming through my vassal DeGaulle.

In conclusion, Mercantilism would have clearly exceeded the Decentralization traderoute commerce rate of 145. Additionally, AIs would not have received any traderoute imports from our empire. Finally, running Representation with 3 extra beakers per specialist, Mercantilism economy would have clearly outperformed a Decentralization one.

Alright, let's finish this long writeup =)

1000AD-1230AD period:

1000AD: I capture Tours and Chartes. French wiped out.

1015AD: Military Tradition discovered, and I start my first golden age using the GreatArtist, switching to Pacifism. As discussed earlier, delaying the first Golden Age to such a late period was not the best idea, but I've recently gained control of the Mausoleum of Maussollos in 965AD. On Marathon speed with Mausoleum of Maussollos, Golden Ages last 24(!) turns.

1050AD: Economics discovered and switched to FreeMarket: another suboptimal decision :-) As discussed at 500AD and 1000AD, Mercantilism was clearly outperforming the Decentralization economy. The extra traderoute from the FreeMarket economy would not have affected the superiority of Mercantilism, because Roman empire was controlling almost half of the earth at this point, and not a sufficient number of highly populated AI cities existed as traderoute partners to fill this extra traderoute slot.

1055AD: The first Roman GreatEngineer (6th GP) is born in Istanbul despite the very low odds. Where were you when I needed to build the GreatLibrary =) Should I rush the IronWorks with you or should I lightbulb Steel for uber siege units?

1080AD: Chemistry discovered.

1115AD: Military Science has been discovered. I know it does not lead to any new techs, but it is great for finishing off the game. You can finally build the military academies in your military production cities. AI teching to rifles? You can upgrade your maceman into Grenadier with +50% vs Rifleman. It is a game finishing tech. I built 2 military academies with the GGs I have been saving; one at my capital, one at Karakorum.

Furthermore, since it is a dead-end tech, after you have sized its advantage, ie produced stacks containing 40-50 units, then you can trade it without much fear against the AI.

1120AD: I declare war on Asoka and his master Joao.

Stack composition:

My stacks consist of masses, but masses of maceman (did not upgrade to Grenadier yet, maybe I should have, yea I probably should have =) I would have gotten a 50% strength boost: 8-->12).

My 2 production centers Rome and Karakorum were massing Cuirassiers, 1 unit every 1 or 2 turns. 1 Cuirassier or 1 Cannon costs 200 hammers on Marathon speed, and Rome was producing 213 during a Golden Age. Let's not forget the huge number of CityRaider III promoted Trebuchets. I had 2 main stacks at that point, each consisting of around 30-40 units: 5-8 Spies, 7-10 Trebuchets, 7-10 Cuirassiers and 7-10 Maceman, and a few Musketman.

One of them entered Joao's territory through DeGaulle's border from the east, while the other entered Asoka's lands from the north.

A city raid with minimal losses would go as following:

Spy revolt on first or 2nd try. City cultural and walls&castle defense bonuses reduced to 0.
First CRIII trebuchet usually dies.
Second, third, fourth, and 5th Trebuchets survive and bring all enemy defenders to the lowest possible HP.
Rest is peace of cake.

I used to upgrade my siege units with Barrage, but as mentioned during German siege, I have realized that CityRaiderIII is the ultimate promotion for high win/loss ratios. For every trebuchet lost, I was killing around 6-7 units.

Cuirassiers were usually upgraded to combat III. I could have settled 1 more GG to Rome to produce lvl 4 promoted Cuirassiers (3 barracks + 2 stables + 4*2 GG + 2 Theocracy + 2 Vassalage = 17), but I have used my 9th GG on another military academy. Karakorum and Rome had the inital 6 GGs distributed evenly.

Observations on healing:

I probably should have also produced a single woodsman III, medic III unit using a GreatGeneral. It is, as I learned after my game from the "Unit healing" article, the only way to stack woodsman III healing with that of Medic III. When one of my maceman had reached 30 experience points, I could have used one GG to promote it to 50 exp which would have unlocked the 7the promotion needed for the woodsman III, medic III combo. Such unit would have increased my city capture speed significantly, especially in the earlier parts of the game where I did not have enough troops to continue conquering rest of AIs' lands. Even a single woodsmanIII, medicI promoted unit would have been much help, because healing from one woodsman III and a seperate medic I unit do not stack.

Madurai captured at 1120AD, Hyderabad and Leiria at 1125AD, Madras at 1130AD.

Joao's stack arrives from the depths of his base to face my evenly matched stack by 1125AD. I can't post the beautiful screenshot due to the 5 file attachment limit per post. I had 3 very exciting battles in this game:

1) vs Genghis at Karakorum siege.
2) vs Isabella near Elephantine, destroyed her Conquistador stack, while her stack of 15-20 units were wounded trying to capture one of DeGaulle's cities.
3) vs Joao at 1125-35AD in the open field.

Maximizing score:

I realized at this point that victory was very near and wanted to maximize my score through capturing as many cities as I could before crossing the domination land limit. Newly captured cities avoided building theaters or working any artist specialists. Thus, empire territory did not exceed the domination land limit for quite some time. The newly captured cities added to total population, but did not add much to the land limit, because they were revolting or even after the revolt, they did not grow culturally, except the religious culture bonus.

If I did nothing, my score was falling around 1-2K per turn, thus I had to be quick in capturing as much population as I could and claim victory. To settle some tiles outside cultural borders at final turn, 3 settlers were produced.

1130AD: Karakurum gives birth to the 7th Roman GP, a GreatEngineer. It appears clearly now saving the earlier GPs for future Golden Ages was not necessary, because the Parthenon and Roman forum helped with extra GP generation up until late game.

If I had the chance to replay this game, I would have used the first GreatArtist right ahead for an early GoldenAge. Furthermore, I would have infiltrated an AI with my GreatSpy to steal 5-6 techs. The remaining GPs would have been saved for late game Golden Ages.

1135AD: 9th GG born in Orleans, the city of 5 GreatProphets. My golden age ends same turn.

1145AD: Varanasi captured, 1150AD: Guimaraes captured.

1160AD: Leiria recaptured, had not left enough defenders... Great Artist appears in my GP farm.

1170AD: Vijayanagara captured. 1175AD: Lagos captured.

1180AD: Calcutta and Delhi captured. Indians wiped out. I discover Steel and upgrade all my Trebuchets to cannons. Cannons are simply overpowered with CityRaider III.

1185AD: Having access to steel, I rush IronWorks at Karakorum with the GE who arrived earlier. I also start my 2nd Golden Age using my remaining GreatPeople. Braga captured same turn.

1195AD: Joao's capital Lisbon fell.

1200AD: Evora captured, and Printing Press discovered.

1210AD: Oporto captured.

1220AD: Scientific Method stolen. Land domination limit, 49.24%

1225AD: Toledo and Abydos captured. I notice my score has passed 500K and decide to secure the win. I run artist specialists in my recently conquered, below 30 culture cities. Also, I settle 3 new cities outside my cultural borders.

1230AD: Happy end with 52.73% land, exceeding the domination limit of 51%. Population limit was exceeded by 62.63%-35%=27.63%

AND THE HIGHEST SCORE IN HOF HISTORY :king: (as of February 10th, 2008)
502 553

168465

Pouring my CivIV wisdom into this guide took just as much, if not more, time as finishing this game, so I would appreciate your 5-stars vote :-)

VirusMonster
Jan 28, 2008, 03:41 AM
Reserved for OP

cabert
Jan 28, 2008, 03:43 AM
did you check the HoF tables completely?
i think a few are above the 500k

edit : actually there are a few near 500k but none above.

VirusMonster
Jan 28, 2008, 04:08 AM
well there is no 500K+ entry in the main HOF tables.... I've just looked up all top scores in the opening page of the HOF tables for all difficulty levels. I am the first to cross the limit... I feel so proud =) There is the A_Turkish_Guy's submission, but it is on beta and on Warlords difficulty with cheesy tribal hut abuse =)

Anyway, I am also top scorer in the Immortal standard and now I've just finished scoring 500K at Huge Immortal =) It was time for a writeup =)

I am waiting for my submission approval, then I will post the writeup...

cripp7
Jan 29, 2008, 11:13 AM
can you give a link because I cannot find it in t HOF. Great job by the way!
Cripp

VirusMonster
Jan 29, 2008, 06:13 PM
HOF announces submissions every 10th and 25th of the month. I will have my writeup posted very soon tough. I can put the saves up here too, but to really believe me, you would have to wait the 10th of February =)

Lord Chambers
Jan 29, 2008, 10:49 PM
Organized Religion(+25% to wonder production)

I checked in Worldbuilder, and it turns out you're correct. Organized Religion, which the game says provides a benefit only to buildings, provides a 25% bonus to wonders. I use Organized Religion way more than the majority of the board seems to, and now I'll prioritize it when I'm trying to build wonders as well.

VirusMonster
Jan 29, 2008, 11:04 PM
=) yep, I actually produced a single missionary which I skipped in this report just so my Parthenon city could get Buddhism on time before the chopping was done. Too bad DeGaulle cancelled the Marble trade before the chopping was done. He had 2 free marbles at one point however. I could have even chopped Great Library if he kept providing with Marble =)

Organized Religion is great for peacetime building.

alex sword
Jan 30, 2008, 04:40 AM
:crazyeye: Uh-oh. Respect! ONe question. Does this approach work on Deity? Perhaps, it won't provide you with highest score but I wonder if it does work at all?

VirusMonster
Jan 30, 2008, 09:23 AM
=) yea thank you for the respect... I haven't tried deity except in the old patches where I used to Inca rush. With the new BTS AI, Inca rush is countered with hardcore poprushing, so you can't capture as much land as easily. Maybe someone will still be able to do it, but probably not me.

but theoretically speaking, any imperialistic leader can chop the settlers quickly. I would say imperialistic leaders have an advantage at deity difficulty, because of the early good start. They can also start producing CityRaider III troops quicker than most of the other leaders. The 2nd trait matters less, but it would help if it something that boots your economy, but almost every skill boosts your economy in one or another way in this game =) even protective has it's uses.

there are many quality threads in the forums... try to read the good player's posts and come up with your own style.

I have read the threads on getting to Liberalism first, followed by Nationalism popup, followed up teching to Rifleman before the AI does. I actually tried it in my Large Immortal game, which I submitted to HOF, but could not dominate as early as I wanted.

I think the limiting factor to getting to domination land limit is not the rifleman. well, rifleman before anyone has got them would definitely do, but I think it is the core production and sheer numbers that can make any strategy work. I wished you could see my power graph at various stages of the game. I produced units non-stop in all my cities skipping any infrastructure other than Courthouses, Granaries and Libraries. Maceman + knights + trebuchets + musketman + spies can win any game with the right numbers.

And do not forget the lvl III CityRaider Trebuchets, they are so so so crucial to this writeup. Too bad I have been upgrading them all wrong with Barrage. For every Trebuchet lost, I was killing 6-7 units.

shyuhe
Jan 30, 2008, 11:58 PM
Impressive victory VirusMonster :goodjob: I don't play for domination victories very frequently these days but I may have to boot up some domination games to see how close I can come to your monster score. I think my personal best is around 200k on a non-regenerated start. With an optimized bfc, I think I can push 300-400 but 500 :eek:

Your overcrowded map is a good idea for high level domination games though. If you can take out 2-3 AI then you have a reasonably good chance of winning the game since you'll have so much greater raw production compared to the other AI. Plus, high level AI have huge stockpiles of gold that you can fleece by selling old techs. I will definitely give this a try myself (though probably on emperor as warm-up before I try huge immortal).

VirusMonster
Jan 31, 2008, 12:26 AM
yea I could not believe my score either =) The closest I got was 367K, a standard Immortal domination win. I submitted it to HOF actually, and that score shows up in the tables now.

As you said, it is a good idea to try lower difficulty or smaller map sizes before making it harder for yourself. I lost several games before getting comfortable with this strategy, some very frustrating ones, but it payed off =) And regenaration can get boring at times, because the 2 gem starts are really rare, so really try your best once your preferred starting location appears.

Your overcrowded map is a good idea for high level domination games though. If you can take out 2-3 AI then you have a reasonably good chance of winning the game since you'll have so much greater raw production compared to the other AI. Plus, high level AI have huge stockpiles of gold that you can fleece by selling old techs. I will definitely give this a try myself (though probably on emperor as warm-up before I try huge immortal).

I agree with all you say. Here is one more benefit of playing a crowded map:

Look at my 500BC financial screenshot. The high number of AIs in the game also affected the ResourceTrading profits significantly. The ResourceTrading profits make up for 1/3 of all my expenses, the result is a 30% research boost.

VirusMonster
Jan 31, 2008, 01:56 AM
Impressive victory VirusMonster :goodjob: I don't play for domination victories very frequently these days but I may have to boot up some domination games to see how close I can come to your monster score. I think my personal best is around 200k on a non-regenerated start. With an optimized bfc, I think I can push 300-400 but 500 :eek:

Your overcrowded map is a good idea for high level domination games though. If you can take out 2-3 AI then you have a reasonably good chance of winning the game since you'll have so much greater raw production compared to the other AI. Plus, high level AI have huge stockpiles of gold that you can fleece by selling old techs. I will definitely give this a try myself (though probably on emperor as warm-up before I try huge immortal).


what does bfc stand for btw?

fed1943
Jan 31, 2008, 04:11 AM
Congratulations! Very well played.
If from a quick reading I'm right, is mostly a warmonger strategy, backed by
religious civic, and a good economic plan to avoid an economic collapse.
It looks to depend on a hammer capital with commerce/happy resources. Or,
maybe not, while the score should surely down. I shall try it.
Best regards,

Gone Dark
Jan 31, 2008, 04:17 AM
what does bfc stand for btw?

Big fat cross.

VirusMonster
Jan 31, 2008, 04:18 AM
yea the capital is a strong hammer city; I run Bureaucracy for some time and late game switched to Vassalage for the synergy it provides with Organized. I used PoliceState+Vassalage+Theocracy+Slavery for a long time for empirewide unit production.

So the ideal starting location should have around 6 hills, 1 food resource and 2 gems or 2 gold or 1 gold and 1 gem.

VirusMonster
Jan 31, 2008, 04:21 AM
big fat cross as in the sketch below? The shape resembles a cross? Is that why? =) I guess it makes sense. C for city center and o for outside cultural borders.

o---o
-----
--C--
-----
o---o

cabert
Jan 31, 2008, 04:27 AM
o x x x o
x x x x x
x x x x x
x x x x x
o x x x o
It does look like a big, fat, cross, doesn't it?

Gone Dark
Jan 31, 2008, 04:56 AM
big fat cross as in the sketch below? The shape resembles a cross? Is that why? =) I guess it makes sense. C for city center and o for outside cultural borders.

o---o
-----
--C--
-----
o---o

Exactamundo.

And I forgot to congratulate you on the game. How long did the game take?

VirusMonster
Jan 31, 2008, 05:02 AM
I finished in 2 sessions; I had the PC open and went eating, chatting, doing other stuff probably more than 50% of the time. civ4 timer shows 25 hours..

but probably the learning curve took longer =)

VirusMonster
Feb 16, 2008, 06:39 PM
alright, I see the saves are up in the HOF page =) now this game officially has the highest score in the HOF tables.

http://hof.civfanatics.net/civ4/game_info.php?show=general&dsply=0&entryID=10520

VirusMonster
Feb 18, 2008, 09:42 PM
5 more saves added... some minor edits to the text...

and thank you Thunderfall for including this writeup in the war academy :-)

CliftonBazaar
Feb 23, 2008, 05:54 AM
Congratulations; that's one heck on an achievement, especially with the new BtS AI!

VirusMonster
Feb 28, 2008, 08:48 AM
Congratulations; that's one heck on an achievement, especially with the new BtS AI!

The first EQM congratulates me :) I am honoured.

WastinTime
Feb 28, 2008, 10:03 AM
Nice work. I was wondering if you learned anything about scoring differences in BTS. I think BTS got a huge score bump, but I'm not sure. Your game is a great achievement and your writeup too. I hate to be the one to rain on your parade, cus I don't know for sure, but 500,000 on BTS might only be like 300,000 on vanilla. Some investigation needs to be done.

VirusMonster
Feb 28, 2008, 10:14 AM
I like the rain :) I will investigate whether the scoring got a bump in BTS, but you are the first one I heard this from. On the contrary, the AI is poprushing units much more effectively in BTS, and inital rushes fail more than they did in Vanilla(well not this game, but in general).

and you are one of the best players in the forum too, so thank you for your kind words as well :) I enjoyed studying your deity cultural victory games.

edit: I found this thread on scoring discussion between Vanilla/Warlords and BTS. BTS vs. Vanilla/Warlords Discussion (http://forums.civfanatics.com/showthread.php?t=237526)

My conclusion of the discussion in this thread is that BTS and Vanilla/Warlords scoring systems are hard to compare, yet probably more or less equal. According to the thread, the math behind scoring in both BTS and Vanilla/warlords seems the same. Base score calculations (land, pop, tech, wonder) seem the same as well.

BTS is harder to score, because of the better AI, defensive AI poprushing, espionage spending, the new tech requirements, slightly lower traderoute profits according to UnconqueredSun (see his post at traderoute profits are less in BTS (http://forums.civfanatics.com/showpost.php?p=6453284&postcount=88), and exponential inflation (was linear in Vanilla). On the other hand, BTS has a few more turns added toward the end of the game, so total number of turns in a BTS game is more than a Vanilla game. Consequently, when you do the scoring calculation (current turn/total number of turns), a slightly(by how much?) lower value is returned in BTS. This value, according to the discussion in the thread, seems negligible, but nevertheless increases the final score by a little amount compared to a Vanilla game. Here are some graphs:

date vs. number of turns graphs on different games speeds for Vanilla/Warlords vs. BTS (http://forums.civfanatics.com/showpost.php?p=5824731&postcount=18)

and the original scoring system explained post:

Civ4 scoring explained (http://forums.civfanatics.com/showpost.php?p=3354736&postcount=82)

Let's not forget the random events... I think random events could both increase or decrease the final score in BTS, so I am undecided on their effect to the final score. For example, EliteSwords mission could be very strong if you get it while preparing your initial Praetorian rush, but in this game, I don't think any quest or random event affected the final score much, well I got the lucky Greed quest :) it gave me 6 free unpromoted Praetorians :) Probably saved 20 turns for producing that many units.

I also feel the city revolt mission could be increasing the speed the cities are captured compared to Vanilla where you had to produce 10-20 cats just so you could reduce the city defences in 4-5 turns. I still use the siege for the colloteral damage, but I wait less time reducing city defenses.

gtg now, take care...

WastinTime
Feb 28, 2008, 02:35 PM
If it matters, I find random events to be a positive effect, making the game easier. (Too small to effect score enough for this discussion though.) The fact is, almost any bad event can be bought off with like 10 gold. There is very little risk.

I looked at that formula too and I think I found the big difference. The currentTurn/Maxturn is the big change. (In other words, how quickly you finish the game) is exponential! BTS moved the max turn out as much as 300 turns. That probably gives the huge bump in score I perceived.

If my calculations are correct, a score of 500,000 on BTS would get roughly 346,000 in vanilla just due to this one factor. I haven't studied the formula that closely, so don't quote me on those numbers....yet.

When it rains, it pours.

WastinTime
Feb 28, 2008, 03:13 PM
My next rough estimate sets 500,000 on BTS to be just over 250,000. It all depends on what turn to end on.

Let me do your exact game, finish on turn 546.

Turn/MaxTurn is .364 BTS and .455 Vanilla

The short answer is that causes your score to almost double the Vanilla score on a game ending around 546 turns. An earlier finish will have less effect. Late game scores would be 4 times greater in BTS.

The increased # of tech and # of wonders does lower your BTS score a bit, so I'd estimate 500,000 BTS is right around 300,000 in vanilla at turn 500-600. Hey, right between my first 2 rough estimates.

The slight timeline changes also work against you, making the BTS score even more inflated...another long story.

edit: Let me add a plug for separating BTS HOF completely. Score is just another reason they are too different to combine in one set of tables. Just like Space Race had to be split into Space Colony. BTS scores will bury the greatest Civ IV vanilla scores. The best games in that category will be forgotten. Same goes for Conquest, but it's Vanilla that buries BTS. I can make a case for culture too. We are not honoring the all the right games in the HOF...my opinion. Treat BTS like CIV 5. You would never try to combine Civ 3 and 4.

WastinTime
Feb 28, 2008, 03:36 PM
Take a look at some examples: (maybe Kamino is the greatest score hound ever, but you be the judge):

Duel Diety:
1 Kamino 443464 760 AD
2 WastinTime 249260 680 AD

I remember my only goal that game was high score, so its not like I just got domination quickly. I was milkin' it. I even finished earlier than Kamino, but he got 443,000!

Tiny Immortal:
1 Kamino 479838 400 AD
2 Bozso77 261266 1286 AD
3 randomlyaware 251583 410 AD

Guess who played BTS?

VirusMonster
Feb 28, 2008, 07:32 PM
ok, I have just read your post... you could be right, but I need to dig out the math first, because your rain is washing all my charisma here on this thread :) jk... I am also curious what my finish date corresponds to in Vanilla, since apart from the highest score, I also seem to have the earliest finish date for Huge Immortal Domination.

I also wondered how Kamino got those high scores... one of his Immortal Tiny submissions almost reached 500k... it was something like 479k... ah you already mentioned that game in your post. I opened up his saves, and my idea is he is killing all enemy units with quechuas and leaves them down to 1 city.. Then he milks the whole map without crossing to domination. Did you apply a more or less similar tactic? Since with BTS 3.13 installed, I can't open your vanilla saves, can you load up your game rivaling the 479K score by Kamino and tell me the land, population, tech, and wonder ratios at final turn?

Also, can you tell me for my future analysis how you know that my game ended at turn 546? Where did you get this number? and where did you get the max turn numbers for vanilla and Bts? I want to open Matlab and plug this scoring formula in for Vanilla and BTS :)

In the links I brought up, there is probably sufficient information to bring this confusion to an end...

----
now reading the edit to your last post... I am not sure why space coloy and space race got split btw... Can you enlighten me and other readers of this thread?

----

I agree with your observations on quests, they only seem to make the game easier except the few negative 10-20 gold ones. On the other hand, the AI gets them too, no?!

----

And don't underestimate that most Vanilla scores , especialyl those super quick BC large and huge domination games by Incans are all achieved through AI not putting more than 2-3 archers per city :-/ I am not trying to say what you say on the scoring system could not be true, rather I have felt while trying to beat previous scores in the deity tables on BTS( and trust me I tried many times) I usually end up facing a lot more defenders than in Vanilla, so I could not dominate as early or maybe(almost always) fail.... The defending AI poprushes much more effectively and it greatly delays your domination finish date... however, I am not sure whether this is a strong enough reason to balance out finish turn/max turn getting lower in BTS. I started to agree with you that BTS scoring should be thought as seperate than Vanilla scoring...

------------

I also think there is some truth to HOF not honoring the right games, and tomorrow, I will crank out the math :) Take care for now, it rained where I live today, I love rain :)

WastinTime
Feb 28, 2008, 07:47 PM
I did the same tactic for a high score as you described.

To get turn 546, I look at your "player log" tab on the game submission.

Total turns show at the top of the screen in-game (if you turn them on in the options.) 1200 vanilla, 1500 BTS for marathon.

Space Race/Colony are split because no one thinks a Space Colony game will ever beat the old Race. It changed too much.

Beating the Immortal BTS AI on Huge was great work. Your writeup should focus on that achievement and give advice towards that goal. Your game doesn't really do much to milk the score, so the title about 500K is misleading. I think 500K will end up being just an average score (for a huge domination). I think someone with a lot of time on their hands will come along with 800K+, probably 1 million.

Kamino
Mar 01, 2008, 01:13 AM
Hi all,

First, congratulations to VirusMonster for this nice game.

I am happy there is another guy who compete for high score.
It will be a challenge for me to score all the immortal line. :)

I agree with you on the settings except :
1) Low sea level,
2) Despite the smarter AI, I think Quechua rush is still, and by far, the best strategy. AI are harder to conquer, but easy to cripple. One Q is enough handle an immortal AI. He is the best worker stealer (Q with Woodsman II is a nightmare for AI :D)

@ WastinTime :
It is obvious things are different in BTS.
You can't compare BTS and Vanilla/Warlors.
Scoring is easier.
Conquer/Domination date is more difficult for big maps on high levels.
AI is smarter.

Sorry for my poor english (I am french)

VirusMonster
Mar 01, 2008, 01:30 AM
Oooo.... mr. Kamino arrived as well :) I love the French... Thank you for your kind words mister.

I am cranking out some numbers to calculate the effective score for BTS and Vanilla at the moment... it will take some time to avoid any mistakes, but I am on track...

nice new tip for quechua rushers btw :) everyone do woodsman II from now on :) I also agree with you that low sea levels would probably have been better for higher scores.

VirusMonster
Mar 01, 2008, 06:23 AM
Hi all,

As promised, I plugged in the whole scoring formula into Matlab and am confirming that WastinTime is right :) I think I should make a seperate article for this topic and post in the HOF forum.

Here are my particular findings:



% Matlab code for scoring BTS vs. Vanilla/Warlords
% You can copy paste this code straight into Matlab Student edition, and it will do the math and get the numbers out.

popfactor=5000;
landfactor=2000;
techfactor=2000;
wonderfactor=1000;

% data from the 500k game
popmaxscore=1227;
landmaxscore=1720;
techmaxscore=334;
wondermaxscore=310;
poprawscore=677;
landrawscore=757;
techrawscore=120;
wonderrawscore=125;
turnonwin=546;
maxBTSturns=1500;
maxVanillaWarlordstuns=1200;
immortaldifficultyadjustor=1.8;

% not sure about initial pop,land, tech, and wonder values so I calculated for 2 possible scenarios

popinitialscore0=0;
landinitialscore0=0;
techinitialscore0=2;
wonderinitialscore0=0;

popinitialscore1=1;
landinitialscore1=21;
techinitialscore1=2;
wonderinitialscore1=5;

popscore0=(popfactor*poprawscore)/(popinitialscore0+((turnonwin/maxBTSturns)*(popmaxscore-popinitialscore0)))
landscore0=(landfactor*landrawscore)/(landinitialscore0+((turnonwin/maxBTSturns)*(landmaxscore-landinitialscore0)))
wonderscore0=(wonderfactor*wonderrawscore)/(wonderinitialscore0+((turnonwin/maxBTSturns)*(wondermaxscore-wonderinitialscore0)))
techscore0=(techfactor*techrawscore)/(techinitialscore0 * power(floor(techmaxscore/techinitialscore0),(turnonwin/maxBTSturns)))

popscore1=(popfactor*poprawscore)/(popinitialscore1*power(floor(popmaxscore/popinitialscore1),(turnonwin/maxBTSturns)))
landscore1=(landfactor*landrawscore)/(landinitialscore1*power((floor(landmaxscore/landinitialscore1)),(turnonwin/maxBTSturns)))
wonderscore1=(wonderfactor*wonderrawscore)/(wonderinitialscore1+((turnonwin/maxBTSturns)*(wondermaxscore-wonderinitialscore1)))
techscore1=(techfactor*techrawscore)/(techinitialscore1*power((floor(techmaxscore/techinitialscore1)),(turnonwin/maxBTSturns)))

totalscore0=popscore0+landscore0+wonderscore0+tech score0
totalscore1=popscore1+landscore1+wonderscore1+tech score1

adjustedscore0=immortaldifficultyadjustor*totalsco re0
adjustedscore1=immortaldifficultyadjustor*totalsco re1

% adjustedscore0 = 53515
% adjustedscore1 = 519280 so let's assume this score is almost correct, because it is very close to the actual 502553 in game score...

% now with maxturns=1200 for vanilla and warlords, and the adjustedscore1 becomes: 279740

%Conclusion: WastinTime is right unless the 519280 score randomly appeared so close the the actual 502553 score :)



Although I am getting a very close number, I can't replicate the exact 502553 ingame score at Matlab, since I am not sure about the initial population, land, tech, and wonder scores.

They could be 0, 0, 2 (for 2 ancient era starting techs), 0 or 1(for capital initial population), 21(for initial land squares), 2(for 2 initial ancient era techs), and 5 (due to palace national wonder). I calculated for both set of values and found that the 1, 21, 2, 5 combination gets a very close score to the in game final score of 502 553.

Another reason I can't replicate the exact in game score could be rounding errors or lack of information about the actual scoring formula. I did check score calculation formulas for both BTS and Vanilla/Warlords in the XML and Python files. There does not seem to be any change except the max turns getting 1500 in BTS.

so in conclusion =) WastinTime is right :) While BTS games might be somewhat harder to win due to various reasons mentioned in previous posts, BTS scoring system is significantly more advantageous than Vanilla or Warlords.

I must, however :), defend that this Huge Immortal domination game ended at turn 546, while the only other submission for Huge Immortal Domination, despite being played on Vanilla huh ;), ended at turn 673. For large Immortal submission, the best score is 259381, and it still is lower than the normalized score for this game, ie assuming maxturns for BTS is 1200 :) So as WastinTime suggested, this game is still very good for Huge Immortal Domination, both for Vanilla and BTS. I hope it will stay that way for a long time.

neckface
Mar 04, 2008, 10:04 AM
:clap: :clap: Well done, 5-star writeup! :thumbsup:

VirusMonster
Mar 06, 2008, 02:44 AM
:clap: :clap: Well done, 5-star writeup! :thumbsup:

Thank you bro :) I have just solved the whole dilemma on spy math and updated the article accordingly. Make sure to read the recently updated Espionage Economy vs. Manual Research (http://forums.civfanatics.com/showthread.php?t=260902#EEvsManualResearch) section of the article, then go build some spies :-) Updated contents are:

technology cost explained
espionage base cost explained
arithmetic of the espionage mission cost modifiers explained
espionage mission cost modifiers explained
espionage point spending modifier explained
:commerce: to :science: conversion ratios compared between EE and manual research
most effective way for generating :espionage:
spy detection analysis with links to game mechanic articles
mission success rate analysis with links to game mechanic articles
spy production and maintenance costs analysis for the average Medieval scenario
pure EE vs. hybrid EE

oyzar
Mar 13, 2008, 05:20 AM
Nicely done, minor quibble is that beakers actually get multiplied by 20% just from having one prereqs(and 20% more for each additional optional prereq) so beakers are a bit higher than you show in your analysis. Additionaly burrecracy capital in a small empire often carry most of the weight of research(with an academy having 75% base research bonus) giving between (0.5*1.75+1.25*0.5)*1.2=1.74 without other know modifier and 2.262 beaker per commerce in the mid game. Getting those espionage multipliers doesn't always come free, although sometimes it do... Obviously spy spceialists still rock but not as hard as you say, only need a 5+ commerce title to match it if you get favourable conditions... They still by far outpreform scientists by their raw output but in the long run academies are sometimes very good, contradictory to popular belif though you are right in that if you are not going to use the great scientist for some specific military advantage(liberalism slingshot for something, for example) or for academies, you are better off running spies.. Unfortunatly there is no way to run alot of spies early in the game and as such using spies for great people is nowhere near as fast as if you make scientists/merchants/artists as these are way easier to get alot of slots for. This part of your analysis is a bit too optimistic in the light of the EE. The analysis of the cost of spies is rather good, except for the loss of a cottage doesn't actually translate into lost cottage turns but into lost town turns as it will take longer for the cottage to reach town etc. so "Say this citizen was working a cottage with +3 commerce, then the total lost commerce for 12 turns becomes: 12*6*3=216." I am guessing this is a riverside hamlet or a village, but lets say it is village, then you only lose 1 more commerce as the village will still likely grow to town before you get printing press 12*6*4=288 so a minor difference, but the main gist of it still holds true. I belive against the AI EE can still be exploited but it isn't as totaly dominating as you say here, if there were more spy slots available earlier in the game it might have been different but as it is, that isn't the case and as such normal research can often do what you want better when you factor in the delay in getting the techs as well as the AI's not actually researching what you want. In adition due to the rather high risk factor EE have very high variance and as such it could be very good for trying to milking the game as good results from stealing is indeed favourable compared to research. High variance means you will have some great results and some bad reults, with normal research you don't have the same variance and even if normal research would on average outpreform EE, in the lucky case EE will win as normal research don't have the oportunity to get lucky.

I am sorry for the lack of structure in the post but hope it is somewhat understandable. Nice game anyways. I certanly don't have the patice a huge marathon game would need so big props for that.

VirusMonster
Mar 13, 2008, 06:09 AM
nice post, oyzar... thank you for your corrections... I'll look up what you said about prerequisite bonus arithmetic and update the article soon... proofreading & editing this article took me forever, so I really appreciate your feedback :-)

VirusMonster
Mar 13, 2008, 08:02 PM
Nicely done, minor quibble is that beakers actually get multiplied by 20% just from having one prereqs(and 20% more for each additional optional prereq) so beakers are a bit higher than you show in your analysis. Additionaly burrecracy capital in a small empire often carry most of the weight of research(with an academy having 75% base research bonus) giving between (0.5*1.75+1.25*0.5)*1.2=1.74 without other know modifier and 2.262 beaker per commerce in the mid game. Getting those espionage multipliers doesn't always come free, although sometimes it do... Obviously spy spceialists still rock but not as hard as you say, only need a 5+ commerce title to match it if you get favourable conditions... They still by far outpreform scientists by their raw output but in the long run academies are sometimes very good, contradictory to popular belif though you are right in that if you are not going to use the great scientist for some specific military advantage(liberalism slingshot for something, for example) or for academies, you are better off running spies.. Unfortunatly there is no way to run alot of spies early in the game and as such using spies for great people is nowhere near as fast as if you make scientists/merchants/artists as these are way easier to get alot of slots for. This part of your analysis is a bit too optimistic in the light of the EE. The analysis of the cost of spies is rather good, except for the loss of a cottage doesn't actually translate into lost cottage turns but into lost town turns as it will take longer for the cottage to reach town etc. so "Say this citizen was working a cottage with +3 commerce, then the total lost commerce for 12 turns becomes: 12*6*3=216." I am guessing this is a riverside hamlet or a village, but lets say it is village, then you only lose 1 more commerce as the village will still likely grow to town before you get printing press 12*6*4=288 so a minor difference, but the main gist of it still holds true. I belive against the AI EE can still be exploited but it isn't as totaly dominating as you say here, if there were more spy slots available earlier in the game it might have been different but as it is, that isn't the case and as such normal research can often do what you want better when you factor in the delay in getting the techs as well as the AI's not actually researching what you want. In adition due to the rather high risk factor EE have very high variance and as such it could be very good for trying to milking the game as good results from stealing is indeed favourable compared to research. High variance means you will have some great results and some bad reults, with normal research you don't have the same variance and even if normal research would on average outpreform EE, in the lucky case EE will win as normal research don't have the oportunity to get lucky.

I am sorry for the lack of structure in the post but hope it is somewhat understandable. Nice game anyways. I certanly don't have the patice a huge marathon game would need so big props for that.

I rearranged the :commerce: to :science: conversion ratio discussion to include a (+12%) spy maintenance modifier. I also changed the cottage +3 commerce to village +3 commerce. At that point in the game, this excess citizen will most likely work one of the weaker tiles, since other tiles will be already occupied through other citiziens, thus no riverside villlage.

Then, I run out of character length in the 3rd post of the article :)

I might rearrange the 3rd post to include:


Small empires benefit more from Bureaucracy effect to higher :commerce: to :science: conversion ratios (your 2.262 ratio is very realistic I think, so on small maps, manual research>>EE)
Bureaucracy is not the best large&huge map size civic, since to dominate a larger map size, you'll need more than a single production center.
Along with the previous point, Vassalage might be the better civic than Bureaucracy in large&huge maps and if empire land size is among the largest
On a large&huge map, if you are stuck with a relatively small empire, manual research is still makes more sense than EE
EE has more luck component than manual research, thus increase your espionage power rating for less variance in EE results. Either go pure EE or go pure Manual Research.
Vassalage+Organized synergy helps to cut down total expenses by 6-10%. Assuming early game 1:1 :gold: to :commerce: conversion rate, an organized leader is better off with Vassalage and EE.
After Nationhood, stick with Nationhood for the +25% :espionage bonus and drafting. Also switch to almost pure EE at this point.
After Nationhood, focus research Constitution as soon as possible to make the EE shine and to be able to assign 3 spies per city with Courthouses and Jails.
Holy cities are very important for EE, ie focus warmongering to capture holy cities. Holy city = 2.95 :commerce: to :science: ratio, highest in the game
Early game infiltration missions are imba, because it gives immense espionage power superiority that cuts your espionage costs and reduces spy detection chances. Don't settle, don't build the ScotlandYard, infiltrate instead! :)

VirusMonster
Mar 14, 2008, 10:05 AM
Oyzar, thank you again for your feedback. The 1.2 prerequisite modifier does make the ManualResearch significantly better against EE. Average case of EE still outperforms ManualResearch, but as you pointed out, in a hybrid EE, the variance in the mission success rates can be risky. A (mainly) pure EE then becomes much more preferable than a hybrid EE to reduce spy detection chances even further.

I have updated the article to include considerations to the poinst you raised. Check the new table of contents here (http://forums.civfanatics.com/showthread.php?t=260902):


technology cost explained
espionage base cost of technology stealing explained
city revolt base cost explained
espionage mission cost modifiers explained
espionage point spending modifier explained
arithmetic of the espionage mission cost modifiers explained
religious modifier arithmetic explained
spy detection analysis with links to game mechanic articles
mission success rate analysis with links to game mechanic articles
tips for low spy casualties
spy production costs for the average Medieval scenario
spy maintenance costs for the average Medieval scenario
total mission overhead
luck factor in EE
:commerce: to :science: conversion rates for EE and Manual Research
Bureaucracy effect to EE and Manual Research compared
effect of larger map sizes on Bureaucracy economy
Vassalage+Organized+EE synergy
summary comparison of conversion rates between EE and Manual Research
most effective way for generating :espionage:
pure EE vs. hybrid EE


edit: spread culture mission effect to mission cost modifier added. Each spread culture mission adds +5% to city culture. 5 consecutive missions reduce the mission cost roughly by %5*0.5*5=12.5% Spread culture missions make the EE even stronger, since they are very cheap, and their effect carries over to future missions at no cost.

ck07
Mar 14, 2008, 09:24 PM
I began with the assumption that if I caused a revolt all the defending units would take 50% damage, just as they do in 'normal' revolts. But when I did it they took no damage.

Have I missed something?

VirusMonster
Mar 15, 2008, 06:39 AM
Spy revolt nullyfies city defenses for 1 turn. Say city had walls(+50%) or cultural defense(+60%). When you use the spy for a city revolt mission (base cost 500 :espionage: points), these defense bonuses don't apply to the defenders. Without these defenses, it becomes much easier to capture the city.

Units take no damage from the city revolt mission, but their ability to use the city cultural or walls defense goes away for 1 turn.

edit: I updated the section on "capturing cities with minimal losses" to include what city revolt missions do.

killercane
Mar 18, 2008, 02:14 PM
I think I will try to play some of these. What were your experiences with Darius? I dont know if someone else asked but why not continue on to Biology and milk a bit before victory?

VirusMonster
Mar 18, 2008, 05:39 PM
I think I will try to play some of these. What were your experiences with Darius? I dont know if someone else asked but why not continue on to Biology and milk a bit before victory?

I am currently trying a HOF Deity game (small size, continents map) with Darius. I'll post a writeup if I succeed winning :-) There is a thread I recently created where I asked for feedback before the game is over.

Could I have milked this game with Biology? hmm tough question... I might test it out and report how much higher, if any, I could get with Biology. I was losing 1-2k score per turn at end game, so I thought waiting for Biology 5-6 more turns would only reduced by score below 500k, but again, I could have gotten Biology much earlier. My goal was crossing 500k, not particularly milking.

I also tried to be honest with my suboptimal play in this game. For example:

late first GP and consequently late academy
had too many GPs waiting, doing nothing. when game ended I had 2 extra GPs and was in the middle of a Golden Age
did not switch to Mercantilism for no reason
missed to build the Great Library, because I traded Aesthetics with the Marble trading AI before my chops were over, etc...)
could not declare war on Peter on time, lost 10-15 turns because his vassal agreed to one of my demands (probably a gold demand)


I could have probably scored a bit higher, 550k? 600k? :) 600K would be my future goal if I ever attempt again... I want to give other leaders a chance before I try Caesar again.

killercane
Mar 21, 2008, 05:07 AM
Well your article certainly points out some interesting and enlightening things. I am playing Darius, and find him quite effective at retarding the first 8 or 10 AI. I have 17 cities at 500 BC, just revolted to CS, and am running at about 10 turns to acquire paper if the gold holds up. Still no courthouses as we are not at 0 gold flow yet. Currency is really a key tech, and allows for 100% research for quite some time selling to the retarded AIs.

VirusMonster
Mar 21, 2008, 06:30 AM
Well your article certainly points out some interesting and enlightening things. I am playing Darius, and find him quite effective at retarding the first 8 or 10 AI. I have 17 cities at 500 BC, just revolted to CS, and am running at about 10 turns to acquire paper if the gold holds up. Still no courthouses as we are not at 0 gold flow yet. Currency is really a key tech, and allows for 100% research for quite some time selling to the retarded AIs.

17 cities at 500BC? :) what difficulty and expansion pack are you playing? and what mapsize? I only had 10 cities at 500BC in this game.

edit: techs researched at 500BC in my game were CoL, Literature, Calendar, Currency, and Monarchy. I did not switch to Caste System, because I wanted to utilize building poprushing in my recently developing cities. What techs did you have at 500BC?

2ndedit: I agree getting Currency early and trading for AI gold reserves is very important. I remember doing something along the same lines in this game as well.

killercane
Mar 21, 2008, 09:20 AM
17 cities at 500BC? :) what difficulty and expansion pack are you playing? and what mapsize? I only had 10 cities at 500BC in this game.

edit: techs researched at 500BC in my game were CoL, Literature, Calendar, Currency, and Monarchy. I did not switch to Caste System, because I wanted to utilize building poprushing in my recently developing cities. What techs did you have at 500BC?

2ndedit: I agree getting Currency early and trading for AI gold reserves is very important. I remember doing something along the same lines in this game as well.
Those +CS. Immortal of course. I didnt super expand until after alphabet and math. BTS AI is pretty terrible in the early going.

VirusMonster
Mar 21, 2008, 09:37 AM
What is CS? civil service or caste system , can you be more exact with your terminology?

edit: your game sounds like you'll beat my 500k score :) tell me more about it :) are you sure it is 500BC and not 500AD? What are the map size and map type? Did you not have trouble conquering with Immortals? and I am curious how you got that many cities even though you said you started conquering later than I did.

killercane
Mar 21, 2008, 01:29 PM
What is CS? civil service or caste system , can you be more exact with your terminology?

edit: your game sounds like you'll beat my 500k score :) tell me more about it :) are you sure it is 500BC and not 500AD? What are the map size and map type? Did you not have trouble conquering with Immortals? and I am curious how you got that many cities even though you said you started conquering later than I did.
Civil Service. Yes, 500 BC on huge pangea. Immortals are great :); the key is to never let the AI hook up metal. Send the first 8 to capture workers and pillage everything on the surrounding 8 or 9 civs. I produced 2 warriors first to capture workers, you have to get to them in the first 40 turns or so or they will cover the workers with archers. Its slow going but effective. Im sure a better player than I can score 750K at least once the optimal strategy is figured out.

Theoretically, you only need about 9 civs' land for domination. So if you leave the other 8 alone, they should be able to fund your research to biology via trade. I might try to figure out if the corporations are worth the research. Certainly from a score standpoint, BTS is a far different creature than previous incarnations of Civ 4.

Here are some autosaves, Im not much for writeups. Your writeup is quite excellent and inspirational, I know I couldnt compare to that.

VirusMonster
Mar 21, 2008, 02:33 PM
Civil Service. Yes, 500 BC on huge pangea. Immortals are great :); the key is to never let the AI hook up metal. Send the first 8 to capture workers and pillage everything on the surrounding 8 or 9 civs. I produced 2 warriors first to capture workers, you have to get to them in the first 40 turns or so or they will cover the workers with archers. Its slow going but effective. Im sure a better player than I can score 750K at least once the optimal strategy is figured out.

Dude, I am sure you'll pass my score this game :-) I have looked up at the 1000BC save, we seem to have same number of cities at 1000BC. I had 10, you had 9, but you have alphabet, mathematics, construction, and priesthood, whereas I only had Alphabet at that time. Your 1AD score is 60k more than 1AD score of this game.

Your starting position is insane :) 4 gold hills and 4 foodplains+cow :) Persians get the +2 health UB, so chopping the forests shouldn't be a problem. You got the first GreatScientist much earlier than I did (1310BC vs. 510BC), so the early academy helped big time with classical era research.

Theoretically, you only need about 9 civs' land for domination. So if you leave the other 8 alone, they should be able to fund your research to biology via trade. I might try to figure out if the corporations are worth the research. Certainly from a score standpoint, BTS is a far different creature than previous incarnations of Civ 4..

True, domination land requirement is 51% with 17 AIs, so beating 9 of them should be enough. Playing Low Sea levels will also give you more land at the end to milk with Biology.

Here are some autosaves, Im not much for writeups. Your writeup is quite excellent and inspirational, I know I couldnt compare to that.

Thank you for the saves and the compliment :) I hope my writeup would be a good reference people trying to cross the 500k barrier :-) I am glad you liked it and hope you pass 600k in your Darius game. Too bad I must deliver my world champion title to you pretty soon :-)

edit: I looked at your 1AD save further. Your research at equilibrium (30%) is 134 :science: My equilibrium research rate at 1AD was 225 :science:, but again you are ahead with so many techs and cities :) once your cities grow to the happy cap, your empire will be very strong. One downside I see is that you control no wonders at this point, but you are about to capture 4 of them at Rome. I had the Pyramids, Shwedagon Paya, and the Parthenon as key wonders in my game, but you'll still do fine without them. You have the Jewish holy city; I strongly suggest you switch to pure EE :-) Go Vassalage+EE. You should check out the :commerce: to :science: conversion rates comparison section in my EE vs. Manual Research article, it might convince you further. Good luck :)

killercane
Mar 22, 2008, 10:47 PM
Well I hate to hijack your thread but I finished with 704,550. Darn AI declared on me at the end with his 3 vassals so I had to end it. Went from 730k to 692k in that turn :( and it even registered on the final score graph. Luckily I was at 50.1%-50.3% for a long time and only had to take a couple of cities; the score was going up at 4K and 5K a turn so there is no telling what it would get to before bottoming out.

I think 900k is easily possible, and perhaps even a cool million. I reached Biology in the 800 or 900 range. I lost 7 turns in anarchy post medicine when a spy popped rather than a scientist. Almost all my GPs were for golden ages outside of Sids Sushi Merchant and Academy.

I disagree about espionage. Normally it is very good in say a multiplayer environment, but here getting to cavalry or curassiers quickly is essential, as is getting to Biology and Medicine. Immortal AI just doesnt research as fast on marathon as you need them to so you have to do most of it yourself with gold from them. Fast cavalry means you dont have to whip as much as you did in your game. No fooling with slow moving units means your war tech milestones are immortals->cats->maces->curassiers->cavalry.

On Sushi company, it seems much better than Cereal Mills. First it comes much earlier, and second you seem to have many more resources for it. In this I had 50+ resources which gave each city 13 food. For little filler cities and island cities with only seafood the effect is massive with a granary. The only problem is it costs so darn much. Next time a Wall Street and earlier banks will be needed as expenses became pretty tremendous. Timbuktu the corporation headquarters was something like 40 maintenance per turn AFTER court and organized bonus! Probably a golden age at the end to switch to environmentalism for the last push (cant earlier as the +25% corporate costs will kill you) is best.

So to sum it up: Darius, harass 9 or 10 AI, steal workers as much as possible, expand at a fast and steady rate and research at the maximum (use wealth if you need to). Get to cavalry and finish off everyone.

WastinTime
Mar 23, 2008, 01:06 AM
Interesting. What map did you play? and what date did you finish?

Wouldn't an earlier finish date score higher? or is the population boom from bio/sushi just too large.

VirusMonster
Mar 23, 2008, 03:51 AM
Congrats :) I can understand why you prefer manual research if you want to be first to reach Cavalry. Your hijacking is welcome :-)

killercane
Mar 23, 2008, 04:54 AM
Interesting. What map did you play? and what date did you finish?

Wouldn't an earlier finish date score higher? or is the population boom from bio/sushi just too large.
Just pangea. It would be nice to look at other things such as global highlands or some other map to see if you can get a significantly larger domination limit.

I dont see how winning earlier would help. Pop is your biggest boost, and the score was still going up significantly at the end here, 1155 AD. In just post biology phase it went up 200k IIRC.

Here is a screenshot from 975 AD and hitting 600k; I dont know what the per turn score increase is but it was pretty significant, a few thousand at least.

VirusMonster
Mar 23, 2008, 05:03 AM
Just pangea. It would be nice to look at other things such as global highlands or some other map to see if you can get a significantly larger domination limit.

I dont see how winning earlier would help. Pop is your biggest boost, and the score was still going up significantly at the end here, 1155 AD. In just post biology phase it went up 200k IIRC.

I believe the domination limit increases if you reduce the number of opponents on the map, but reducing the number of opponent AIs could make a different game, with different diplomacy.

edit: tested a pangea huge map with 10 AI opponents. The land limit became 56%, 5% increase compared to 17AI opponents.

WastinTime
Mar 23, 2008, 10:31 AM
Hmm. 17 AI gives you a few more free cities, and workers than only 10 AI, but isn't that much larger drain on your focus? I suppose it adds more resources to the map. More food might look good, but I think it raises the bar on population. A low food map scores better for the same population. But then the corporations are stronger with more food.

I'd think you'd want to beat them all down and then kill em all on the last turn. Would that give you a conquest? or Dom? Doesn't matter if you're going for score I guess.

couple quick questions: How much score penalty does immortal get compared to diety?

Is there a penalty for playing large instead of huge? or is huge just the best way to get a massive population?

killercane
Mar 23, 2008, 04:13 PM
Dunno the ins and outs of maps; Im thinking currently Terra will yield a nice score.

More land and more pop makes huge better than large. I dont know the score penalty for deity compared to Immortal, but the tradeoff there is you would need many more troops and infrastructure vs. AIs being able to help a bit more techwise.

VirusMonster
Mar 23, 2008, 04:40 PM
Hmm. 17 AI gives you a few more free cities, and workers than only 10 AI, but isn't that much larger drain on your focus? I suppose it adds more resources to the map. More food might look good, but I think it raises the bar on population. A low food map scores better for the same population. But then the corporations are stronger with more food.

If you read MrCynical's Power of Sushie article, he discusses a +20 :food: Sushi Cooperation from one of his games. KillerCane had 50 sushi resources in his game, 50*0.25=roughly +12 :food:. He was using low sea levels compared to my high sea level submission. I believe the random shore line could also play a factor. (I was using a solid shoreline).

In my Huge Immortal submission, Sushi would give 21*0.25=+5 :food:. In my large Immortal submission for HOF, I had 13*0.37=+5 :food: from Sushi. Again, for my standard Immortal submisson, I would gain 9*0.5=+5 :food: from Sushi. These last 3 figures are all from Pangea map type. I assume MrCynical's example is a water-rich map, possible archipelago, with definitely more sushi resources and can push for +15-20 extra :food: range.

As you can see from my examples, more or less all map sizes will yield the same number of sushi extra food. Map type, sea level setting, and shore setting seem to affect the number of sushi resources however.

The number of AIs have little effect on the number of Sushi resources, but since the capital start location generator usually puts food resources around, I assume maps with more AIs would yield slightly higher number of Sushi resources. The extra food resource will increase the maximum possible population score, but this effect is rather negligible compared to the +1-2 :food: effect through Sushi in all cities.

I'd think you'd want to beat them all down and then kill em all on the last turn. Would that give you a conquest? or Dom? Doesn't matter if you're going for score I guess.

I used this tactic you described in my game. As the end approached, I captured as many cities as I could without crossing the domination limit. I doubt you can conquer all, but you could definitely increase your final score significantly by 50-100k.

couple quick questions: How much score penalty does immortal get compared to diety?

Final score gets multiplied by 1.8 for Immortal and 2.0 for Deity.

Is there a penalty for playing large instead of huge? or is huge just the best way to get a massive population?

There is no penalty for map sizes. Larger map sizes increase the maximum population limit and consequently allow higher base scores, since each population unit adds +1 to the base score. Each land square also adds +1 to the final base score.

Low sea setting on Killercane's game compared to my high sea setting also increased the maximum possible population score from 1600sh to 1900sh. This %20 increase translates to a higher final population that could make up at least +50k in final score.

So, I believe best way to increase your score through Sushi would be on a water-rich huge map :-) But you would have to build ships there and tech for Astronomy, slightly deviating from the Sushi tech path. Immortals and Praetorians would be less effective as well.

Terra, huge, low sea levels sounds best, but you won't be able to hook up those rare health/luxury resources until Astronomy.

killercane
Mar 23, 2008, 07:15 PM
Only land tiles count for domination, right? So what about archipelago and the "free" population sea tiles give as compared to milking in civ 3? Can a massive archi map of 1 tile islands give a great score despite the tedium?

I began a new game and it progresses well. Built Oracle and took CoL and as such should reach CS at 1000 BC with about 17 cities. Another thing about lots of AI is there is more gold to fuel research, it seems there is always someone to sell Wheel or Meditation to for a few hundred gold. I skipped construction this time around due to cat costs and just have been avoiding hill cities and capitals. I also think I will avoid having vassals outside of Mansa as you have much more control over pop growth.

Sushi is strange, it seems to round up your food, so if you have 49 resources you get +13 food rather than the expected 12.

VirusMonster
Mar 23, 2008, 07:36 PM
Only land tiles count for domination, right? So what about archipelago and the "free" population sea tiles give as compared to milking in civ 3? Can a massive archi map of 1 tile islands give a great score despite the tedium?

I opened a tiny pangea map and counted the land tiles and compared to the maximum possible land score. They match, so coast, ocean, and fresh water lakes don't count toward the land score.

An archi map could give a higher score, but there are too few archi entries in the HOF with high scores. So I assume something is more difficult about archi maps, possibly the transportation of the army and the extra hammers spent on ship contruction. Maybe it is just a mental barrier.

I began a new game and it progresses well. Built Oracle and took CoL and as such should reach CS at 1000 BC with about 17 cities. Another thing about lots of AI is there is more gold to fuel research, it seems there is always someone to sell Wheel or Meditation to for a few hundred gold. I skipped construction this time around due to cat costs and just have been avoiding hill cities and capitals. I also think I will avoid having vassals outside of Mansa as you have much more control over pop growth.

I agree that the higher number AIs are better for research. More AI = more gold after Currency :) Your start looks excellent again.

Sushi is strange, it seems to round up your food, so if you have 49 resources you get +13 food rather than the expected 12.

Yea, contrary to almost all math in CivIV, Cooperations round up, instead of down :)

VirusMonster
Mar 25, 2008, 05:49 AM
omg :) I entered maximum population (3181) and land (2711) values from a Huge Archipelago map (tiny islands, low sea levels with roughly 200 :crazy: sushi resources on the map) into the earlier Matlab Civ4 scoring script I wrote. Yea, you did not read it wrong: 200 sushi resources equal to +50 :food: per turn on a huge map. If you don't believe me, go give yourself satellite tech in the world builder and count yourself :)

Next, I assumed an as successful game where same precentage of land and population as this 500k game would be achieved on that map as well. This game ended on turn 546, but I assumed a more successful game, like the one KillerCane had that ended at turn 500. 1 000 000+ score is possible :)

Then, I plugged in maximum population (2690) and land (3810) values from a Huge Terra, low sea levels map into the equation. I assumed an as successful game where same precentage of land and population as this 500k game would be achieved on that map as well. This game ended on turn 546, but I assumed a more successful game, like the one from KillerCane had that ended at turn 500. Again, 1 million scores appeared :)

I think Archipelago is better than Terra at this point for highest Civ4 scores. Terra might get an earlier finish date, but the Sushi bonus is just too large to pass. Say you control half of the earth toward the end game and achieve additional sushi resources through trading. At least +30 :food: per turn through sushi :) Population growth every 2-4 turns. That is imba. Question becomes which leader to use :) Darius, Caesar, Joao, or Williem. You need the Immortals to expand as quickly as KillerCane, so I would stick with Darius. Sailing must be researched early however for Galley transports. Workerstealing is also problematic.

WastinTime
Mar 25, 2008, 08:48 AM
I don't think arch will work. You're forgetting the lack of forest chops, few hills (you'd have to rely on slavery), no early conquest (free cities/workers), and very slow tech with water squares. You probably can't get to sushi in time. Maybe aim for 600 turn finish instead.

oyzar
Mar 25, 2008, 09:30 AM
Would non-huge maps sometimes be better? As you can get more food more easy from sushi on such maps?

WastinTime
Mar 25, 2008, 10:01 AM
Would non-huge maps sometimes be better? As you can get more food more easy from sushi on such maps?

I was thinking the same thing when I asked if there is a score penalty for less-than-huge. I don't think there is. Smaller map means less beakers/tech. It also means you can hold a larger % of land before domination and get a larger % of the land in the end.

Best of all, it means you wouldn't have to play on Huge! Not sure how small you can go before the lack of space/food catches up to you. I'll guess standard size is the way to go.

MrCynical
Mar 25, 2008, 10:48 AM
If you read MrCynical's Power of Sushie article, he discusses a +20 Sushi Cooperation from one of his games. KillerCane had 50 sushi resources in his game, 50*0.25=roughly +12 . He was using low sea levels compared to my high sea level submission. I believe the random shore line could also play a factor. (I was using a solid shoreline).

In my Huge Immortal submission, Sushi would give 21*0.25=+5 . In my large Immortal submission for HOF, I had 13*0.37=+5 from Sushi. Again, for my standard Immortal submisson, I would gain 9*0.5=+5 from Sushi. These last 3 figures are all from Pangea map type. I assume MrCynical's example is a water-rich map, possible archipelago, with definitely more sushi resources and can push for +15-20 extra range.

That particular example was from a large Big/Small map with a high sea level. I've found +20 is usually achievable without too much trouble on archipelago's of any type, fractals, hemispheres and Big/Small. Medium/Small would probably give a high number of seafood resources as well. More AIs do have a noticeable boost to the seafood quantities due to the tendency of the map generator to beautify starting locations with it.

My record for Sushi is +35 food per turn, though that was near the end of a Big/Small game with Domination deactivated, and I had most of the world's land area.

VirusMonster
Mar 25, 2008, 01:49 PM
edit: I deleted my last message, because I run some test on my Matlab CivIVscoring simulator.

For a huge archipelago map with tiny islands and low sea setting, max pop was 3181 and max land was 2711.

For a standard archipelago map with tiny islands and low sea setting, max pop was 1556 and max land was 1104.

I assumed both maps had similar raw score to max score ratio for population and land scoring. (824 pop rawscore / 458 land rawscore) for standard, and (1684 pop rawscore / 1124 land rawscore) for huge.

When I plugged those numbers into the formula, the huge map gave 1 000 000 + score, while the standard map gave 700k + score. I think the formula does take into account map sizes through the higher value of rawscores. Although the ratio of rawscore to max score remains the same, since rawscore grows linearly, whereas maxscore value in the formula, ((floor(maxscore/initialscore))^(turnselapsed/maxturns)), has an exponential growth, the final score seems to be higher for larger map sizes.

Then, I remembered that technology costs are less in standard sized maps than huge. 1.3/1.5=0.86 cheaper tech costs could mean reaching Sushi 14% faster. For the calculations above, I had assumed game would finish at turn 500. Including a faster tech speed by 14%, game would end at the 500*0.86=433rd turn.

When I plugged 433rd turn into the above formula, I received a 900k score for a standard sized map, still lower than a 1.0-1.1 million huge map score, but close enough.

In conclusion, huge maps are still the way to go to achieve highest scores. The HOF also does not allow more than 10 AI opponents for standard maps, ie less number of beautiful starting positions, less Sushi resources.

Methos
Mar 27, 2008, 11:06 AM
I hate to burst your bubble, but in our March 25th update (http://forums.civfanatics.com/showthread.php?p=6645441#post6645441) Killercane had a 704,552 score (http://hof.civfanatics.net/civ4/game_info.php?entryID=11412) game. I'm curios what the cap is.

r_rolo1
Mar 27, 2008, 11:18 AM
There should be one, but I doubt that it is inferior to 1 million....

P.S and now a mod has to change the title... :(

VirusMonster
Mar 27, 2008, 11:50 AM
I would like to rename this article to "Guide to 500k+ scores through Huge map domination on Immortal difficulty", but could not, because I got stuck with the inital naming :) Perhaps Methos can change this thread's name?

Also methos, can you create one more post on this thread after the last page of my article? Cabert's quick post prevented me from posting a summary&mistakes&how to score higher kind of last page. Character limit is 30000, and most of my pages are almost full. I don't want to rearrange pages, since the hierarchy of the article would be lost. I didn't originally plan this article to be this long, but there is still some details I would like to add :)

I want to keep my sig the way it is :) or I could add a small scripted 2nd at front :)

If there will be ever a 1 million score, it requires a fantastic starting position just like Killercane's 700k game, with 4 gold mines, 1 cow and 4 foodplains. Map type also requires continuous land for Persian Immortals to harass successfully. A lucky Terra huge low sea levels could cross 1M+, but none of my regenerated Terra starts had 3 or 3+ gold or gems in the BFS. Perhaps Pangea is the only generous map script :)

Methos
Mar 27, 2008, 12:05 PM
I would like to rename this article to "Guide to 500k+ scores through Huge map domination on Immortal difficulty",

Also methos, can you create one more post on this thread after the last page of my article?

Done and Done

r_rolo1
Mar 27, 2008, 12:06 PM
Why don't you use Mapfinder?.... Just Type " 4 gold" and go dinner ( I suppose that it will take a while :lol: )

VirusMonster
Mar 27, 2008, 12:11 PM
Done and Done

omg, done so quickly. I love you :)

I should probably change my sig as well, but I won't change the first paragraph of the article, just so I can remember I had the highest score once :)

Why don't you use Mapfinder?.... Just Type " 4 gold" and go dinner ( I suppose that it will take a while :lol: )

Yea, I should probably learn how to use Mapfinder properly. Regenerating a single Huge Terra map takes at least 15-20 sec on my PC :)

edit: changed first paragraph to include between which dates this game remained as highest score.

WastinTime
Mar 27, 2008, 02:24 PM
Why not play on Diety level. You get a 10% score boost, right?

And you can play less AI (for larger dom limit).

And you get more free cities/workers.

I still think your conclusion that Huge maps are best uses some faulty assumptions (like rawscore:maxscore ratios are the same). And misses a few other factors (like a higher dom limit on standard.)

Unfortunately, I don't have the patience for marathon, so I'm content to watch this thread and see if someone gets the million. There's also the Vanilla record high score (which was on Cheiftain?) to think about.

VirusMonster
Mar 27, 2008, 02:39 PM
Why not play on Diety level. You get a 10% score boost, right?

And you can play less AI (for larger dom limit).

And you get more free cities/workers.

I still think your conclusion that Huge maps are best uses some faulty assumptions (like rawscore:maxscore ratios are the same). And misses a few other factors (like a higher dom limit on standard.)

Unfortunately, I don't have the patience for marathon, so I'm content to watch this thread and see if someone gets the million. There's also the Vanilla record high score (which was on Cheiftain?) to think about.

You might be right. A standard map with maximum number of HOF allowed 10 AI opponents gives 56% domination limit, 5% more limit than Huge map size. Reducing the number of opponents to the minimum allowed 6 increases domination limit to %64. This extra land could give higher population and a higher rawscore/maxscore ratio in the final score. On the other hand, worker stealing and Immortal harass would be much more difficult, because the AI will have much space to expand to find a Copper or Iron resource. I guess 1M+ scorer would have to figure out which mapsize&type is better for himself :)

I think Deity might be too hard :-/ But yea, if you succeed, you will get a 2.0/1.8=11% higher end score. Worth the try. This %11 could make the difference between 900k and 1M.

killercane
Mar 27, 2008, 03:03 PM
I would like to rename this article to "Guide to 500k+ scores through Huge map domination on Immortal difficulty", but could not, because I got stuck with the inital naming :) Perhaps Methos can change this thread's name?

Also methos, can you create one more post on this thread after the last page of my article? Cabert's quick post prevented me from posting a summary&mistakes&how to score higher kind of last page. Character limit is 30000, and most of my pages are almost full. I don't want to rearrange pages, since the hierarchy of the article would be lost. I didn't originally plan this article to be this long, but there is still some details I would like to add :)

I want to keep my sig the way it is :) or I could add a small scripted 2nd at front :)

If there will be ever a 1 million score, it requires a fantastic starting position just like Killercane's 700k game, with 4 gold mines, 1 cow and 4 foodplains. Map type also requires continuous land for Persian Immortals to harass successfully. A lucky Terra huge low sea levels could cross 1M+, but none of my regenerated Terra starts had 3 or 3+ gold or gems in the BFS. Perhaps Pangea is the only generous map script :)
I think you should play some more and Im sure you will get a larger score. I just finished my second game unfortunately, hitting the domination limit too early at 589k and just having got biology (turn 440ish). So back to the drawing board.

The second game was mapfindered and got another 4 gold start. The first game was randomly generated :eek:. Gems would certainly be better. I think I will try some Deity maps on terra and see what mapfinder comes up with. From a patience standpoint, these games are much easier than doing a huge civ 3 game.

I read where more want a writeup. I will try to do that this weekend. Really the techniques are simple and nothing is revolutionary. An outline:

1) find a nice commercial start with at least 2 commerce resources
2) build 2-3 warriors, 2 asap working high shield tiles. Go wander about until you can find some workers unprotected on the outside of the BFC and steal them without an archer killing you.
3) crank out a settler for horses at about size 3, especially if you can work 2 golds and a floodplain/foodbonus tile with no starvation.
4) initial research is mining->animal husbandry->wheel->bronze working->oracle path
5) send immortals out to all the neighbors. If you have one close with only 2 archers in, send 3 to take the city. Once you have amassed enough immortals, begin taking the non capital cities and farming the AI for workers and towns. Guard/pillage any and all bronze/iron/horses. If you need to make peace, do so, and return to them shortly. There are many AI to harass. Avoid hill cities as much as possible.
6) Research writing, build a library, and kill your gold hills in favor of 2 scientists. Continue building immortals.
7) Build Oracle pre 1500 BC for CoL. Metal casting you dont need, and you have to build it quick to assure you obtain it so CS slingshot is a risky endeavor.
8) Currency is the most important tech, it allows you to run high research at deficit. I must go through 50K gold doing research in the early-midgame. Add catapults (construction) and proceed to nationalism/education and then liberalism for mil tradition slingshot to build currassiers. Always sell tech for gold to continue your research at a high rate. Keep taking cities for gold. You should never go broke really with Darius.
9) You should reach mil tradition about 1 AD. Upgrade Immos/horse archers and roll through everyone as best you can until cavalry arrive.
10) Obtain cavalry, and then proceed to biology and watch your domination limit.
11) Milk it for all its worth. Give away some low food cities in favor of higher food locations.

WastinTime
Mar 27, 2008, 03:20 PM
so no corporations?

killercane
Mar 27, 2008, 04:03 PM
so no corporations?
Thats a whole 'nother chapter. Sids Sushi is the One, put it in your holy city (you hopefully got a priest to build the shrine for whatever religion most of the AI are) along with Wall Street since you have such high costs using these corporations. Try to obtain as many coastal cities as possible and the resources of course. One tile islands are very nice.

Executives cost 200 hammers at marathon, and I have tried just purchasing them for 500ish gold a pop. You can turn off research at medicine if you have Physics already and the great scientist. Pop off a golden age here if you can to go into the -25% corporate cost civic; maybe you can take environmentalism work later after most everything is bought.

National Park you can probably put in your highest food city along with Globe so you dont have to raise the lux slider to compensate for a single city's high growth rate.

Another thing is to obtain Mausolleum of Mausolos as early as possible. Its worth 2 golden ages at least. I would hate to build it, rather take it, but perhaps it is indeed worth building. I havent had it in time in either of the two games so far.

GP Needs Rundown
Scientist-academy
Priest- shrine for uber religious city (worth +70 gold when put in last go around)
Scientist- education
Merchant- Sushi (free at Econ)
Scientist- Golden age (free at Sci Method)
Artist- Golden age (free at Music)
Spy- Golden Age (free at Communism)
+3 more of various types for your 4th golden age

Total of 96 turns in golden age (about 20% of the game).

Things for further study:
Research post medicine/refrigeration- go on to Future Tech if score is still increasing? What is the value of future tech?
Christo Redentor- worth it or no? No more need for golden ages to civic change and allows for easier environmentalism.

VirusMonster
Mar 27, 2008, 04:43 PM
Things for further study:
Research post medicine/refrigeration- go on to Future Tech if score is still increasing? What is the value of future tech?
Christo Redentor- worth it or no? No more need for golden ages to civic change and allows for easier environmentalism.

Future techs (+1 raw score) have almost no effect on final score, since finishing early matters much more. You could probably change to environmentalism on the last turn of your 4th Golden Age and avoid needing Christo Redentor.

Holy cities on a huge map are great. I wished I captured more in my 500k game.

VirusMonster
Mar 28, 2008, 09:26 PM
4) initial research is mining->animal husbandry->wheel->bronze working->oracle path

5) send immortals out to all the neighbors. If you have one close with only 2 archers in, send 3 to take the city. Once you have amassed enough immortals, begin taking the non capital cities and farming the AI for workers and towns. Guard/pillage any and all bronze/iron/horses. If you need to make peace, do so, and return to them shortly. There are many AI to harass. Avoid hill cities as much as possible.

7) Build Oracle pre 1500 BC for CoL. Metal casting you dont need, and you have to build it quick to assure you obtain it so CS slingshot is a risky endeavor.

This strategy is a bit luck dependent, because:

1) Initial worker stealing could be problematic.
2) Horse are rare. Not every 2 gold or 2 gem start will yield horses nearby. I tried a Huge Terra game where I researched Fishing&Sailing&Writing, just so I could connect the distant Horses using Open Borders, rivers, and coastal traderoutes.

172757

3) Late horses mean late harass. Late harass means, AI has better chance to produce a few spearman :-/ A few spearman would definitely beat your harassing Immortals.
3) If you delay researching IronWorking, how would you know which mine to pillage? Last game, I thought I had a great start, but spears appeared because I delayed researching IronWorking. Priesthood tech path is risky.
4) Oracle got build at 1770BC in my recent Terra Huge game. Can't really beat that date without Marble around :-/ Or I should have chopped 5 forests at 60 hammers each and poprushed the rest, but did not manage to steal enough workers initially.


In conclusion, I believe producing the inital worker yourself with Darius is best idea. Next, 2-3 warriors can be sent to worker stealing. Research path should be (Animal Husbandry (to work pigs or cows in BFC)-> Mining (to speed up research) -> Wheel (Immortals can be produced now) -> BronzeWorking -> IronWorking. I think Oracle is a big gambit, and the Oracle tech path might delay other areas of research. I would rather tech quickly to Currency to trade away AI gold.

killercane
Mar 29, 2008, 09:15 AM
This strategy is a bit luck dependent, because:

1) Initial worker stealing could be problematic.
2) Horse are rare. Not every 2 gold or 2 gem start will yield horses nearby. I tried a Huge Terra game where I researched Fishing&Sailing&Writing, just so I could connect the distant Horses using Open Borders, rivers, and coastal traderoutes.
3) Late horses mean late harass. Late harass means, AI has better chance to produce a few spearman :-/ A few spearman would definitely beat your harassing Immortals.
3) If you delay researching IronWorking, how would you know which mine to pillage? Last game, I thought I had a great start, but spears appeared because I delayed researching IronWorking. Priesthood tech path is risky.
4) Oracle got build at 1770BC in my recent Terra Huge game. Can't really beat that date without Marble around :-/ Or I should have chopped 5 forests at 60 hammers each and poprushed the rest, but did not manage to steal enough workers initially.


In conclusion, I believe producing the inital worker yourself with Darius is best idea. Next, 2-3 warriors can be sent to worker stealing. Research path should be (Animal Husbandry (to work pigs or cows in BFC)-> Mining (to speed up research) -> Wheel (Immortals can be produced now) -> BronzeWorking -> IronWorking. I think Oracle is a big gambit, and the Oracle tech path might delay other areas of research. I would rather tech quickly to Currency to trade away AI gold.
Well, I take this as a guide on how to achieve an optimal score in BTS, not just win on Huge Immortal. So no, making a worker first yourself is not the best idea, as producing two warriors should give you size 2 with 2 workers as opposed to size 1 still when you research mining or animal husbandry. Free is the best game in town!

1) Maybe, but I find it is 85% successful. It takes practice actually though, funny as that sounds.
2) Maybe again, but out of 5 maps started on these settings I had horses within 10-12 tiles every time.
3) Well maybe. Kill their metal first!
4) Pillage all the mines. If they have a bunch of mines AND have hooked up iron AFTER researching all that way you are either too slow or the AI is too far away to mess with. If a mine appears on plains or grassland, obviously thats what you want to pillage.
5) Sure enough. This is why I didnt build it first time around, its 50/50 with the new prioritization on it by the AI. But optimally you want to build it. The superoptimal start is CS slingshot but research is too slow unless you really want to risk it going late. Maybe with only 10 opponents you can achieve this if you are able to reach them all in time to keep them from working hammer tiles. Hard to do.

So future tech is out unless you have nothing else to do. I was thinking the +health and happy might be useful. Good discussion!

VirusMonster
Mar 30, 2008, 08:41 PM
I used mapfinder to regenerate 120 Terra Huge starts. Only 1 of them had 2 gold resources :)

Then, I had a fantastic start. Mining and BronzeWorking popped from the hut. Had Horse nearby. Built Oracle on time and choose Code of Laws. Harassed very successfully.

How did I ruin it? Well, make you sure are in good diplomatic relations against a Bronze or Iron owner. Adjust your religion to be friends with him. Also, as KillerCane suggested send Immortals to all AIs nearby. I only harassed 3 of them. I should have harassed like at least 6 :-/ The AI won't attack your Immortals with archers except when he has like 5-6 archers inside the city.

VirusMonster
Mar 31, 2008, 07:20 PM
argghghhhhh :mad: Another fantastic start ruined because I pushed my chances too much :-) I got a very cool 2 gold start. Successfully stole 3 workers before 3250BC. This time I sent 1-2 Immortals to 6 AIs, and on top, the WarChariot Quest showed up. Reward is CombatI to all Immortals, so I didn't want to miss it. Pillaged all metal and mines. The AIs could not do anything. Some of the very distant AIs declared war, but they were so far away that it did not matter.

Now, here is how I ruined this game. I stayed with my initial 2 cities until 1770BC and tried manually researching Mathematics, so I could get Currency with Oracle and also chop it with +90 :hammers: each instead of +60 :hammers: But I also did not want to reach Mathematics quickly, because Math would obsolete the WarChariot quest. I needed to produce 12 WarChariots and just the turn after I completed the WarChariot Quest, I discovered Mathematics.

Wonderful you think? :) Now I should be able to get Currency with the Oracle. I had already pre-chopped forests near my capital, so I should be able to build the Oracle in 1 turn.

But what happened :) The one AI nearby I left untouched, Isabella, because I thought she would build Pyramids with the Stone nearby, completes the Oracle :( Puff...

Here is the save: 172923

I also realize now that probably the earlier Oracle would have made more :gp: and the gap in :science: between Code of Law and Currency is minimal. I should have build the Oracle before 2000BC, chosen CoL, then none of this would have happened. Or harassed Isabella too, but then I would miss the chance to control the Pyramids :-/

I am gonna play safer if I try again.

killercane
Mar 31, 2008, 08:46 PM
I looked at your save. You have the immortals fortified everywhere; maybe this is just so you can look over your deployment, but I find that moving them around opens up a lot of possibilities, like there in Peters land after moving I see a worker working that can be taken next turn. BTS AI doesnt compute 2 move units very well. Also your workers are fortified, surely they can be doing something low priority nearby and not resting; sometimes you can begin a cottage or mine on the forest while you wait.

Also, I would look to expand more than 2 cities to get a 3rd immortal producer. Just find a pop 2, 2 archer, non hill AI city that you arent at war with and send 3 immos there to take it; or bait the AI to move his archer stack to a peripheral city and take the capital. These opportunities should be plentiful in the early going. Your research looks good, and Oracle would have been nice on such a nice map! :goodjob:

I think safe is the way to go for sure. No reason to risk nice maps with crazy slingshots.

Methos
Mar 31, 2008, 08:50 PM
@killercane: Any chance of a write up on your 700k game?

ParadigmShifter
Mar 31, 2008, 08:56 PM
Izzy seems to love building the Oracle in my games.

VirusMonster
Apr 01, 2008, 10:12 AM
@Paradigmshifter: As you said, the problem with Izzy is she acquires way too many religions :religion: It is very usual in my games that she gets Buddhism, Judaism, and Confucianism (through Oracle). Ideally, you want the Egyptians get the early religions so they can get an early prophet through their Obelisk and build the religious shrines. Then, hopefully, you can capture their holy city :)

Izzy does not seem to get any GreatProphets until much later in the game.

@killercane: yea a 3rd city is a must. Or a better located horse city that can actually grow to size 4-5 while producing those early Immortals. In the save I posted, there was no good food resource around that horse city.

My Immortals were patrolling the mines, since I was being paranoid about AI discovering Iron somewhere :) If you saw them fortified, it is most likely because they did an oppurtunistic attack on one of AI archers and were healing their wounds.

VirusMonster
Apr 01, 2008, 07:30 PM
Another attempt ruined, because I produced my inital Immortals too late. I managed to produce a 2nd settler due to a 3gem start and had 3 cities, but then messed up by attempting to hook up Marble. Workers were busy hooking up the Marble instead of the Horse, and the Immortals arrived too late.

I believe you can build 2 settlers successfully and still complete the Oracle in time near 2000BC, but you need to get those Immortals out as soon as possible.

killercane
Apr 02, 2008, 07:16 PM
@killercane: Any chance of a write up on your 700k game?
Working on another, when it gets done I will give a good writeup.

VirusMonster
Apr 08, 2008, 09:23 PM
edit: I've just looked at your saves again. You had 9 cities at 1000BC, not 17 :-) Wow, I had a great start then with 10 cities at 1200BC :)
-------------------------

I have tried sufficient games to say that you can't always pillage all nearby Bronze and Horses successfully. Even if you have a great start, warmongering AIs have a high chance of hooking up horse or bronze before your Immortals appear on their borders. There is a high chance you are surrounded by AIs who already possess Chariots or Spearman.

The problem is that your inital city gets build too late. In your 700k game, you build your 2nd city at 2400BC, and captured your first 2 cities near 1900BC. Well in my Darius games, it is not uncommon for AIs already have spears by 2150-2100BC :-/

Compared to my game, where I already had 4-5 cities at 2400 BC, your initial war production comes in too late. I'm thinking maybe researching BronzeWorking after Mining might be a better idea to chop 2 settlers before the inital rush. You'll also get your first city earlier. Only the capital should build barracks, second and third cities should poprush Immortals at size 2.

---

my recent attempt: 3175BC, and the nearby Caesar hooks up Horse :-/

--

yet another attempt: 3475BC: Lincoln builds pasture on horse.

killercane
Apr 09, 2008, 05:09 AM
edit: I've just looked at your saves again. You had 9 cities at 1000BC, not 17 :-) Wow, I had a great start then with 10 cities at 1200BC :)
-------------------------

I have tried sufficient games to say that you can't always pillage all nearby Bronze and Horses successfully. Even if you have a great start, warmongering AIs have a high chance of hooking up horse or bronze before your Immortals appear on their borders. There is a high chance you are surrounded by AIs who already possess Chariots or Spearman.

The problem is that your inital city gets build too late. In your 700k game, you build your 2nd city at 2400BC, and captured your first 2 cities near 1900BC. Compared to my game, where I already had 4-5 cities at 2400 BC, your initial war production comes in too late. I'm thinking maybe researching BronzeWorking after Mining might be a better idea to chop 2 settlers before the inital rush. You'll also get your first city earlier.
Hard to say. I just go with the flow. With 10 opponents you dont need to pillage much, if any. Just go round up the neighbors.

P.S. I played a Terra map with not really ideal conditions (settle on PH, grassland cow, 1 gold capital) , but it went well and Im working on the writeup finally.

VirusMonster
Apr 09, 2008, 05:27 AM
omg :) 1 million, bravo :)

classical_hero
Apr 09, 2008, 09:20 AM
Congratulations Killer. :clap:

tycoonist
Apr 09, 2008, 03:51 PM
world record not just beaten but smashed. they should put you in the Guinness Book of World Records

r_rolo1
Apr 09, 2008, 04:05 PM
I wonder how high this can get..... probably 1,3 , 1,4 million ?

In fact this only shows that the higher scoring in game = 15k + is simply innadequate ;)

VirusMonster
Apr 11, 2008, 01:28 AM
I've looked up his saves :) The max possible population is something like 2800, and at victory date he had 2700 score on population :) He fought his way to control more than half of the main continent and settled almost all of the New World. He could have finished the game much earlier, but milked like mad. Actually, the turn he won, his score was still going up like 2-3k per turn.

Sushi was giving +14 :food: Not as ideal as a water-rich +25-35 :food: map, but still good.

His starting position and early game is far from ideal, but he stick to it. Early Academy near 1500BC. He had both Stone and Marble early on which helped him build some nice wonders. He did not build the Oracle in the 2000-1500BC period, but there was no Isabella this game, so he still managed to finish it between 1300-1000BC.

I've looked at this 1390BC save. He controls 12 cities with a total of 30 population. Research has almost halted, but the Oracle and cottages will speed things up soon. And the problem of AI massing archers to guard their capital is also hurting him when capturing London.

Really nice game actually, but I think you need a very good PC to survive late game :) He had like 140 cities in the late game, and I had major delays opening the city advisor screen :)

Killercane :goodjob: again for being the first person to cross 1M score!

killercane
Apr 11, 2008, 07:30 AM
edit: I've just looked at your saves again. You had 9 cities at 1000BC, not 17 :-) Wow, I had a great start then with 10 cities at 1200BC :)

I think the 17 cities was my 589k game that I didnt submit. I still have the saves if you want those.

world record not just beaten but smashed. they should put you in the Guinness Book of World Records

This made me laugh. Thanks Tycoonist and Classical!

killercane
Apr 11, 2008, 07:37 AM
I've looked up his saves :) The max possible population is something like 2800, and at victory date he had 2700 score on population :) He fought his way to control more than half of the main continent and settled almost all of the New World. He could have finished the game much earlier, but milked like mad. Actually, the turn he won, his score was still going up like 2-3k per turn.

Sushi was giving +14 :food: Not as ideal as a water-rich +25-35 :food: map, but still good.

His starting position and early game is far from ideal, but he stick to it. Early Academy near 1500BC. He had both Stone and Marble early on which helped him build some nice wonders. He did not build the Oracle in the 2000-1500BC period, but there was no Isabella this game, so he still managed to finish it between 1300-1000BC.

I've looked at this 1390BC save. He controls 12 cities with a total of 30 population. Research has almost halted, but the Oracle and cottages will speed things up soon. And the problem of AI massing archers to guard their capital is also hurting him when capturing London.

Really nice game actually, but I think you need a very good PC to survive late game :) He had like 140 cities in the late game, and I had major delays opening the city advisor screen :)

Killercane :goodjob: again for being the first person to cross 1M score!
Im still working on the writeup. RE: early game-my research sucked balls and Mansa had priesthood. Luckily he didnt build it in time and I could take Currency to solve all my research problems. This also shows how a few extra commerce tiles would really help; the start really wasnt good. About the only good thing on this map was that India was a pushover early on and built Stonehenge. Delhi then became my wonder city which produced GP for golden ages.

The slow pc and making too many mistakes was pretty much why I stopped when I did.

VirusMonster
Apr 11, 2008, 07:43 AM
After seeing your sucky early game, I thought that even a 1.5M+ game is possible. You recovered really nice. If you need help with html formatting on your writeup, let me know. Otakijubski taught me nice table of contents tricks ;-)

killercane
Apr 11, 2008, 07:48 AM
Well I'll probably do well even getting enough screenshots in :). Can you plug 3200/2800 population and winning about 50 turns earlier in Matlab (assume other numbers are the same as that game) and see what you get? Im thinking of trying to eliminate most AIs and stay at war with the last few to prevent them from building Hanging Gardens until the last turn. Its hard to do so far but possible.

VirusMonster
Apr 11, 2008, 07:52 AM
What is your PC configuration Killercane? Mine is 1.8Ghz Athlon with 1GB RAM.

VirusMonster
Apr 11, 2008, 08:18 AM
Well I'll probably do well even getting enough screenshots in :). Can you plug 3200/2800 population and winning about 50 turns earlier in Matlab (assume other numbers are the same as that game) and see what you get? Im thinking of trying to eliminate most AIs and stay at war with the last few to prevent them from building Hanging Gardens until the last turn. Its hard to do so far but possible.

Alright, I plugged in the numbers :) First, I plugged in your game data to the formula just to see whether it is displaying a correct score. Yep, you had 1,085k, formula shows roughly 1,100k; so, I assume it works correctly neglecting rounding errors.

Next, I plugged in 3200 for population. Remember in your 1M+ game maximum population accoring to the scoring system was 2800.

Then, I made a mistake :) You have asked me to give you an approximate score for this game if it were finished 50 turns earlier. Well, you finished on turn 603. So, if I were to do what you said, I should have plugged in 553. But I was careless and plugged 503 :)

Then, I saw the 2M+ score and realized something must be wrong :) So fixed it back to turn 553 as victory date, and 1,64M appeared.

Looking at your not ideal start, I am convinced that 1,5M+ scores are possible with a good PC :)

WastinTime
Apr 11, 2008, 09:41 AM
There is the A_Turkish_Guy's submission, but it is on beta and on Warlords difficulty with cheesy tribal hut abuse =)

What is the hut abuse you speak of? Is it just the 5% chance to get a free settler or worker?

VirusMonster
Apr 11, 2008, 10:23 AM
That is a pretty old post of me. You don't get free settlers on Immortal difficulty. Nor do you get free workers from huts. He was using Asoka for that game, because workers can travel 3 squares. Then, he was hitting like 30-40 huts early game. He delays settling first city and searches for huts with the inital settler as well. I am not sure about the probability of worker&settler appearance. But almost always he would get 3-4 Settlers, like %10 of the time maybe. And many workers of course.

WastinTime
Apr 11, 2008, 10:53 AM
At Warlord level, it's only 5% settler, 5% worker, so that's not really very abusive. I was hoping for something more powerful to make it possible for lower difficulties to score 1 million. They still might make 1 mill, but can't get the ultimate high score (probably.)

P.S. Cheiftain/Settler level gets 10% settler, 10% worker.

VirusMonster
Apr 11, 2008, 11:01 AM
Well, I know for sure that he tries many starts until he gets sufficient settlers and workers. Even if it is as low as 5%, since to replay any start for 30 turns does not take much time, he'll have a good start within several attempts. He also does not have to waste time on map regeneration, anything goes. Then he settles 4 cities after 30 turns. 4 cities in 30 turns is slightly abusive to me :) Remember this is marathon speed, so 30 turns during very early game under normal conditions should only be enough to produce a single worker. It is like you are playing two times stronger version of the AI deity bonus of 2 settlers.

WastinTime
Apr 11, 2008, 11:19 AM
Then he settles 4 cities after 30 turns. 4 cities in 30 turns is slightly abusive to me :) Remember this is marathon speed, so 30 turns during very early game under normal conditions should only be enough to produce a single worker.

That's not even as abusive as the old Inca rush on Diety. In 25 turns (or less) you can wipe out 2 other AI civs, and of those 4-5 cities 3 of them are prime capitol city locations. And you get 4 workers too! Of course that rush only works that well on vanilla civ.

VirusMonster
Apr 11, 2008, 12:24 PM
well, I wrote the Quechua rush article on Deity if you remember, and no, in 25 turns (on marathon speed) you can only produce 4-5 Quechuas under best odds. Add 10 turns until your quechuas travel to first city location. Capturing the 2nd city, ie possibly capital, would again require waiting another 25+10 turns until the second set of 5 quechuas appear on enemy borders.

The extra good thing about Warlords hut rush is that he would collect 600 gold from the huts allowing him to run 100% science for a long time.

I respect his tactic tough, don't think because I called it abusive, it is in any way less smart than other strategy out there. It is just that when majority of the players in the forum play regular games with their inital city building the first settler, scoring high due to hut settlers is not as appreciated.

Shoot the Moon
Apr 11, 2008, 12:51 PM
Don't forget that with the hut popping exploit you also get a lot of free techs from huts. I'm not saying it is enough to overcome the score penalty, but it is still incredibly overpowering.

VirusMonster
Apr 11, 2008, 01:21 PM
True, that is just as strong as the extra 600 gold. You'll need to have the first city settled for huts to give techs. He was also searching for huts with the initial settler, but yea after he settles the first city, he can also pophut some early game techs. If lucky, he could even discover expensive ones.

killercane
Apr 11, 2008, 01:57 PM
I have been getting animal husbandry and the wheel a lot from huts on Immortal. And of course Astronomy is almost always free on Terra maps. My thinking now is to conquer everyone prior to their ability to build HG and this would allow you to get free music from a hut on the new continent as well and not worry about losing it to an AI by having to research it.

You will never get a million points on lower levels due to the score multiplier. You will research faster and have more $$ to buy settlers and improvements but it really still boils down to the population, early win bonus, and the multiplier.

VirusMonster
Apr 11, 2008, 02:07 PM
How Astronomy free on Terra maps? :) You use huts from the New World?

killercane
Apr 11, 2008, 04:09 PM
How Astronomy free on Terra maps? :) You use huts from the New World?
Send over caravels with explorers/scouts once you get Optics or whatever the caravel tech is. There are so many huts there that you are bound to luck up and get Astro.

WastinTime
Apr 11, 2008, 04:50 PM
well, I wrote the Quechua rush article on Deity if you remember, and no, in 25 turns (on marathon speed) you can only produce 4-5 Quechuas under best odds. Add 10 turns until your quechuas travel to first city location. Capturing the 2nd city, ie possibly capital, would again require waiting another 25+10 turns until the second set of 5 quechuas appear on enemy borders.

I didn't know you wrote a quechua rush article. 4-5 Quechuas...sounds like your strategy was to help people do a lengthy, safe rush with 99% chance of success. Fine advice for that case, but I'm talking about an aggressive risky rush (10% success.) I hate quechua and only play Inca or Marathon if I'm forced to for a certain victory. So I spent very little time/effort, but still easily captured the #1 Tiny Diety Conquest spot at one time. I'm not bragging, I'm just saying it doesn't take that many tries or that much experience with "the rush" to make it happen. That game had 4 cities in 22 turns. Of course Moonsinger wouldn't allow that record to stand, so she did it in 19 turns. I was being generous saying 25 turns!

VirusMonster
Apr 11, 2008, 05:44 PM
KillerCane what is the probability of getting a different tech than Astro? How can you be so sure the tech you get will be Astro?

WastinTime, I've just looked at Moonsigner's Tiny Conquest game log. As you said, it is different from my article in that it takes the ultimate risky path :-) My article still has a risky quechua rush section, but not as risky as the one you explained. Moonsinger wiped the first civilization out before they settled their 2nd city. The capital only had a single archer defending. I don't know how that is possible on deity.

Secondly, he kills two more archers in the open field with the same quechua that wiped out the first civilization. And his capital I assume must have blocked the 2nd civilization's path to settle their 2nd city as well. Then, he lures enemy archers out two times with worker sacrifice. Finally, he captured the remaining capital with a freshly build Quechua.

Well, that is some risky strategy for sure :) I wasn't away of it, thanks for pointing it out, but it will only work on tiny maps unfortunately :-/ so not much use for the highest score discussion.

r_rolo1
Apr 11, 2008, 05:53 PM
Virus, you can only get techs from the Classical and Ancient age from huts, except Music and Astro... if you have Drama and all the classical and below techs, if you pop a tech, it has to be Astro :D

tycoonist
Apr 11, 2008, 05:56 PM
KillerCane what is the probability of getting a different tech than Astro? How can you be so sure the tech you get will be Astro?

he can't be sure but on huge there are tons of huts out there. the chance of popping a tech is 10% i think (???) and the astro is the last tech huts will give. so pop enough huts and you are bound to be given a tech. and that tech will normally be astro (assuming you've discovered the others)

r_rolo1
Apr 11, 2008, 06:10 PM
Not exactly 10 %, tycoonist.....

Here is the <goodies> part of Immortal section of the handicaps file (in BtS of course ;) )
<Goodies>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_MAP</GoodyType>
<GoodyType>GOODY_MAP</GoodyType>
<GoodyType>GOODY_WARRIOR</GoodyType>
<GoodyType>GOODY_SCOUT</GoodyType>
<GoodyType>GOODY_EXPERIENCE</GoodyType>
<GoodyType>GOODY_HEALING</GoodyType>
<GoodyType>GOODY_TECH</GoodyType>
<GoodyType>GOODY_TECH</GoodyType>
<GoodyType>GOODY_BARBARIANS_WEAK</GoodyType>
<GoodyType>GOODY_BARBARIANS_WEAK</GoodyType>
<GoodyType>GOODY_BARBARIANS_WEAK</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
</Goodies>
As you can see, if you use a non scout, the probability of popping a tech is 2/20 = 10% like you said. But for a scout/explorer ( not sure about spies ,but I assume the same... if I'm wrong correct me ) ,that is killercane's scenario ( you can't ferry normal units in Persian caravels ... ;) ), it becomes 2/13 ( scouts can't pop barbs from huts ... and I'm overlooking the rare instances that a scout receives nothing from a hut because it draws barbs ten times in a row ( awful coding IMHO... ) Edit: I made the calculations... in Immortal the probability of getting nothing from a hut popped by a scout is (7/20)^10 = 2,75*10^-5 ..... far easier to get a pike beating a gunship :lol: ), a little more than 15%

In most cases in Immortal you'll get Astro from a hut if you can pop 10 huts after having Music... in a huge Terra map it is not that hard.

VirusMonster
Apr 11, 2008, 06:24 PM
Where can I learn which techs belong to Ancient and Classical age? Astronomy is not a classical age technology, is it?

WastinTime
Apr 11, 2008, 09:01 PM
it will only work on tiny maps unfortunately :-/ so not much use for the highest score discussion.

Why do you say it only works on Tiny? The same thing can be done on huge. You wipe out 2 AI and have 4 cities. Only 8 AI left.

I did the whole thing with the original free Quechua, I never built any. The odds aren't that bad. You use the worker lure as you said.

The reason this discussion is not going to be fruitful is because the worker lure trick doesn't work in BTS. That's the kind of start you would want though. If you can emulate that kind of start on BTS Diety, you're on your way to 2 million. Maybe 2 or 3 quechua can do it instead of the 1. Even if it's only a 5% success rate, it's easy to start a couple dozen games and get one that works.

r_rolo1
Apr 12, 2008, 02:42 AM
Where can I learn which techs belong to Ancient and Classical age? Astronomy is not a classical age technology, is it?
Best way is to go to worldbuilder and check which techs are added when you choose to add techs per era. And astro is not by far a classical tech ;)

VirusMonster
Apr 12, 2008, 03:22 AM
and then how do you explain Astro and Drama are still in the list of pop technologies? Could there be more hut techs that you are not aware? Is there any article on this?

VirusMonster
Apr 12, 2008, 03:39 AM
I have been getting animal husbandry and the wheel a lot from huts on Immortal. And of course Astronomy is almost always free on Terra maps. My thinking now is to conquer everyone prior to their ability to build HG and this would allow you to get free music from a hut on the new continent as well and not worry about losing it to an AI by having to research it.

Isn't Music a Medieval era technology? How can a hut give Music?

Drama is still classical age btw, Astronomy seems to be the only non-ancient and non-classical age technology huts give.

r_rolo1
Apr 12, 2008, 03:48 AM
Sorry , I meant Music in my posts instead of Drama :hammer2: Killercane is right ( by some unknown reason I always confuse Drama with Music ;) ).

Probably there is a article on that, but I'm not aware of that... AFAIK only Music and Astro ,from the post classic era techs, can be popped by a hut.

VirusMonster
Apr 12, 2008, 04:09 AM
Well, that is so nice to know :) They are pretty expensive, and you might not have to research for them, but you need early Optics if you want to be the first to Music.

killercane
Apr 12, 2008, 07:44 AM
Terra was chosen. It has a nice domination limit, and it also means you have a nice new world to colonize, picking out the best food spots well before the AI can grab them up. It also provides a safety net in case you destroy too many AIs; you normally HAVE to get to the new world to reach domination so you could theoretically eliminate most of the AIs in the old world without worrying about domination. Finally, using Terra essentially gives you a free tech, Astronomy; there are so many huts on the new continent on huge size, that ferrying over a scout/explorer with caravels gives a very good chance to pop it out, saving several turns of research. The following are the techs that can be popped from huts, perhaps you can get more than astronomy with a fine tuning of tactics:

=====Ancient=====15=
Fishing
Hunting
Mining
Mysticism
Agriculture
Archery
Priesthood
The Wheel
Masonry
Pottery
Animal Husbandry
Sailing
Bronze Working
Writing
Horseback Riding
=====Classical======10=
Literature
Iron Working
Mathematics
Drama
Monarchy
Calendar
Construction
Compass
Currency
Metal Casting
=====Medieval======1=
Music
=====Renaissance=====1=
Astronomy

Speed- Marathon

Low Sea Level

10 AIs- this gives you a 56% domination limit as opposed to 51% with 17 AIs.

Barbs- None

Goody Huts- On. On deity you may consider turning huts off as they help the AI too much, but on Immortal you benefit even if you never pop a single hut. The AI researches comparatively slow, and you do too in the beginning. If a couple of AI pop key techs early, you get a research bonus for them knowing the tech. Usually you can get a couple hundred gold as well, allowing for maximum research in the early going.

Rivals- Mansa, the Germanies, the Indians, Mehmed, Washington, and Elizabeth

(from whatever it is Im writing, article or guide or whatever you want to call it)

ck07
Apr 12, 2008, 11:23 AM
Not exactly 10 %, tycoonist.....

Here is the <goodies> part of Immortal section of the handicaps file (in BtS of course ;) )
<Goodies>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_LOW_GOLD</GoodyType>
<GoodyType>GOODY_MAP</GoodyType>
<GoodyType>GOODY_MAP</GoodyType>
<GoodyType>GOODY_WARRIOR</GoodyType>
<GoodyType>GOODY_SCOUT</GoodyType>
<GoodyType>GOODY_EXPERIENCE</GoodyType>
<GoodyType>GOODY_HEALING</GoodyType>
<GoodyType>GOODY_TECH</GoodyType>
<GoodyType>GOODY_TECH</GoodyType>
<GoodyType>GOODY_BARBARIANS_WEAK</GoodyType>
<GoodyType>GOODY_BARBARIANS_WEAK</GoodyType>
<GoodyType>GOODY_BARBARIANS_WEAK</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
<GoodyType>GOODY_BARBARIANS_STRONG</GoodyType>
</Goodies>

.....


I have tried to edit this to eliminate the "map" result, which I hate. Used notepad. But after several attempts I was still getting results where I would pop a hut and get no result plus some assorted odd behavior. My conclusion is that there must be a hidden carriage return that I'm failing to delete.

Has anyone succeeded in editing this?

r_rolo1
Apr 12, 2008, 11:27 AM
The best place to ask that is in the Creation and costumization, but I'll toss my dime.....

All of the entries of the <goodies> have 20 entries... have you deleted the map entry or you changed maps for something else?

ck07
Apr 12, 2008, 11:30 AM
Terra was chosen. It has a nice domination limit, and it also means you have a nice new world to colonize, picking out the best food spots well before the AI can grab them up. It also provides a safety net in case you destroy too many AIs; you normally HAVE to get to the new world to reach domination so you could theoretically eliminate most of the AIs in the old world without worrying about domination. Finally, using Terra essentially gives you a free tech, Astronomy; there are so many huts on the new continent on huge size, that ferrying over a scout/explorer with caravels gives a very good chance to pop it out, saving several turns of research.

But if you do more than pop huts, what about the maintenance penalty for too many cities on another continent? Tried this in a continents game and found that the penalty was as huge as described. Releasing them as a vassal: 1) loses that land toward your domination %, which is annoying since capitulations apparently do count; and 2) causes any current vassals to leave you.

Moving the capital is no solution: even if you wind up with more land in the Americas, the difference in development is bound to be huge.

?

r_rolo1
Apr 12, 2008, 11:48 AM
Forbidden palace or versailles, maybe? SP would be a option, if they didn't blocked corps :(

ck07
Apr 12, 2008, 12:07 PM
Forbidden palace or versailles, maybe? SP would be a option, if they didn't blocked corps :(

FP and Versailles eliminate the normal distance penalty, but do they affect the "wrong continent" penalty?

SP?

r_rolo1
Apr 12, 2008, 12:42 PM
State property... sorry for the jargon ;)

killercane
Apr 12, 2008, 05:18 PM
But if you do more than pop huts, what about the maintenance penalty for too many cities on another continent? Tried this in a continents game and found that the penalty was as huge as described. Releasing them as a vassal: 1) loses that land toward your domination %, which is annoying since capitulations apparently do count; and 2) causes any current vassals to leave you.

Moving the capital is no solution: even if you wind up with more land in the Americas, the difference in development is bound to be huge.

?
The only solution I know is to switch back and forth between State Property (to gather $$$) and another civic (free market or environmentalism at the every end); I assume you mean during the milk portion here or are you talking in a normal game?

The maintenance is one of the biggest problems; hence Darius rather than someone like Inca.

VirusMonster
Apr 12, 2008, 05:20 PM
It would be best to conquer the inital continent as much as possible to avoid needing colonization. My feeling is that colonies are same as vassals and count only 1/2 toward land and population score.

VirusMonster
Apr 13, 2008, 07:19 AM
The only solution I know is to switch back and forth between State Property (to gather $$$) and another civic (free market or environmentalism at the every end); I assume you mean during the milk portion here or are you talking in a normal game?

The maintenance is one of the biggest problems; hence Darius rather than someone like Inca.

Hmm, but late game you seem to farm everyting. I don't see much cottages in your final save. I guess that did not happen in a single turn, so there was a long period of time where you did not take advantage of the Financial extra commerce bonus. How is Darius helpful in that case to compensate expenses?

The main advatage of Darius is the fantastic early harass with Immortals. They are pretty good at city capturing as well. And financial does help to reach the Sushi earlier and stock some gold for late game maintenance expenses, but late game financial starts getting useless if you want to take advantage of the farms.

On your final save, the total maintenance you pay per turn is around 10k :gold: Say a town is generating +8 :commerce: under FreeSpeech. Then, say you cottage the 20 squares around a city, and add in the gold multiplier buildings at 100% rate. 320 :gold: from a single city. To compansate your expenses you will need roughly 10 000 / 320 = 30-35 fully cottaged cities. Rest can be set on full farms. On final save you had 140 cities, but roughly half of them were on the New World, so they only counted 1/2 toward land and population limit.

Organized trait was saving you 886 :gold: per turn on final save. Compared to the total 10k expenses, Organized is not a very important late game money saver, but nevertheless, it was compansating a more significant portion of total expenses in the early and mid-game.

I think Darius is probably the best leader for a shot on future 1M+ scores. You might not believe it, but Charlamange is the best leader to cut down late game costs. He would have reduced the 10K expenses down to something around 6K.

Late game math on Charlamange vs. Darius would be:

3.5k saved from maintenance for Charlamange, 1.3k saved from inflation for Charlamange again, -850 for not being Organized, -1k for not being financial (20 more :commerce: per city * 25 full commerce cities * 100% gold multiplier). He would outperform Darius in late game by 3k gold per turn.

In conclusion, Charlamange beats any Financial or Organized leader in late game, including Darius. Yet, he won't have an as good early game, so probably not the best 1M+ scorer.

WastinTime
Apr 13, 2008, 09:51 AM
Thx for digging up those numbers. I would have thought Org was better than that, but I see Distance and Corp. maintenance is the real problem, not civics. For corporations, how does the -25% from Free Market stack with -75%? probably not -100%. You still need to pay the ~19% of total corp cost.

On a water map, Financial would get a boost from all the water tiles.

I suppose Zulu is also something to consider (-20% maint) but then you need a barracks in every city.

Imperialistic (from the burger king) would be a nice trait for all the settlers. Too bad protective is so sad. Maybe he would work well on a lower difficulty.

VirusMonster
Apr 13, 2008, 11:12 AM
The way they stack is 0.75 * 0.5 = 0.375 total maintenance. The 10k :gold: total expense in final save of KillerCane's 1M game was with this factor.

Charlemagne would have 0.75 * 0.25 = 0.1875, a significant reduction.

Zulu is interesting, because the Impi starts with Mobility making it pretty much like a Persian Immortal. I guess it could harass as successfully, but it does not have the 50% bonus against archers, so capturing cities will be problematic. On the other hand, it can be promoted CityRaider I&II to balance it out. Still not as good as an Immortal.

I am not exactly sure how the Ikhanda and Courthouse effects are put together. A single (0.5+0.2)=0.7 reduction multiplier (70% reduction) or 0.5*0.8=0.4 (60%reduction). The 10% difference is rather significant, since it could traslate roughly to 1k :gold: per turn in the final turn.

edit: I checked on WorldBuilder, they are added for a total of 70% reduction. That is almost as good as a Rathaus. I am starting to think Zulu might be the new ultimate 1M hero :) Barracks get build faster too due to Aggressive trait, and you want to build the Granaries early too for growth, so perfect combo.

killercane
Apr 13, 2008, 11:29 AM
This is very interesting. No one mentioned the cheap courthouses with organized though, and there arent any other traits that jump out at me as more important, only UUs and UBs. Financial makes all water tiles +1, windmills +1, and all that jazz aside from cottages. Getting electricity or whatever gives +1 commerce to windmills provided a sizeable (300 gpt?) jump.

WastinTime
Apr 17, 2008, 01:19 AM
I "theoretically" broke 2 million!

600K on Cheiftain = 2 million on Diety (I think)

So now I just need the same (or better) population, land, wonders, and tech ratios that I got on the Chieftain--Epic speed game along with a similar finish date.

Land = no brainer, you'd have the % land in either game.

Pop = should be easier to get cities planted on Marathon speed. Epic was really slowing me down. Cash flow might be a problem. I bought a lot of stuff.

tech = this will be the hard one. Trading w/ Diety AI helps the early/mid game tech, but late game will be hard to match the easy teching on Cheiftain.

wonders = easy. I didn't build many wonders. Should be able to work some more in.

finish date = easy to beat my 1520 AD finish. Score will skyrocket if I finish around 1200 or 1300.

killercane
Apr 25, 2008, 04:28 PM
Here is a rough draft of that article if it helps someone. I dont have time to finish right now with a baby due any day and work, but instead of it sitting on my hard drive for a month or two I will post it here for whoever would like it to read. It is unfinished but I will hopefully get to resume finishing it in a few weeks.

@WastinTime- dont forget city maintenance at Deity and its much harder to keep the AI from building the Hanging Gardens... did you really have 3.5x pop? So a huge map would have 10,500/3000 population?

Background

Inspired by VirusMonster's high score game, I began playing some Huge Immortal and Deity games in order to see how high a score could be attained. The results are interesting; I was able to score 700k my first game! Many people have asked for a writeup, but I thought I could hit a million with the BTS scoring system, so I tried a bit more. The final result was 1,085,000 points. Here is how it was achieved and why the high scores can reach higher. Hopefully someone else will take up the challenge.

BTS Scoring

Im certainly not an expert in BTS scoring, but the gist of it is the added turns in BTS allow for a higher multiplier with the BaseScore. In other words, a domination victory can be achieved in 400 turns on huge marathon, regardless of CIV version; however, BTS grants you a bonus since the max turns are significantly higher. Furthermore, the CIV equivalent of agricultural civs, Corporations, theoretically allows for EVERY city to have 15 extra food at least from Sids Sushi. The AI is a bit better, but just calls for different tactics than previously (ie, they are still bad). Add in events like the health generating event and BTS is the only choice for scoring high.

Civ Choice

I chose Darius. He has 3 things going for him: 1) An early UU that can take cities by itself; 2) The organized and financial traits- organized is very significant as pointed out by A Turkish Guy and VM for managing large empires and financial furthers that ability; even though you wont have many cottages for the majority of the game, the bonus to water tiles, special tiles, and windmills is huge; 3) The UB is pretty good, surprisingly, adding what is essentially +2 food.

Since this game I have tried Inca and Ragnar on deity with similar settings and really, you can use any civ. Inca can hit a wall if you have too many cities (no organized trait), and industrious does not make up for the lack of organized at all, it’s merely a small bonus in a couple of wonder building cities. Their UB is good, and the UU is excellent if you can economically handle having large numbers of them roaming around the map with catapults. Ragnar has swordsmen almost as strong as praets coming out of the gate (aggressive), is financial, and his galleys move 4 tiles with the UB and circumnavigation bonus on archi tiny islands. You can island hop and try to get free hut technology with his starting scout and a galley.

Map Settings

Terra was chosen. It has a nice domination limit, and it also means you have a nice new world to colonize, picking out the best food spots well before the AI can grab them up. It also provides a safety net in case you destroy too many AIs; you HAVE to get to the new world to reach domination so you could theoretically dominate all the AIs in the old world without worrying about domination. Finally, using Terra essentially gives you a free tech, Astronomy; there are so many huts on the new continent on huge size, that ferrying over a scout/explorer with caravels gives a very good chance to pop it out, saving several turns of research. The following are the techs that can be popped from huts, perhaps you can get more than astronomy with a fine tuning of tactics:

=====Ancient=====15=
Fishing
Hunting
Mining
Mysticism
Agriculture
Archery
Priesthood
The Wheel
Masonry
Pottery
Animal Husbandry
Sailing
Bronze Working
Writing
Horseback Riding
=====Classical======10=
Literature
Iron Working
Mathematics
Drama
Monarchy
Calendar
Construction
Compass
Currency
Metal Casting
=====Medieval======1=
Music
=====Renaissance=====1=
Astronomy

Now, there are several other map types to be tried, including Big and Small and Archipelago and similar map types. These might also yield high scoring potential.

Speed- Marathon

Low Sea Level (maybe you might use high seas for less land and more sea population?)

10 AIs- this gives you a 56% domination limit as opposed to 51% with 17 AIs.

Barbs- None

Goody Huts- On. On deity you may consider turning huts off as they help the AI too much, but on Immortal you benefit even if you never pop a single hut. The AI researches comparatively slow, and you do too in the beginning. If a couple of AI pop key techs early, you get a research bonus for them knowing the tech. Usually you can get a couple hundred gold as well, allowing for maximum research in the early going.

Rivals- Mansa, the Germanies, the Indians, Mehmed, Washington, and Elizabeth

Attacking Huge Immortal

Expansion: Rapid expansion is always grand. Your friend in expansion is the cost/benefit ratio. Since settlers on marathon cost 300 hammers, any city you can take with less than 5 Immortals (60 hammers) lost is a net win. Since most unthreatened AI cities have 2 archers defending, and immortals have favorable odds against non-hill, non-cultural defense cities, military is almost always the best build early on. Immortals are even the best unit usually when there are 5 archers in a 40% culture AI capital; theoretically, each of the first 5 should damage the archers enough to allow 5 more to capture the city. With the retreat bonus of Immortals, you should still be able to make a net gain in those situations, not even counting the city improvements like granaries that often come with these types of cities and the terrain improvements already in place.

Furthermore, you do need quite some workers to be successful on this level. The easiest and most cost effective method to acquire at least 3 workers in the early going is to worker steal. BTS AI is better at covering their workers with archers, but still many times they will leave the worker all alone during the first 100 turns to be captured by a warrior. Sometimes they are just irrigating on the outer portion of the BFC, sometimes they are roading to a second city, but if you are patient you will capture one from each civ you try. I like to "Skip Turn" rather than Fortify the warrior so I dont forget about him sitting there and miss worker stealing opportunities. Never steal one in a situation in which you will end up on a tile with no defensive bonuses and that is right next to the capital so that an archer can kill you; the exception to this is if the worker is mining metal there before you get immortals. Then, by all means attack the worker; even if your warrior is killed the AI will often delete the worker anyways even if there is no more threat. If you have some aversion to worker stealing, building your own workers works fine, it just takes longer. For marathon level in the early going, I like to have 3 workers per city, which enables 3 turn forest chops, 7 turn irrigations, and 4 turn mines. Once your empire is a sufficient size in the ADs, you can turn each and every city to build a worker and before long you will have 100+ to help irrigate and make population.

Military: 4 initial warriors to start, at least the first two produced using the highest shield tiles available at the expense of growth. You never know if the warriors getting to an AI quickly will pay off, but the benefits of a new worker earlier are immense.

The second phase is production of Immortals. This phase lasts until 500 BC or so. The concept is simple: make as many as possible and take cities, workers, and harass the AI. If playing as another civ, swords/cats/axes is the concentration.

The third phase is Immortals+. This series of wars may be fought with cats, axes, spears, maces, etc., depending on your tech level at the time and current empire needs. Cats do two things: defense against spears and softening up of cultural defenses so that you can again attain a satisfactory cost benefit ratio. They also can collateral damage, but I generally refrain from using them to attack cities unless I have to or they are promoted to CR3 to achieve high odds. It is better to lose a 60 hammer Immo than a 100 hammer cat in city attacking.

The final phase is Currasiers+. Once longbows appear, you need a bit more punch for your attack, and curassiers provide that; they can be combined with airships to increase their odds of winning or later be turned into cavalry for a final assault. The AI should never reach rifles unless you let them or frankly, you are doing something wrong, wild, or crazy.

Of course, if you desire to build the Hanging Gardens, you have to play a completely different game. Your wars will be conducted entirely with cats and ancient units, and you have to eliminate everyone down until you have only 1-3 AIs left, hopefully in subpar food locations. This is a two edged sword; you can do whatever you want at this point, but your research will be bad (no foreign trade routes) and the AI will not trade anything they have researched (no problem, just steal things like Feudalism if they make it that far). Finally with this method you must make many more settlers than in the game I mention here; you cant take cities from the AI as they wont have many good ones to give!

Exploration: Your initial scout will do a lot of things. With no barbarians, he is not at risk of dying unless you are in a war with an AI. He will pop huts and get a general description of the landmass, but most importantly he will find horses in your nearby vicinity for your first produced settler to acquire.

AI Research: For the purposes of a game like this, you really want the AI to research well. They will provide several nice techs, metal casting, machinery, engineering, calendar, ironworking, etc. Most importantly they will research alphabet early as a priority, which means you dont have to research it and can concentrate on economic technologies, like Currency and Education.

Research: Initially, you want to acquire Mining, Bronzeworking, Animal Husbandry, Pottery, and Writing. If you plan on building the Oracle, naturally you would want to research to Priesthood before Pottery and Writing, and then enjoy a research bonus on Writing without having delayed finishing the Oracle. You can take a variety of techs with the Oracle, but CoL, Currency, and if you are lucky, Civil Service or even Education is possible.

Walkthrough

So, in the year 4000 BC, King Darius decides to settle on a plains hill, with a cow and a gold hill. The greatest start in the world it is not, but it will do for our purposes here. Really this game was to try the Terra map out, and work on the endgame strategy, which it accomplished.

Phase 1 to 1000 BC

As you can see in the replay, I stole workers from France, India, America, and England during this period. Worker stealing is the key to fast growth, and you really need 3 in the early going for your capital to provide 3 turn forest chops, 4 turn mines, 2 turn roads, and 5-7 turn irrigations depending on terrain. I think 2 commerce tiles is all you really need, and if you can get pottery early, several floodplains can make up for the lack of a gems or gold.

The first builds are typically 3-4 warriors to go out worker stealing in all directions. In worker stealing, naturally you want to try to grab workers while they are on the outside of the BFC. If they are covered by archers, just wait until the AI builds a second city (turns 50-75 roughly); at some point they will leave them unprotected and you can move on in. A helpful tip I use is to “skip turn” for each of these waiting warriors each turn so you dont forget about them and can act as soon as possible. Also, you can camp a resource if its on the edge of their territory; they will get around to it eventually. The AI likes to irrigate special tiles, and if doing nothing else road around their territory. An animal husbandry resource like cattle they will wait usually if they have other tiles to work; so if you are camped by a cattle, and it gets up to turn 60 or so, go out looking for the second city and catch the worker roading.

The next project is a settler to found by horses (if playing Persia/Egypt). You want a high shield location with decent food, as this city will produce quite some military. You can pre-road if possible, with the goal to hook up horses asap. Here horses were close to the west, and I could hook them up and then road towards India, who had Stone and Stonehenge built.

Indian war

Pretty quick matter overall, built 7 or so Immos and declared. Received good RNG vs. the archers at Delhi and didnt need reinforcements to take Varanasi. Finished the road over to Delhi from Persepolis, and sent the Immos west to Napoleon.

Napoleonic War

Another quick matter. Moved in and grabbed his cities. Nothing much else to report as it was straightforward. I seem to remember he captured Delhi with a wandering archer but I took it back asap.

Washington and Elizabeth

Managed to make some spears, but eliminated the copper asap on both of them and poked around until I had them where I wanted them with enough forces in place. BTS AI will move units to reinforce a threatened location better, so if you camp one city on the periphery, often they will move archers from the capital which you can kill in the open field and move to take the capital. You shouldnt have any units threatening the capital while you do this however as they will sit there.

In attacking cities with 20% culture and no hills, I usually send 2 immos for every archer, and if there is 40% defense or hills involved, 3 immos per archer.

At the end of the English war, I am on the verge of bankruptcy. Luckily I was able to plod along to Priesthood barely and take currency. Currency is the most important ancient age tech for games like this, as you can trade gold from the AI and enjoy the 2nd trade routes for your existing cities to further solve your gold problems. To achieve a fast tech pace, early on you will only have the libraries and academies, so you really want to run high research at a deficit to distance yourself from the AI. Dont forget to look for gold trades after you build wonders.

During this phase, Delhi built Pyramids and then Hanging Gardens; since I hadnt sent immos to harass the far away civs, I couldnt wait to build the HG. I typically have two wonder building cities during this time, the best production city and a Moai enhanced coastal city to build The Great Lighthouse, Colossus, and anything else lying around like Temple of Artemis. I wouldn’t worry about wonder distribution and GPP pool contamination as you will just use most of the great people for golden ages anyways and you will need a variety of them. The most important wonders are Pyramids and Mausolleum of Maussolos

Events

There are several events in the HOF mod that will be helpful to you. Consult Ori’s Strategy article on Events for full details.

1) Health tests- free +2 health with possiblity of -1 population. If you get this event early before growing to size 2, there is no penalty.

2) Guns not Butter, Marathon Runner, & Sports League- free golden ages if you have the right wonders under your control. If you get these events, Taj, and 5 GP enabled golden ages you can spend 216 turns in GA with the MoM!

3) A few of the others are marginally helpful, like Blessed Sea (free prophet), Horse Whisperer (+1 food for stables), and Warships (+2 commerce for harbors).

Phase 2 to Education

Once Currency is in, and you are playing for building HG/reducing all AI to insignificance, your main purpose is to maximize science through tech selection. Lit (+15 science (2 free scientists+Representation+library), CS (bureau), Metal Casting (Colossus), Calendar (dye/incense plantations and the powerful Mausolleum) and Education (Oxford) are all important to concentrate on.

Espionage

Since you have whittled the pool of AIs down now, you can concentrate your EPs on them to 1) Steal a few filler technologies and even an important one like calendar or aesthetics and 2) keep tabs on their research and what they are building (like the darn HG). There is no reason to detour to Music for the free artist at this stage if the AI isnt in a position to claim it. You can pop music from a hut later if you play on Terra; Big and Small might be a different story as there may not be as many huts if you are using massive continent/small island setting.

Technology

You probably will need a few courthouses in the farther away places, and this is also a good time to build the Forbidden Palace. The FP needs to be built as soon as possible in a farther away location as its benefits are immediately significant. If CS/Education will provide you with the best science outcome, shoot for those. A scientist for Education isnt bad if you have 11 cities w/ libraries ready to build unis AND you have stone available for the capital to bust out Oxford. If you have several coastal towns, now is the time to get a coastal wonder powerhouse going, with Moai (Stone again!), Great Lighthouse, and Colossus to add a bit of a boost to your bottom line. This city might even need the National Epic to produce your GPs in solid numbers.

Philosophy is a nice to have at this point so you can take a golden age and switch to CS/caste/Pacifism.

Phase 3 to Biology

Economic civics like Free Market or Mercantilism can add to your bottom line significantly. 50 cities x an extra 6 beakers or 3 gold+3 beakers can certainly make a sizeable contribution to the economy. Really Mercantilism might be better to beeline than education if you have many cities and Oxford isnt worth as much due to 10-20% science slider.

Technologically, you want to head to Biology first, Communism second. I tried both approaches and it seems getting the growth first is more important in your trip since you cannot spread Sushi in Communism anyhow. This may be different from the other jargon in this article (I have to vet and clean it all up). This is where you want to pop Astro, and free up Sci Method for research, and on to Liberalism->Biology.

Wall Street should be a priority in one of your holy cities; it will house your Sushi Corporate HQ as well, and cut down on the vicious corporate maintenance costs. Gold producing buildings should be your priority at this time rather than unis or observatories (probably the last things you would want to build at this time).

Finally, this is the time when you should be producing settlers and workers. I like to switch all cities to workers for 2 rounds, and get up to at least 150. Hagia Sophia hasnt been as much a of priority previously, but perhaps it should be built sooner rather than later.

Phase 4: The Endgame Milk

There are a number of interesting things that I have discovered RE: the last phase of the game.

Civics

For the most of the builder turns, you should be in Universal Sufferage, Nationhood, Serfdom, Pacifism, and Free Market. US is obvious, it allows you to buy rather than whip away precious pop, and FM provides -25% corporation costs. The others are seldom used government styles so why are they included?

-Nationhood has no upkeep and 25% espionage, also a barracks provides +2 happies. The upkeep can be quite significant, and eliminating that gives you more $$ than bureau or free speech. The barracks bonus will come in handy for your large cities. The 25% espionage is handy for keeping tabs on what the AI is building, and destroying the Hanging Gardens if they start building it. I also destroy the aqueduct in the HG city, and the AI seems to mill around and build a lot of other stuff until they get back around to making the aqueduct and starting HG again.

Free speech also has the unfortunate culture effect that of doubling Sushi culture, and those perfectly placed food cities by floodplains will get a second and third border expansion earlier, and force you to either plant a city in the desert or just ignore them as wasted tiles. At this point the extra commerce from cottages is frivolous. You can probably use FS with communism/representation in your latter part of the research phase.

-Serfdom- workers work twice as fast and LOW upkeep. The benefits for growth are obvious with faster workers (8 turn irrigations of grassland w/ Hagia Sophia!) and the low upkeep is important for cash.

Pacifism- no upkeep and faster GP for those crucial 5th and 6th golden ages. If you have some extra GP lying around at the end, try bulbing some technology if you wont have enough for a final golden age.

Some other civics that will be useful:

-Representation- during your research phase to Radio, this will be your key civic. In such a large empire, you research at 30% without state property, and the Rep bonus is huge.

-State property- should be your next stop after Biology. Cutting the majority of city maintenance allows you to run 60-70% science and finish up your last bits of research. The bad thing is no spread of sushi while running it :(.

-Free speech- as mentioned above, can provide a cash/research boost for a while pre-Sushi as long as culture doesnt get out of control.

-Environmentalism and Free Religion- both are used at the very end, Environment for the extra health growth and FR for some extra happiness and small research boost to finish out a tech or two. Note the +25% corporate maintenace with environment, if you are not through spreading Sushi you will soon go broke by continuing to expand it.

Actual Milking Pros and Cons

The two most important worker improvements are farms and windmills. We want explosive population growth, and therefore will farm over everything, leaving those research/cash boosting cottage tiles (now towns) for last. The only things you dont want to farm over are: 1 each of your health/happy items, and pigs. Cows? Irrigate them. Sheep and bananas? Doesnt matter if you irrigate or pasture/plantation them; the only consideration is having a few extra of these "neutral food" healths to trade to the AI for gpt, if available. The +2 food for farms in biology is very significant.

On windmills, naturally you want to keep some production cities to build wonders like Rock and Roll and Christo Redentor up until the end. Electricity provides a large (15%?) boost to your economy based on the additional commerce output for these improvements. It is also on the way to Radio.

So what should cities build? I usually build growth, unit, happiness, health. This allows your cities to keep growing; happiness ahead of health is important as an unhealthy citizen still produces food whereas an unhappy one does not. This works out to a build pattern like:

-granary, lighthouse, catapult/archer/immortal (if you never take HBR!), theatre, market, colloseum, grocer (+5 w/Persia), aqueduct, hospital, with harbors thrown in for coastal cities at some point in the middle.

You want to build the most hammer efficient city improvement available for the city’s needs. What I mean by this is, for example, a unit is the cheapest way to maintain happiness, next is a theatre which will provide +2 happy with dye. You may build a market before the theatre if there are significant happiness problems in a high growth city. A colloseum provides +1 happy regardless of lux slider. The apothecary is a great build for unhealthiness, and the aqueduct is cheaper hammer wise than the hosptial in terms of health per hammer. If you still need happiness, you can try spreading religion and building temples for smaller benefits, but if the problem is empire wide you will need to raise the culture slider.

Another note: Once Communism is in, you will go on a buying spree, since every city simply must have a courthouse with Sushi. Otherwise the maintenance is in excess of 100 gpt! 50 gpt is a lot easier to swallow. It is also probably more efficient to buy market, apothecary (or equivalent), and a bank before you buy the other buildings. Now a city limited to size 15 or so growth probably doesnt need these things, but for most of your cities they are necessary so might as well buy them early, enjoy the benefits longer, and get the best return on your investment.

Wonder Buying

Another thing with Communism is the Kremlin. Once it is complete you can buy wonders for less than 10K gold with a turn or two invested. Since you make 3-5K gold at this stage, its not big deal. I have now taken to buying Redentor, Rock N Roll, etc. and it is very helpful. Your two or three wonder building cities can then build other things like execs and settlers. This also means that you only need to set off enough golden ages at a time when you know you can reach Radio for civics changes.

City Placement
Where should the cities be placed? At this point in the game you are making cities for the high growth spots on the map with the most food. You should also look to see how many land tiles are in a new city’s radius, and attempt to minimize those. 1 tile islands with seafood resources are the best spots. They are negligible in terms of domination limits but provide a lot of high growth potential with Sushi.

Floodplains can be great, but the effect is minimized by having desert and moutains present. One desert in the radius wipes out the bonus from 4 floodplains! If you have 2 deserts and 6 floodplains, an all grassland irrigated city comes out as the better spot to place a city.

I would shoot for 150-200 workers at this point to fully improve everything in the fastest possible time. An easy way to do this is to turn every city to workers, and buy them/produce them naturally before you reach Sushi. You will have +/- 150 cities but many of those are already improved, so 200 should be enough with Hagia Sophia and Serfdom.

Building the Hanging Gardens at the end of the game

If you went the route of destroying the AIs early and keeping them from building the HG, you can get a nice 3-10% score boost at the end of the game by finishing the HG on the last turn. To do this, you should build the HG up until one turn left, and leave it in the queue until the end. On the final turn, settle as many city spots as possible on the map and/or attack as many AI cities as possible and take those. You can make hundreds of extra population by doing this.

Im sure there are several other things that do not come to mind atm. But basically, with the endgame you want to do a couple of things:

1) Grow population. This is accomplished by keeping your people healthy and happy, and having enough workers to completely irrigate/windmill the whole of your territory, which should be the most food rich land on the map. Further, spreading Sushi should provide about 15 extra food per city.

2) Avoid the domination limit. Be very careful with Sushi's extra culture, it will easily get you past the limit unless you pay close attention.

3) If you can achieve max population (3000 citizens) by turn 500, you should make 2 million points on Immortal.

VirusMonster
Apr 26, 2008, 01:57 PM
Hey, nice writeup Killercane, it was a good read. I also looked up WastinTime's 600k submission on Chieftain. I can tell you for sure that you won't be able to get Education from the Oracle at Deity difficulty :) Your tech speed will be cut by half at the very least, so 2M+ not as easy theoretically as you claim :) I have never managed to get even Civil Service from the Oracle at Immortal difficulty, Currency is probably the best pick.

In the meantime, I have been wasting some super good starts, 4 gold, 4 gem, combination of luxury resources of all sorts :) Most important advice I can give to the readers of this thread would be producing sufficient number of Immortals to capture the cities. Killercane's writeup gives some average numbers for city capturing. 2 Immortals per archers on a non-hill, non-culture city. 3 Immortals per archer for a 40% culture or hill city. Do not try a 40% culture hill capital without catapults. Just capture the rest of the cities and pillage mine and stay in war until you can bring a single catapult the nullify the defenses.

I also don't think you can delay building the Hanging Gardens until the last turn on higher levels. Perhaps it is just me :) On a huge map, the distant AIs will be too hard to reach to pillage all metal and horse resources. On a standard size map, it could be possible. Also, the prebuilded buildings start losing hammers after 30 turns if you leave them too long in the queue.

I am also interested in hearing WastinTime's experience with Charlamange :-) Can you post some saves up here WastinTime? 2500BC,2000BC,1500BC,1000BC,500BC??... I want to know how many cities you had at these dates.

edit: still no answer from WastinTime regarding the saves... so just forget it... After some wonderful starts, I am convinced Shaka is the best leader for 1M+ scores... Warrior rush to capture first capital :) 1400BC Bureaucracy on Immortal :)

VirusMonster
Apr 26, 2008, 02:30 PM
1) Not taking Communism with Liberalism. Research is no problem with State Property. SP is the greatest civic ever for your massive empire...well, other than the fact that you cannot build Sushi or executives for Sushi while running it. I thought you would be able to run the thing but just not derive benefits; alas, you cant spread it at all in SP.


You mean instead of Biology one should pick Communism from Liberalism?

WastinTime
Apr 27, 2008, 09:03 PM
@WastinTime- dont forget city maintenance at Deity and its much harder to keep the AI from building the Hanging Gardens... did you really have 3.5x pop? So a huge map would have 10,500/3000 population?

Yea, I realize Diety will be nothing like my Chieftain game. I was just teasing mostly, but I'd like to try to emulate that game as much as possible.

3.5x pop, yes. I believe it was 4200/1200 approximately. I'm convinced that huge maps are not ideal for high scores. You just can't make that many cities that fast. You'd want at least 300 cities.

Noobilator
May 09, 2008, 07:37 PM
This thread seems to be a gathering of the greatest experts ever known in Civilization IV. The information this thread offers is very indepth. The replies proved to be very useful too.

This thread convinced me the power of Organized. I used to think it is a worthless trait. I was wondering if someone can give me an advance analysis of leaders traits and the notable UU/UB.

killercane
May 13, 2008, 04:21 PM
Well, you mean leaders with organized traits or overall? If overall, the only UUs you need are early (chariot types/sword types) that have the ability to kill archers at a good cost/benefit ratio OR another tactic is let the AI use their bonuses and grow their cities large w/ improvements while you beeline to Cataphracts/Curassiers/Cavalry or their UU equivalents.

killercane
May 14, 2008, 12:31 PM
Small update in the long strategy post above.

VirusMonster
May 14, 2008, 04:02 PM
Small update in the long strategy post above.

what is the update?

WastinTime
May 14, 2008, 04:38 PM
Sorry I missed the request for cities at certain dates:

2500BC : 5 cities
500BC: 12 cities
70 AD : 20
750 AD : 37

killercane
May 16, 2008, 12:54 PM
what is the update?
Some stuff in the pre-Education and pre-Biology discussion.

I tried some smaller maps but isnt your theoretical MaxScore always going to weight your total score downward in the calculations? That is what it looks like and it seemed to bear that out in the milking phase.

WastinTime
May 17, 2008, 03:55 PM
I tried some smaller maps but isnt your theoretical MaxScore always going to weight your total score downward in the calculations? That is what it looks like and it seemed to bear that out in the milking phase.

You're right, I didn't see that effect in the formula until I looked closer. A population score of 1000-raw/1000-max (ratio 100%) gets beat by a pop score of 2000-raw / 3000-max (66%).

EDIT: Even with this set-back. I remain convinced that huge maps are not the best for high scores. In the example above, you still need to DOUBLE your raw population just to get roughly the same score. I think it's going to come down to a map that is large enough so you don't hit the land limit so soon, but small enough to get a high pop ratio. Really little maps also have resource problems (like some resources are completely missing from the map.) I'm thinking Standard size will score best. That would be nicely designed (or just lucky) by Firaxis.

killercane
May 19, 2008, 03:29 PM
You're right, I didn't see that effect in the formula until I looked closer. A population score of 1000-raw/1000-max (ratio 100%) gets beat by a pop score of 2000-raw / 3000-max (66%).

EDIT: Even with this set-back. I remain convinced that huge maps are not the best for high scores. In the example above, you still need to DOUBLE your raw population just to get roughly the same score. I think it's going to come down to a map that is large enough so you don't hit the land limit so soon, but small enough to get a high pop ratio. Really little maps also have resource problems (like some resources are completely missing from the map.) I'm thinking Standard size will score best. That would be nicely designed (or just lucky) by Firaxis.
Well archi maps should get you the highest but you have to overcome the speed of conquering limitation of having to build boats. Perhaps I am too set on building HG; HG is a major reason why Immortal can best Deity's scoring bonus though (earlier victory date and pop bonus from HG can add up to over .2 scoring... assuming you settle all available land on the map the final turn and get those pop numbers doubled with HG= 10% pop boost (300 pop/3000 max pop).

So the question is, what size map will allow for the fastest conquering on an archi map and thus the highest score? I didnt find standard better than huge, so that only leaves large to experiment with.

End of game HG is buildable on huge immortal; you can take all AIs out before it becomes a problem. How many AIs can you take out on a Deity map before they get the chance to build HG? Can it be done on standard archipelago? Im thinking yes. So that leaves Deity standard vs. huge Immortal.

In the final calculations, what value would you put on the lower/higher maxscore on a huge vs. large vs. standard map?

WastinTime
May 19, 2008, 06:10 PM
Your valuation of HG seems inflated. I'm pretty sure it was only 15k -20K bonus on my 600K game. So about 3%.

A 10% increase is population is not a 10% increase in score.

10% is also a lofty goal for pop boom. If your other cities have grown properly to size 20-25. Then HG is only 5% or less pop growth. So 3% score increase sounds about right.

Do you really think that a 300 city game is going to score best? Won't it take too many extra turns to get up that high? I would think a map that can support 200 cities is big enough. Maybe 150 cities could score 2 million if done fast enough.

What was your colony maintenance/turn? something like 10,000? On a island map it is zero.

killercane
May 20, 2008, 07:37 AM
Your valuation of HG seems inflated. I'm pretty sure it was only 15k -20K bonus on my 600K game. So about 3%.

A 10% increase is population is not a 10% increase in score.

10% is also a lofty goal for pop boom. If your other cities have grown properly to size 20-25. Then HG is only 5% or less pop growth. So 3% score increase sounds about right.

Do you really think that a 300 city game is going to score best? Won't it take too many extra turns to get up that high? I would think a map that can support 200 cities is big enough. Maybe 150 cities could score 2 million if done fast enough.

What was your colony maintenance/turn? something like 10,000? On a island map it is zero.
Well I said 10% pop boost 300/3000. That will be less than actual growth. I have it calculated at 5-8% as the value for HG. That plus early win bonus as you will save 200 turns on immortal on a bigger map not having to fight so many enemies is >.2?

But it should be simple to figure out. Attached is the scoring formula again. What is the maxScore for say a huge immo map vs. standard deity vs. huge deity and 2) what amount does pop play in the final statistics. I forgot algebra long ago...

maxScore : The theoretical maximum rawScore for the map. Can also be seen in the above breakdown.

factor : A weighting for the score components. By default these factors are as follows:
pop = 5000
land = 2000
tech = 2000
wonders = 1000

currentTurn : The current turn number. Note that this does not start at 0 for games other than the ancient era!

maxTurn : The turn limit for the game.

Raw Score:
Pop: 1 per pop
Land: 1 per tile
Tech: 1 per Ancient tech, 2 per Classical, 3 per Medieval, ..... 7 per future
Wonders: Don't know, but not worth much anyway

killercane
May 21, 2008, 09:07 AM
So maxscore is or is not a constant? I got 2.397 million as MS for the million point game but... cant make heads or tails of it.

WastinTime
May 21, 2008, 09:32 AM
So maxscore is or is not a constant? I got 2.397 million as MS for the million point game but... cant make heads or tails of it.

It's different for each map, but it's constant for the duration of the game.

killercane
May 21, 2008, 10:24 AM
If initial score is 28, how do you get 1.085 million for the following numbers from that game?

Pop- 2755/2842
Land- 2115/3881
tech- 209/334
Wonders 125/310

MaxScore is 7367. Turn is 604/1500. Total score (needed or not?)= 7589.

VirusMonster
May 21, 2008, 06:49 PM
Just plugin those numbers to the matlab formula I posted in the first few pages of this thread... All work is already done. Download matlab student edition (should be free) and copy paste my code into a new file, then call the file name from the command line. The code should run and return your 1M score. I actually did it long time ago to figure out what made you score so high. A combination of finishing earlier than my 500k game and also having higher population was the main reason for your 1M score.

All scores are calculated seperately then added togehter using a weighting. 50% for pop, 20% for land, 20% for tech, and %10 for wonders.

And you've also posted the wrong formula to calculate the pop score. There are 2 formulas, and the pop formula uses the exponential version.

Here is the correct formula for the population score.

popscore=(popfactor*poprawscore)/(popinitialscore*power(floor(popmaxscore/popinitialscore),(turnonwin/maxBTSturns)))

The power of the exponential term is (turn on victory / max turns) and this term goes to the denominator of the formula.

Then, the numerator of the formula has your raw score.

All other values are constants (maxscore, initialscore, weighting) so they dont matter in maximization.

Now, how would you maximize a formula that has a linear rawscore component in the numerator vs. an exponential term? The exponent of the exponential term ranges between between 0 and 1, depending on the value of (turn on victory/max turns). Then, the later you finish the game the greater the exponential term will be, hence the greater the denominator term, and hence the lower the total value of the formula.

On the other hand, if the raw score in the numerator of the formula is too small, even if you get a relatively small denominator value for finishing early, the final value of the formula won't be so high.

So how would you optimize the settings to get the highest population score, which corresponds to half of the final score?

Realize that the numerator of the formula (rawscore factor) grows exponentially in most of our games. Data from my 500k shows population doubled from 100(at 1AD) to 200(at 500AD) and then 400(at 1000AD) every 500 years.

But in my game I did neither have sushi, nor biology for the extra food on farms.

The ideal scenario for high score would get Biology and Sushi as early as possible (ie means very fast tech speed) and make use of them for a long time to maximize score.

Doubling the population every 500 years (50 turns between 1AD-500AD), (100 turns between 500AD-1000AD) compared to the 1500 total turns on BTS games translates that your score is actually getting higher, because to come up even with the score decrease for finishing 10% later of the total number of turns, the population raw score just has to double.

Read this last sentence more carefully if it does not make intuitive sense. The exponent term in the formula (turn on victory/max turns) ranges between 0 and 1. Take a calculator and take exponent of your max popscore(any constant number will do as well). Make the exponent to be 0.1, 0.2, 0.3, etc... all the way up to 1. You will see that the game expect your pop score to double every 10% of the game.

In my 500K game, between the 1AD-1000AD period (150 turns=10% of the 1500 max turns), the population did more than double, in fact it went 4 times higher, so my score went up as well.

Now Sushi and Biology change the dynamics to a great extent.

How fast do you grow with Sushi and Biology? Well let's assume you have 200 cities near end game at 50% land mass. Let's assume you get +15 food from Sushi(for Terra, other maptypes give more). I assume map is grassland rich (ie tropical). Also let's assume you farmed most of your land and get +2 extra food per each farm. So 20 farms = +40 food. Add the Sushi it is roughly 55 extra food per turn.

How much does it take to grow the city with a granary? Let's say an average 4 turns, because you can only get the maximum +40 farm food if you have enough citizens. At lower city sizes, the population growth will be mainly through the Sushi and not the farms. Then the +15 extra food would mean growth roughly every 4 turns.

Then, in 150 turns, which is 10% of the total 1500 turn game lenghth. You could grow a city to 40 citizens. Well, you can't sustain a city at size 40 because of the health&happiness cap. Well you can if you are fine with angry citizens and expect the overcrowded penalty be compensated through the many farms and Sushi. Sizes 25-30 are sustainable without the future techs, but you must build the right buildings.

So as you can see once the Sushi is spread to all your cities, in less than 10% of the total game lenght, in which the scoring formula expects you to double your score, you can skyrocket your population score, ie half of your total score.

Say you had 200 cities. 200x25 city size would effectively give 5000 population score.
For 300 cities and 35 city size would give 10 500 score. To simulate WastinTime's Chieftain population score, you would need 300 cities at size 35 for a Huge Terra map where max popscore is roughly 3-3.5k.

--------------------------------------------------------

So once the Sushi is spread to all cities, your empire should reach its maximum possible city sizes (ie 25-35 citizens) in less than 100 turns.

How fast can you spread Sushi? Well you can definitely poprush all executives in your already highly populated cities. They would cost you 2-3 population, not a big deal considering the city will grow back pretty soon with Sushi. You can also buy the executives for 500 gold. Say you have 20 large enough cities near endgame. Say you can also buy 5 executives every turn. So every 30 turns, you get 20+100 executives. If you have a total of 200-300 cities. Almost all of them should have executives running toward them in less than 60 turns.

Let us also not forget that many settlers must be built to achieve a total of 200-300 cities. Say at least 100 settlers must be built for 300 :hammers: each. That is a hard task. I expect at least 50 of your cities will have halted population growth for at least 20 turns to produce 50 settlers. So producing 100 settlers will take 40 turns. If you want to go for more settlers, let's say Sushi and goldbuying would allow you to get your settler goal in 50 turns.

So total milking phase lasts:
1. 100-150 turns spent on growing citizes to size 25-35.
2. 60 turns for producing Sushi executives.
3. 50 turns for producing settlers to achieve 200-300 total cities.

So milking lasts 200 turns. 200/1500, %13 of the total game length. Game would expect you to increase your total score by 2-2.5 in this time period. How much would score increase in this time period? Well your total empire population would reach from half the max population score(assuming before sushi and farms, you already had half the maximum population, data taken from my 500k game) to roughly 2-3 times the maximum population score. Don't forget population only makes up half the total score.

Then, the milking phase will exceed the formula expectation by (2.5/0.5)/2.5=2 times. The final score should roughly double maybe a bit more than double in the milking phase.

You can more than double the final score if your empire population before milking was half the max population score, but near end game before milking, it is likely that your empire population is already near half the max popscore.

---------------------------------------

Now, let's answer the big questions you have raised Mr. KillerCane :)

Deity or Immortal? Huge or Large or Standard?

First deity or immortal. Alright. 0.2 score difference is huge, but at what cost?

If you can finish a Deity game as early as an Immortal one, assuming completion of the milking phase as well, then of course Deity is better. Unfortunately, Deity is much more difficult in both expanding, keeping up in tech, conquering, etc.... In fact, if you look at the average earliest finish dates for standard size maps on Immortal and Deity for BTS, you will see no entries for dominaton or conquest.

Still let's assume Inca or Persian UU harass can get you early domination. Even then, once you are at war with all AIs, you might have a hard time finding a tech trade partner to boost up your research. Ok, let's assume you have kept Gandhi as your tech ally.

Research also happens slower for you on Deity difficulty. All these factors could end up delaying your victory date by at least 150 turns compared to an Immortal domination victory. I doubt under optimal odds your victory will be delayed by more than 150 turns compared to Immortal.

Then, say your score gets delayed by 150 turns. Well as I pointed out earlier, game expects you to double your score every 10% of the game length. So instead of 1.8M score for finishing on Immortal, you would get 2.0M/2=1M for finishing on Deity 150 turns later. Deity will give you 2.0/1.8 =1.1 times more score than Immortal.

The bad news is that the scoring formula expects you to increase your population by 1.1 times every %1.5 of the total game length. (25-30 turns on marathon speed for BTS)

Thus, your margin for error on Deity is less than 25-30 turns. If the game takes you more than 25 turns longer to finish on Deity than Immortal, then you are better off playing Immortal.

----------------------

Based on the discussion above and given the margin of error being so small between Deity and Immortal scoring system, I am sure Immortal is the best difficulty to try the highest scores. In fact, lower difficulties could score higher for finishing earlier.

----------------------------

Now map size discussion...

Let's look back at the population score formula. For a huge map, low sea lvls you get roughly 3k max population score.

Let's say standard map size gets 750 max population score.

The milking phase in standard size maps will last shorter. We assumed 200 turns milking phase for a Huge map in earlier discussion, say for a standard map milking lasts 100 turns.

The difference is roughly 100 turns, assuming everything else being equal.

Now, is this 100 turn difference worth the lower max score on standard size maps? 100 turns is %6 game length for BTS marathon. The formula expects the population to increase by 1.6 in this time period. Can this 1.6 expectation be compansated through the higher max score on larger map sizes?

Let's also make the final raw score to be 2 times more than the max score.
max popscore for huge = 3k , for large = 1.5k, for standard = 750.

popscore=(popfactor*poprawscore)/(popinitialscore*power(floor(popmaxscore/popinitialscore),(turnonwin/maxBTSturns)))

huge map = 5000 * 6000 / (1 * ((3000/1) ^(600/1500))) = 1 220 000

standard map = 5000 * 1500 / (1*((750/1)^(500/1500))) = 824 000

let's add large map for completeness.

large map = 5000 * 3000 / (1*((1500/1)^(550/1500))) = 1 026 000

So when would a 750 max population score map come up even in population score with a 3k max population score map?

Setting the scores equal and solving the equations for turn on win (500 and 500+x),

huge map = 5000 * 6000 / (1 * ((3000/1) ^(675/1500))) = 824 000....

So if you finish 175 turns later than the standard size map 500th turn finish date, your population score comes even with the 500th turn standard map score, assuming final empire populations are the same.

On the other hand, if you finish 175 turns later, your tech and wonder scores will probably go down. They compose 30% of the total score. How much will they go down? Well in 150 turns, if you do nothing, your score halves. They will halve as well.

Your land score, 20% of total score, will follow same dynamics as population, because large map sizes will have roughly 4 times more land than standard map sizes. Relatively speaking, if you were to finish on same turn, your land score from a huge map size would be 4 times higher. If you were to finish 150 turns later, your land score from a huge map size should still be 2 times higher than the standard map size.

Overall, since the population component of the total score is so important, larger maps will score higher despite later finish dates by roughly 150 turns.
--------------------------------------------------

Alright conclusion:

Huge map sizes finished at turn 675 or earlier get higher scores than standard sizes finished at turn 500 or later.

----------------------

Summary:

1. Immortal beats Deity, because the 2.0/1.8 = 1.1 times increase in final score would only be favorable if the game were finished not later than 30 turns than on Immortal. BTS Deity will really slow things down more than 20-30 turns so Immortal is much better than higher scores. The gap between Emperor and Immortal(1.8/1.6) is another 35 turns, the gap between Monarch and Emperor(1.8/1.4) is roughly 60 turns. The gap between Chieftain and Immortal (1.8/0.6) is 240 turns. These numbers are approximations, based on a 1000 max population score map.

2. Beating a huge map size at turn 675 or earlier will give you a higher score than beating a standard size map at 500th turn or later, asssuming all else being equal.

-------------

If you use this info of this post in crossing 2M+, you must acknowledge my name :)

killercane
May 21, 2008, 08:04 PM
Now thats a 5 star post. Thanks for answering VM! I dont understand all you said but the conclusions seem spot on IME.

The next question is, what map size gives you the most bang for your buck out of sushi? Huge is .25 per resource, while standard is .5. Is there any map out there that we can use that to make standard>huge? In other words, how many more seafoods are there on huge than standard to get the free population and does that overcome the maxscore limitations? I would certainly rather play a smaller map than huge so I wish WT was correct.

VirusMonster
May 22, 2008, 04:22 AM
Well open up your algebra books to understand the math going on... use windows calculator :) The x^y function in the windows calculator is used to take exponents.

-------------------

Now, let's answer your question on Sushi food bonus for different map sizes.

Let's say there is a map type where Sushi gives more food on standard than Immortal. How much more food do you think you will get on standard?

I would say not more than 5 :food:, because Huge map sizes have on high sea levels roughly 2x land mass compared to a standard map type, hence, roughly 2x more sushi resources. If you tweak the setttings and play a low sea level map with many small land masses, you will actually get more final sushi food bonus on Huge map sizes, because the total land mass for a Huge map could be potentially reach 4k, while the average standard map size maximum is 1k... 4 times more land will get roughly 4 times more sushi resources, even if less, I doubt it will be less than 2 times of standard map sushi resources.

Despite all this, say you are getting more sushi food on a standard map. How much more do you expect to get? 5 more food sounds reasonable to me. How much will this 5 extra food shorten your milking phase?

In my previous post, I had a total of 55 :food: in calculating city excess food generation at size 20. 20x2 :food: from farms + 15 :food: from sushi = 55 :food:

Say you are getting 5 more :food: for playing on standard and you get a total of 60 :food:

Well after you have spread your cities, your growth rate will be cut by roughy (1-(55/60))=91% for playing a huge map. How many turns will lose that for playing a huge map? In my earlier post, I assumed 100 turns would be enough to grow your empire to the happiness cap, well now on standard it should take you 90 turns to grow to the maximum limit.

These 10 turns are negligible compared to the huge 150-175 turns saving you get for playing a huge map and getting a higher population raw score.

Again, as discussed in earlier posts of this thread, if Sushi gives you 35-40 :food:, consider yourself lucky. Based on my earlier counting on Sushi resources on a Huge Archi map, from roughly 200 Sushi resources there, i would say 50 :food: is the maximum you can get from Sushi on any map type. The practical maximum will be lower than 50 :food:, because:

1. the more resources Sushi consumes the more maintenance you will pay, and there is some difference between 20 :food: vs. 40 :food: sushi.

2. You might not have access to all sushi resources immediately due to diplomatic reasons, lack of settlers, etc..

---------------------

Conclusion:

Even if smaller map types could give you a bit more (1-10) :food: for sushi, the resulting effect of growing your cities faster does not compansate the big scoring difference huge maps get for having higher population and land rawscores.

WastinTime
May 22, 2008, 10:12 AM
Good work VM, there are a lot of numbers there...and a lot of assumptions. The biggest one is that you assume a larger maxscore (a larger map) = larger rawscore. Not necessarily true.

the resulting effect of growing your cities faster does not compansate the big scoring difference huge maps get for having higher population and land rawscores

Let's be clear, a higher MAXscore for the larger size map will LOWER your final score if you get the exact same population, land, etc. Not the same percentages, the same raw numbers.

It's only the promise of higher RAWscores that could raise your score. So far, this promise (that a 2x larger map produces 2x larger population) is unfulfilled. KC had a population of 2755 on Huge. I had 4200 on Standard. Now I know I only got there because the easy difficulty, but that's not the point. The point is: a standard map could support a population of up to 5000 and this is plenty for a 2 million score. It doesn't matter that Huge could support a population of 10,000 or 20,000. You can't get that high in the turns you have to play the game.

Look at it this way: Let's say you wanted to beat my 600K score on chieftain. You've concluded that it can be done by playing a Huge/Chieftain map instead of the standard one I played. Now, an interesting calculation would be: What population would you need to beat 600K ? I'll take a rough estimate and say 7000. Do you really think you can get the same end date and grow your population that much higher?

Here's another way to look at it. Assuming you don't run out of space to grow on Standard or Large, there is nothing about Huge that will let you make settlers faster, or make corp execs faster, etc. Instead, Huge gets a slower tech pace AND a huge sushi penalty (I was not aware of that.)

I don't have a lot of numbers, but my gut is still telling me that Huge has no chance at the highest score. I think KC will prove that when he gets a chance to take his play strategy to other map sizes. I can't get up the desire to play it myself.

killercane
May 22, 2008, 02:09 PM
Good work VM, there are a lot of numbers there...and a lot of assumptions. The biggest one is that you assume a larger maxscore (a larger map) = larger rawscore. Not necessarily true.



Let's be clear, a higher MAXscore for the larger size map will LOWER your final score if you get the exact same population, land, etc. Not the same percentages, the same raw numbers.

It's only the promise of higher RAWscores that could raise your score. So far, this promise (that a 2x larger map produces 2x larger population) is unfulfilled. KC had a population of 2755 on Huge. I had 4200 on Standard. Now I know I only got there because the easy difficulty, but that's not the point. The point is: a standard map could support a population of up to 5000 and this is plenty for a 2 million score. It doesn't matter that Huge could support a population of 10,000 or 20,000. You can't get that high in the turns you have to play the game.

Look at it this way: Let's say you wanted to beat my 600K score on chieftain. You've concluded that it can be done by playing a Huge/Chieftain map instead of the standard one I played. Now, an interesting calculation would be: What population would you need to beat 600K ? I'll take a rough estimate and say 7000. Do you really think you can get the same end date and grow your population that much higher?

Here's another way to look at it. Assuming you don't run out of space to grow on Standard or Large, there is nothing about Huge that will let you make settlers faster, or make corp execs faster, etc. Instead, Huge gets a slower tech pace AND a huge sushi penalty (I was not aware of that.)

I don't have a lot of numbers, but my gut is still telling me that Huge has no chance at the highest score. I think KC will prove that when he gets a chance to take his play strategy to other map sizes. I can't get up the desire to play it myself.
Well maintenance is always the thing. On huge you will have less maintenance, and therefore can make settlers and executives more quickly. Once you get past point X, your tech pace is better than on a standard map.

Breaking it down into game turns, on marathon it takes roughly til turn 100 to produce a settler, and turn 125 to begin taking AI cities. This part goes quickly until you reach 0 research and have to build wealth. Huge will allow several more cities than standard and the building effects are then exponential. If we assume it takes 150-200 turns to take a new city from pop 1 to pop 25, the only way for standard to make 2 million is if your tech pace allows you biology and sushi considerably sooner on standard rather than huge.

But play some standard maps and prove me wrong! I havent yet played a late game on standard map but I suspect you can make 1.5 million, but I think 2 million will only be for huge maps.

One thing that has not been mentioned, does Maxscore take into account just 1 citizen per grassland or does it take into account biology's two citizens per grass?

WastinTime
May 22, 2008, 03:48 PM
I'm not sure why you two (KC and VM) seem so biased toward the conclusion that huge maps are better. Instead of trying to figure out which size is best, you both are trying to twist the numbers to make Huge sound better. Your analysis includes some crazy assumptions/conclusions

150-200 turns to grow from 1 to pop 25? That makes no sense. It's probably more like 50 turns or less. But more importantly, what is the point? How does that lead to your conclusion that standard maps are limited to 1.5 million.

Now (I think) you're claiming that on standard size maps you will run out of money and have to stop founding new cities. You say huge maps won't have this problem because of lower maint. I don't buy that. You'll have the same number of cites on standard or huge at turn 200, and at turn 300, etc.

It wouldn't help much for me to play. My play style would be so different it still wouldn't shed much light on the map size debate. It's best if KC does his same strategy on standard/Immortal.

VirusMonster
May 22, 2008, 04:06 PM
To KillerCane: max score only takes into accout pre-biology maximum possible sustainable population.


To WastinTime:

"Look at it this way: Let's say you wanted to beat my 600K score on chieftain. You've concluded that it can be done by playing a Huge/Chieftain map instead of the standard one I played. Now, an interesting calculation would be: What population would you need to beat 600K ? I'll take a rough estimate and say 7000. Do you really think you can get the same end date and grow your population that much higher?"

What you are missing is the rawscore/maxscore precentage for Huge does not have to be as high as a standard one for a huge map. For example, say a standard map has 1000 max score and a huge map has 3000 max score. And let's say you reach 3000 population on victory date for standard. For a huge map to score as high as your standard score, he does not have to reach 9000 population on the same victory date. He can get away with much less population to score just as high.

How much less? Say your turn for victory on Immortal is 600. Actually if we scale your 1520AD finish date on Epic to marathon speed, your end date would be like turn 650-700.. That is pretty late of a victory imho....

Anyway now let's turn back to your 3000 vs. 1000max score for huge and standard.

In the formula, the denominator would be:

For standard map size, 1000^(650/1500) = 20
For huge map size, 3000^(650/1500) = 32

Now the rawscores, for both mapsize will be divided by these values. Say on finish turn 650, standard map size has 3000 population score. The resulting value would be 3000/20=150. To reach same value on a huge map on 650 victory date, 150*32=4800 rawscore would be enough.

According to what you might be saying, you would expect a rawscore of 9000 to come even with the standard map score. But if you investigate the formula carefully, the denominator term is just an exponential term, whereas the nominator term is linear. Thus, you can get away with relatively much less rawscores on Hugemaps and still score just as high as standard.

In summary, to answer your question, yes I think if you can reach 3000 population on turn 650 for a standard size map, I can reach 4800 population at turn 650.

Even if I miss reaching 4800 population at turn 650. Due to the growth explosion after sushi and biology, when i reach 9000 population on turn 700, my score multiplier will be:

9000/(3000^(700/1500))=214

whereas your score multiplier for finish at turn 650 with 3000 popscore on a 1000maxpopscore standard map was: 3000/(1000^(650/1500))=150.

That is 214/150=1.42 higher popscore..... add the fact you get higher raw landscore... even if the remaining tech and wonder scores could be abit reduced for your 50 turn later finish date.... still a huge map finishing 50 turns later than a standard map would score higher....

How much higher? Roughly 1.2-1.5 times higher end score...

-----

Now, if what KillerCane saying about standard maps achieving 1.5M score is possible.... Then a huge map scoring 1.4 times higher would cross the 2.M barrier..

VirusMonster
May 22, 2008, 04:17 PM
I'm not sure why you two (KC and VM) seem so biased toward the conclusion that huge maps are better. Instead of trying to figure out which size is best, you both are trying to twist the numbers to make Huge sound better. Your analysis includes some crazy assumptions/conclusions

150-200 turns to grow from 1 to pop 25? That makes no sense. It's probably more like 50 turns or less. But more importantly, what is the point? How does that lead to your conclusion that standard maps are limited to 1.5 million.

Instead of calling my analysis crazy, I suggest you point out where you think there is a weak assumption. You play epic speed, and 50 turns might be good enough to achieve 25 population there. On Marathon, however, which should be the preferred playing speed for scoring high, 100-200 turns are just about right to reach 25 population. I had included a 3 part summary of why growing to 25 would take so long. First, you must build the settler, 2nd you must build the sushi executive, 3rd you must wait the city to grow..... How do you expect to grow to size 25 so quickly? If you see anything wrong, let me know.

In fact, if what you say is true even for marathon and growing size 25 takes less amount of time than 200 turns, then the better for a huge map to be able to support higher number of cities quicker.

---------------------------------

Now (I think) you're claiming that on standard size maps you will run out of money and have to stop founding new cities. You say huge maps won't have this problem because of lower maint. I don't buy that. You'll have the same number of cites on standard or huge at turn 200, and at turn 300, etc.

It wouldn't help much for me to play. My play style would be so different it still wouldn't shed much light on the map size debate. It's best if KC does his same strategy on standard/Immortal.

Alright, I have answered this question in my previous long reply. Even if you get same amount of cities for huge and standard on similar dates, being able to support a higher rawscore, despite finishing 50 turns later, would still score higher than a standard map score.

Say you had 100 cities at finish date on Standard, and say you want 300 cities on finish date on Huge. Then to build the settlers, to spread the sushi, and to grow to size 25 would, if you start building them earlier than your finish date, should not take more than 100-120 turns. You think you get lower score for finishing later on huge, but the resulting higher rawscore more than compensates the later finish, at least for 175 turns... look in the first long post where the 175 number comes from.

WastinTime
May 22, 2008, 04:23 PM
Your numbers look good. I did a similiar calculation in my head and used turn 750 instead of 650 (cus then it's just the square-root). So my 4200/7000 estimate is about the same as your 3000/4800. The problem is that you somehow think and extra 2000 population (give or take) is easy. And somehow the extra tiles on the huge map provide this miracle grow in spite of the other disadvantages of huge.

WastinTime
May 22, 2008, 04:40 PM
In fact, if what you say is true even for marathon and growing size 25 takes less amount of time than 200 turns, then the better for a huge map to be able to support higher number of cities quicker.

Sorry, that sounded a bit rude. When you plant a new city with sushi +30 food, it will grow almost every turn (even on marathon). It will be at size 15 in no time and then slow down a bit, but I don't see how any of that matters.

And even if it did matter, how does map size fit into that argument? It comes down to the point I made earlier. The map needs to be large enough so you don't run out of space. What if there was an "ultra huge" map that was 10 times bigger than huge? You would conclude that it is the best map for high scores, but it's not, because you can't fill that space in 700-ish turns.

WastinTime
May 22, 2008, 04:45 PM
Even if you get same amount of cities for huge and standard on similar dates...

Say you had 100 cities at finish date on Standard, and say you want 300 cities on finish date on Huge.

Ok, 1 more time. This is where you are getting lost. First you mention the case where you get the same # of cities on similar dates. Then you say that somehow you "want 300" cities on huge. Like they come free with the map. You still have to build them. I'm saying you can "want 300" on standard size too, just like you first mentioned: same # of cities at the similar dates.

Your score on standard will beat huge with those same 300 cities.

VirusMonster
May 22, 2008, 04:48 PM
I hope you understand that the extra 2000 population does not have to come at the finish date of the standard map score. It can come 50-100 turns later in the milking phase. You would still score more than a standard map would score.

More cities for huge map = higher number of settlers can be poprushed simultaneously.
More cities for huge map = higher number of executives can be poprushed simultaneously.
More cities for huge map = you get more gold to goldbuy the executives.
More cities for huge map = you get reduced maintenance costs and thus can support more cities at any date... The limiting factor to expansion in almost every civ4 game is not military, but rather economical. I am digging how much more % of cities you can sustain at any date on a huge map.

WastinTime
May 22, 2008, 05:10 PM
Ok, I'm done arguing. You are hell-bent on favoring huge maps. If you play a few games, maybe you'll see the light.

VirusMonster
May 22, 2008, 05:19 PM
WastinTime, I now fully understand your argument.

Let me put it this way. Since the rate of the milking phase score earnings are higher than the rate of score loss for finishing later, you would ideally want as big of a map as possible.

Your argument makes sense... say I had a 50 000 max score map (does not exist in current civIV game) and I finished at turn 1000 for reaching 150 000 popscore for such map. Then the score multiplier would be 110... slightly less than the 150 score multiplier for standard... Then we would have assumed such map should not be preferred for highest scores.

But as it happens, you can fill the extra citizens pretty quickly in the milking phase. The extra 100-200 cities can be built super quickly through poprushing and goldbuying. And growth can happen in less than 50 turns (ok, I assume you have a good sushi map) just as you claim with a +30 :food: sushi.

How much extra time can you spend in this milking phase and still score as high as a standard map score? My analysis says roughly despite finishing 175 turns later on huge. you would still get same popscore...

In fact, for a huge map of max popscore 50 000, finishing 175 turns later than 600th turn standard date score: 150 000 / (50 000 ^ (775/1500)) = 602. Compare this popscore multiplier with the 150 standard score multiplier for finishing at turn 600. If you finish on turn 800, then score multiplier would be 400sh... still better than the 150 standard score multiplier for finishing at turn 600. finish on turn 900, and the multiplier would be 220... still better than the 150 standard popscore multipler....

If you finish too late however, your tech, wonder, and scores start to diminish to a great extent. They compose half of your score, thus you can't stay so much in the milking phase, and your popscore multiplier has to be at least 2 times higher than the relative standard earlier finish date score. So for a 50 000 max popscore huge map, somewhere between 800 and 850th turn should be the ideal finish date. Well, that is roughly 150-250 turns later than your 600th turn finish date. So my 175 turn later finish was a reasonable approximation.

I have been very conservative in my calculations and did everything to favor standard map sizes, yet

1. the way the formula is set with the exponential term in the denominator
2. the way milking takes roughly 100-150 turns to reach max pop size for any map.
3. the way you can afford more cities at any date for larger map sizes due to reduced maintenance costs

clearly favor huge map sizes for highest scores.... Not by much, maybe 1.2-1.5 times more score... but nevertheless significant.

killercane
May 22, 2008, 05:23 PM
I'm not sure why you two (KC and VM) seem so biased toward the conclusion that huge maps are better. Instead of trying to figure out which size is best, you both are trying to twist the numbers to make Huge sound better. Your analysis includes some crazy assumptions/conclusions

150-200 turns to grow from 1 to pop 25? That makes no sense. It's probably more like 50 turns or less. But more importantly, what is the point? How does that lead to your conclusion that standard maps are limited to 1.5 million.

Now (I think) you're claiming that on standard size maps you will run out of money and have to stop founding new cities. You say huge maps won't have this problem because of lower maint. I don't buy that. You'll have the same number of cites on standard or huge at turn 200, and at turn 300, etc.

It wouldn't help much for me to play. My play style would be so different it still wouldn't shed much light on the map size debate. It's best if KC does his same strategy on standard/Immortal.
I have tried some standard maps and didnt think the score was progressing as quickly. Therefore I didnt continue through the milking phase. I just didnt get the feeling that I was going to be able to score as high compared to huge. So Im the empirical guy not the numbers guy.

Based on that I think 1.5 million is the goal on standard and that you cant make 2 million there but you can on huge.

You wont have the same number of cities at turns 200 and 300, as they cost more. I dont know how MUCH more but definitely more. I would guess 20%. I dont know where you are getting I love huge maps, I dislike them and would rather play standard (or smaller!).

Mapsize debate is just a subset of the how the heck to score 2 million points debate so if you develop a strategy and do so, i think we would all like to hear about it. Im still trying to figure some new things out, such as archer rush the first civ and then build settler & Immortals, which should give 3 cities early. Also, I have yet to figure out how to use the initial scientist GP; academy is looking like a worse and worse choice as science rate gets so low in the midgame.

VirusMonster
May 22, 2008, 05:35 PM
What you are saying is we should play a map with as low of a maxscore as possible....

That is a true statement..... I totally agree with you..

If on preferred finish date, we still have some population to grow, but cannot because we have to finish at certain date or we would score less than our standard score, then the excess possible population to grow is a lost score and we must pay for having chosing a map size so large that we could not fill on time.... The higher max popscore will cut down on score.

So your argument has a point....

I hope this point was answered in my previous reply... Obviously there is a limit on how late you can finish the larger map.

And if the goal is not achieved within these limitations, you would be better off playing a smaller map size....

But I can poprush 150 settlers in 3-4 turns. Executives can be built and poprushed rather quickly (30 turns at max?!). Executives and settlers can be settled simultaneously within 15 turns. 50 turns for population growth. Total milking phase lasts 100 turns.

So finishing later by 100 turns is still good for scoring, because not too much time is wasted in the milking phase. If we had spent 250-300 turns in the milking phase, then it could be said that standard size maps are the better choice.

VirusMonster
May 22, 2008, 06:05 PM
Mapsize debate is just a subset of the how the heck to score 2 million points debate so if you develop a strategy and do so, i think we would all like to hear about it. Im still trying to figure some new things out, such as archer rush the first civ and then build settler & Immortals, which should give 3 cities early. Also, I have yet to figure out how to use the initial scientist GP; academy is looking like a worse and worse choice as science rate gets so low in the midgame.

I already figured the strategy you talk about: Shaka Warrior rush... Capture the first capital when it only has 1 archer defender right after the first AI settler is sent out of the capital with 2 protecting archers. You need 5-6warriors to beat a single non-hill archer at 20% cultural defense.

I have also experimented with 1 temple priest and Oracle Gp generation to get an early shrine, but spreading this shrine religion requires too much focus on producing the missionaries, thus it might not be the best idea to get a priest instead of a scientists as first GP... I still think your early academy idea is best.

With Shaka warrior rush, and a 3 gold 1 gem capital, I have managed to get Bureaucracy at 1400BC on Immortal Huge. Rest was going really well, but I had not managed to capture as many cities as I would want, so I abandoned the game. With Bureaucracy at 1400BC and an academy, most of your research problems would be solved.... Currency will be discovered rather quickly with Bureaucracy.

Capturing cities will be more problematic with the Zulu UU, but you can just use the UU to pillage strategic resources, and the city captures can be done with your axeman or swordsman army.

killercane
May 23, 2008, 09:40 PM
Well it is only experimentation when I pop archery from a hut, there is an enemy close within 12 tiles, can watch the AI city from a corner spot, and can produce some archers before turn 50 (when the aforementioned settler comes out). So very game specific.

Using an aggressive civ to rush warriors is interesting but it is a short term benefit for a long term negative IMO (replacing a more long term trait with aggressive).

I am going to try to prove WT correct this weekend on a standard map. We will see how it goes.

killercane
May 25, 2008, 12:12 PM
I really dont think there is any way to make 2 million on Immortal WT. Attached is a save to the game I just played. There is still room for pop growth but Washington just built HG and I dont want to continue. The score is about 800k at turn 615 or so. I reloaded after letting a colony loose on another continent, as a result domination % went from 62 to 60 and I accidently won. So another lesson learned.

I think buying Redentor and all the late wonders is the way to go. First build in island cities is going to be a courthouse, with Sushi and no court you pay 120 gpt maintenance on the things. Sushi produced 40 food per turn and was maximally spread.

Standard size is much nicer for micromanaging. I might try another to see about going all the way to future tech.

VirusMonster
Jun 02, 2008, 01:54 PM
I don't understand why people would vote 1 star for any article unless it is total philosophical garbage. 12 voters and rating was 4.75, and now 13 voters and rating 4.46.. someone voted a 1 star :P gotta respect his hidden identity...

VirusMonster
Jun 02, 2008, 02:35 PM
I really dont think there is any way to make 2 million on Immortal WT. Attached is a save to the game I just played. There is still room for pop growth but Washington just built HG and I dont want to continue. The score is about 800k at turn 615 or so. I reloaded after letting a colony loose on another continent, as a result domination % went from 62 to 60 and I accidently won. So another lesson learned.

I think buying Redentor and all the late wonders is the way to go. First build in island cities is going to be a courthouse, with Sushi and no court you pay 120 gpt maintenance on the things. Sushi produced 40 food per turn and was maximally spread.

Standard size is much nicer for micromanaging. I might try another to see about going all the way to future tech.

How large can you grow your cities through having unhappy citizens? 25? 30? 35? 40? WastingTime suggests one can grow his empire on standard just like he would on large, but I don't understand how a city can grow to size 40 in 25-50 turns despite happiness and health caps being around 25 near end game on the 600th turn?

Say you could squeeze 100 cities on a standard map.. Do you think it is possible to grow them to size 35-40 for a total of 3500-4000 population score around turn 600-700?

killercane
Jun 03, 2008, 06:05 AM
How large can you grow your cities through having unhappy citizens? 25? 30? 35? 40? WastingTime suggests one can grow his empire on standard just like he would on large, but I don't understand how a city can grow to size 40 in 25-50 turns despite happiness and health caps being around 25 near end game on the 600th turn?

Say you could squeeze 100 cities on a standard map.. Do you think it is possible to grow them to size 35-40 for a total of 3500-4000 population score around turn 600-700?
About 30-35 with Broadway, Rock N Roll, etc at 0 culture slider. You can always squeeze more happiness out of Nationhood barracks, Free Religion, and the slider. Supermarkets (refrigeration) add health so that should take you about to size 30ish health wise as well.

I think no vassal states is the way to go as well to prevent the AI from making colonies and taking the domination limit lower. Its a frustrating way to win... or lose if you want to look at it that way.

WastinTime
Jun 03, 2008, 09:45 AM
I don't understand the concern for happiness. It has no effect on city size. Once you are working all your food tiles any population over that (happy or not) will not make any food. All you really lose is research, or whatever the specialist would do if it was happy.

I always have vassals off by default. I don't find many uses for them in my games.

VirusMonster
Jun 04, 2008, 01:42 AM
Alright if happiness is not the problem, how much can you grow your cities?

If cities will be built within 3 squares of each other to maximize number of cities on the map without crossing the domination limit, then there won't be 20 farms per city, but rather 10-12.

Say Sushi is giving +30 :food:. Add food of 12 farms = 12 x 4 = 48. Total food generated is 78. You can sustain 39 citizens with that much food. 39 x 100 cities squeezed on a standard map = 3900 population points. Pretty damn good for a standard map.

On the other hand, growing to such city size will get gradually slower. At size 38, food surplus would be +2 :food:, at size 37 +4 :food:, at size 36 +6 :food:. Growing from size 38 to 39 with a granary would take like (33 + (3 x 38))/2 = 76 :food: (I have never had such a large city during my civ4 career)

Then with +2 :food: surplus, it would take 38 turns to grow to size 39. Not worth the milking.

So while you can theoretically grow to size 39, 38 turns spent from growing from size 38 to 39 is such a long period in late game that the score benefit you get from having the extra population will be insignificant.

There must be a reasonable point to grow cities during the milking phase. I think it can be found through backwards growing the cities. Let me do the basic approximation quickly and you can tell me where I've missed anything.

Growing from size 37 to 38 with a granary, this time with a +4 :food: surplus:

((33 + (3 x 37))/2)/4= 18 turns

Pretty good compared to the not practical 38 turns to grow from 38 to 39.

Growing from size 36 to 37 with a granay, this time with a +6 :food: surplus,

((33 + (3 x 36))/2)/6= 12 turns

Growing from size 35 to 36 with a granay, this time with a +8 :food: surplus,

((33 + (3 x 35))/2)/8= 9 turns

Growing from size 34 to 35 with a granay, this time with a +10 :food: surplus,

((33 + (3 x 34))/2)/10= 7 turns

Growing from size 33 to 34 with a granay, this time with a +12 :food: surplus,

((33 + (3 x 33))/2)/12= 6 turns

Growing from size 32 to 33 with a granay, this time with a +14 :food: surplus,

((33 + (3 x 32))/2)/14= 5 turns

Growing from size 31 to 32 with a granay, this time with a +16 :food: surplus,

((33 + (3 x 31))/2)/16= 4 turns

Growing from size 30 to 31 with a granay, this time with a +18 :food: surplus,

((33 + (3 x 30))/2)/18= 4 turns (again 4 turns)

Growing from size 29 to 30 with a granay, this time with a +20 :food: surplus,

((33 + (3 x 20))/2)/40= 3 turns

Before I stop, here is the calculation for a city size 20 and 25.

Growing from size 24 to 25 with a granary, this time with a +30 :food: surplus,

((33 + (3 x 25))/2)/30= 1.8 turns ( roughly 2 turns)

Growing from size 20 to 21 with a granary, this time with a +40 :food: surplus,

((33 + (3 x 20))/2)/40= 1.16 turns (roughly growth every turn, just like WastinTime suggested)

--------------------------------------------------------------------

Summary:

For an overcrowded map, where most cities don't have more than 12 farms to work and sushi giving +30 :food:, it takes the following number of turns for cities to grow.

Cities until 20 population grow almost 1 citizen every turn.
Cities size 20-25 growth every 1-2 turns.
Cities size 25-29 growth every 2-3 turns.
Cities size 30-31 growth every 4 turns.
Cities size 32 growth every 5 turns.
Cities size 33 grow every 6 turns.
Cities size 34 grow every 7 turns.
Cities size 35 grow every 9 turns.
Cities size 36 grow every 12 turns.
Cities size 37 grow every 18 turns.

So total growth phase until size 20 lasts 20 turns. Until size 25 lasts 27 turns. Until size 30 lasts 40. Until size 35 lasts 71 turns. Until size 39 lasts (71+12+18+38)= 139 turns

Waiting your cities to grow 139 turns to max size is pretty long, so better is aiming for size 30 (40 turns) or 35 (71 turns)....

So ideal city size at the end of the milking phase is somewhere between 30-35....

---------------

Btw, have I forgotten to take the effects of unhealthiness into account? Does each unhealthy citizen also reduce the food surplus by -2 :food: in addition to the regular 2:food: each citizen consumes?

I haven't played civIV long time so I am starting to forget the game mechanics, someone refresh my memory :)

-------------------

Anyway, I will assume the above calculation are correct. For a standard size map with 1000 max pop score, 100 cities * 30-35 citizens each = 3500 population score. Then a standard map indeed could have 3.0-3.5x max population score within 40 turns.

----------------

Until how long later could a larger map size achieve same 3.5-4x max population score and still score as high as the standard map size with same population to max pop score ratio?! Well settler and sushi executive production is the limiting factor. Once sushi is spread to all cities it is a matter of only 40 turns to reach practical milking limitation on any map size.

Producing 300 settlers or producing 100 settlers... Producing 300 executives or producing 100 executives? And would it take faster or slower to dominate a larger map size? I would assume slower, but again you might be getting so many great generals and experienced army that could make your advances even faster.

Those are the questions you need to answer to determine which map size would score higher, because growing your cities to size 30-35 would take similar amount of time on any map size.

Poprushing a 300 :hammers: settler would take 3-4 citizens. Poprushing a 200 :hammers: executive would take 2-3 citizens. So poprushing both would take. 500/90 = 5-6 citizens.

Say you had 50 cities with sushi before the milking phase all at size 15. In 3 turns, you will have 100 sushi cities.

Then you must wait some turns for the new cities to grow so you can poprush them. The new cities need granaries to be built as well for faster growth, but let's say you won't worry about unhappiness for a long time, so you will poprush the granary, the settler, and the executive within 30 turns.

Growth happens almost every turn for granary containing cities lower than size 20, so if you wait 15 turns you will have again a city size 15. Also you poprushed the granary as soon as the city size was 3-4, took you 1 turn only.

Now, 100 cities poprush, 100 settlers and 100 executives. You end up with 200 sushi cities. Then, you wait another 5-6 turns for cities to grow their population back. Poprush another set of 100 settlers and executives. You got yourself a 300 city empire.

300 city empire all with sushi took 3+1+15+5=24 turns of poprushing.

---------------------------------

Conclusion:

Total milking phase for a large map would last 24 turns to poprush the settlers and the executives plus 40 turns to grow all cities to size 30-35.

24+40 = 64 turns to stuff a large map size with 300 cities and 9k-10k citizens without crossing the domination limit.

On a standard map size producing 50 executives and 50 settlers would take only a few turns, but again those cities would have to wait 40 turns to grow to size 30-35. Thus total milking phase for a standard map would be roughly 45.

64 turns vs. 45 turns... the difference between milking a standard size map and a large map size with 3x more pop score is only 20 turns.

The advantage of playing a smaller map size is, well, you would have spent less time producing those settlers and executives. And you would have spent less time conquering the world, assuming similar domination speed for both map sizes. How many turns would that make in total? 20-30 turns? 50? or even 100? As I discussed in earlier posts, those number of turns are not good enough to compansate the higher score you would get with the same popscore/maxscore ratio for a larger map.

I know both of you prefer smaller map sizes, but showing the patience to play a large map would pay handsomely in the final score. That is the way the game is.

arcangel7
Jun 17, 2008, 10:17 AM
Many thanks to VM and all who give their help in this guide. I found it very useful to achieve my first Immortal win on Huge/Marathon as Julius with 18 competitors. I followed a similar path and made a domination vistory in 1532 for a score of around 1/4 mill. I was unlucky that the pyramids was built on the other side of the map (a downside to worker stealing on your neighbours perhaps ?) so I couldnt capture and I didnt get techs via trade :(. I still enjoyed the victory and will try as Cyrus b4 I tackle deity.
I had 2 scientists per city and attacked anyone near who built a wonder and although I got to guilds 1st (for knights), I went for race to Lib and won although this meant my opponents got to cuirassor first. Still I got rifle first then had Rifle/Treb/Knight V Curr/Grenadier and then got cavalry soon after and then infantry and completed the win b4 I built an oil well which is new for me :)).
I guess built buildings a few times when I should have built units, but I wasnt in for the big score, just my first win ;)

tycoonist
Jun 17, 2008, 02:19 PM
this had turned into a great article with a lot of in depth analysis and maths. i would just like to thank VM, Wastintime and killercane for this effort.

WastinTime
Jun 19, 2008, 12:54 AM
I don't think there's much time left to get that 2 million score. The new patch 3.17 will hose this strategy a bit. Corp executive limit of 5 (like building missionaries.)

killercane
Jun 20, 2008, 04:32 PM
Well I have been away but I have a game in the milking stages with no real domination limit problems that I foresee. Hopefully I can get it done before upgrading. This new patch was sudden eh?

VirusMonster
Jun 22, 2008, 05:46 AM
Hard to understand why they are putting a limit on missionaries and executives, and again hard to understand why this number does not scale with map size. They really should not be making such changes without asking us first :-)

killercane
Jun 27, 2008, 07:38 PM
I think I have figured it out. Playing standard deity is a bit easier and the research is so much faster maybe 2 million is possible. I think Sushi can be incorporated several turns before turn 400 and thus you have 100+ turns for milking. Now, this is quite a bit better than turn 700 for huge.

The way this works is to eliminate everyone except for 1 civ you get to friendly. Events help with this to cancel out the -you declared on our friend penalty. The friendly AI will build settlers and workboats for your islands AND garner enough land that you wont have to worry about domination % until you declare on them.

THey also will research things like Constitution and Corporation for you, as well as some other things along the way. I messed up a bit in not getting Democracy in the game I just played (emancipation penalty!) but it really looks like a fresh new approach that is much simpler to implement; the AI was giving me +10 food in resources via Sushi and immediately at that. The test game was messed up from the get go by having no city razing on but really the approach looks good from both a time standpoint and ease of play standpoint.

I would also recommend having 2 AIs at the midgame so that both will settle several islands on Big and Small map.

Of course, if you can solve the research bottleneck on huge then that would be better overall still but this approach might yield 1.5 million. THe last huge Immortal I played was only 1.15. So the gameplan is now to hit 1.5 on deity standard and figure out a way to translate higher research into 2 million after looking at the way the game progresses.

killercane
Jul 05, 2008, 09:53 AM
Finished another Inca game over the past month on Deity Standard. Quite an interesting game. I ended up with 2 friendly opponents left, and Sushi incorporated about turn 400 again. Unfortunately my land area was only 50% for the most part. So how to win? Obviously AP was looking good. So I wait for the AP vote in order to win when the score looked about right. Pop rockets up, I hit 1.25 million, but dont get an opportune voting time. 2 votes came too early, and the rest of the votes ended up being to assign me more cities from the AI; between AP reassignments and culture flip, I must have added a dozen cities to the empire.

So no AP vote happens, but I have an engineer for UN, which I researched until 1 turn left to wait for AP vote. So finish MM, finish UN. But as I check my score it has fallen so much from messing around that it is now below 1.1 million. The AI is researching well at this time, so I look at HOF charts... it looks like I can get space race done if I switch to SP rather than use the units on strike free market approach I had been under. Long story short, I lose to the AI in 1618 via space race, despite us having launched at the same time (I was minus 1 casing, guess that is the tiebreaker?).

So... the friendly Deity Standard approach seems to work well in conjunction with going on strike. You just have to have everything in place before you settle all the executives at the same time.

r_rolo1
Jul 05, 2008, 02:15 PM
Are you sure that you and the AI launched in the same turn, killercane? There are no ties in Civ IV ( due to the turn-based nature of the game ). Or the AI launched before you ( hence 1 turn earlier, given that the human is always the first to play in a turn, unless stated otherwise in the beginning of the game ) or your ship exploded ( you should receive a warning of that if you're using 3.17 )

killercane
Jul 05, 2008, 11:59 PM
Are you sure that you and the AI launched in the same turn, killercane? There are no ties in Civ IV ( due to the turn-based nature of the game ). Or the AI launched before you ( hence 1 turn earlier, given that the human is always the first to play in a turn, unless stated otherwise in the beginning of the game ) or your ship exploded ( you should receive a warning of that if you're using 3.17 )
Positive. I had thought during the game that he launched the turn after me. I was fully expecting a victory message, as it kept counting down to 1 turn to victory and then Washington wins a Space Victory! As I looked back through the turn log it showed we launched at the same time (2 engines, all thrusters, just minus the casing). Do you not get a message in 3.13 if your ship fails? That would be odd.

At least I know a bit about space colony wins now; it was an interesting game. Really, It was pretty instructive to look at Sushi and its relationship to culture and space victories.

r_rolo1
Jul 06, 2008, 02:02 AM
Do you not get a message in 3.13 if your ship fails? That would be odd.Yup..... only 3.17 introduced a message informing on SS failure. Before it you need to guess if the ship get there or not :(

But it looks to be a interesting game , in spite of the loss. And that is ultimately what matters ;)

killercane
Jul 06, 2008, 08:06 AM
Yup..... only 3.17 introduced a message informing on SS failure. Before it you need to guess if the ship get there or not :(

But it looks to be a interesting game , in spite of the loss. And that is ultimately what matters ;)
Well it would be a lot nicer if someone played some comparison games :rolleyes:.

r_rolo1
Jul 06, 2008, 08:10 AM
Not me.... I'm a standard map player that has really some dificulties of even winning in Immortal, even more milking the cow ;)

kivanc
Aug 20, 2008, 03:30 AM
Not me.... I'm a standard map player that has really some dificulties of even winning in Immortal, even more milking the cow ;)

i've read in a few more posts as well that standart maps are harder but i have to disagree. my following comments are for difficulties>=emperor

in huge maps you have many enemies. so let's say you have killed 4 AIs before 1AD. as standart, you will still have 6 more AIs. and this makes 4/5 AIs (nearly all AIs except AGR ones) with "-4 you declared war on our friend"

i can say you will be in war against 2/3 AIs at the same time also after 1AD.
but of course, these assumptions are for fractal/pangaea maps. for continents map, OK it is not a big problem. you early rush 2/3 AIs and then you are alone peacefully in the continent. and no AI in otehr continent will know what you slaughter you made in history :D
so continents type maps are great for DARIUS.

if you need a real challenge, then play in highlands map with CYRUS :) it's real fun!


and now my comments for JULIUS:

he is a beast. one of the best leaders in BTS.
A) org/imp has a great synergy.

B) UU is good but not best. because:
i) it's expensive (100: twice IMMORTALS)
ii) 8 instead of 6 and against every type of unit, not melee and archer ok, i mean this is better than a +%33 against melee. but instead of 8/6 extra, a free CR1 would be great.

C)+25% GP rate forum is good but doesn't have a big effect. as you will be having most GPs from cities with NP or NE, that it will only make 1/2 GPmore. most cities won't even have even 1 GP as GP points increase for each new one born.

mystyfly
Aug 22, 2008, 09:40 AM
and now my comments for JULIUS:

he is a beast. one of the best leaders in BTS.
A) org/imp has a great synergy.

B) UU is good but not best. because:
i) it's expensive (100: twice IMMORTALS)
ii) 8 instead of 6 and against every type of unit, not melee and archer ok, i mean this is better than a +%33 against melee. but instead of 8/6 extra, a free CR1 would be great.

C)+25% GP rate forum is good but doesn't have a big effect. as you will be having most GPs from cities with NP or NE, that it will only make 1/2 GPmore. most cities won't even have even 1 GP as GP points increase for each new one born.
Julius is a mediocre leader imo but Julius of Rome is overpowered.

I also don't want to start another discussion about praets (there is for example virusmonsters thread about sword-UUs) but praets are hard to compare to immortals. Immortals might be better than the average defender than praets are and better when looking at cost/yeild; but praets have other advantage like longer livespan (and on marathon it IS very long) and no real counter around.

kivanc
Aug 25, 2008, 02:27 AM
Julius is a mediocre leader imo but Julius of Rome is overpowered.

I also don't want to start another discussion about praets (there is for example virusmonsters thread about sword-UUs) but praets are hard to compare to immortals. Immortals might be better than the average defender than praets are and better when looking at cost/yeild; but praets have other advantage like longer livespan (and on marathon it IS very long) and no real counter around.

ok, you're right. let's not get out of topic, though i like praets and immortals both.
and julius is really good and as i don't use unrestr leaderds much, i judge leaders together with the civ.

anyway, the best success about all what you did in your game is IMO being able to build the wonders and required infrastr while being a warmonger.

knowing when to cease fire and attack again is very important.
knowing which city to capture and which one to raze is also important.
it seems you made the right decisions most of the time in that game.

generally a peaceful time should come after each big war because you are expanding and costs increase, and you are getting behind on tech during war. but as you are expanding and raising cash much thru capturing, you have a good tolerance for -30, -50 GPT for faster teching. and as land is power, you have a very good advantage with larger land afetr war. shortly, period after warring is very important.
beelining tech is also important, depending on the leader you can judge.

killercane
Aug 30, 2008, 07:40 AM
I have played some Rainforest maps and they are nice if you can get enough workers built to clear the jungle. The pop limit seems to be around 3300-3400 for Huge, and the starting locations are amazing in terms of food. These are on your base Rainforest, but I discovered that if you play a non rainforest map where you can assign sealevel to high, and then switch to Rainforest, the results are large lakes that can have seafood in them for the health bonus (health is a problem on RF). Some of these are the problematic cant-build-lighthouses kind however.

Rainforest is nice in that you can use cereal mills rather than sushi, and dont have to worry about the culture of sushi putting you over domination. The 120+ rice on the map generate the necessary food. Your great people farms can reach 400+ GPPs per turn relatively early with such high food areas.

I really just wanted to edge WT's Chieftain score so I quit at 604,900, but it seems I left 50-100K points on the board at 2500 points per turn and population left to be had.

WastinTime
Aug 30, 2008, 08:37 AM
I really just wanted to edge WT's Chieftain score so I quit at 604,900, but it seems I left 50-100K points on the board at 2500 points per turn and population left to be had.

I'm glad to get some sort of comparison, but keep in mind, I had never played a score game. Never used corporations, Never rushed to get Sushi early, didn't use slavery much if at all, I still don't understand the big deal with Cristo Redentor, etc. I was not very good at the many tactics used on this game. So if you're goal was to show that Huge is better than Standard size maps, you would need to play standard yourself. I'm betting you'd get 800K - 1 million on standard or large. However, if you goal was just to kick my butt, then Congrats! nice work! :king:

killercane
Aug 30, 2008, 10:05 AM
I'm glad to get some sort of comparison, but keep in mind, I had never played a score game. Never used corporations, Never rushed to get Sushi early, didn't use slavery much if at all, I still don't understand the big deal with Cristo Redentor, etc. I was not very good at the many tactics used on this game. So if you're goal was to show that Huge is better than Standard size maps, you would need to play standard yourself. I'm betting you'd get 800K - 1 million on standard or large. However, if you goal was just to kick my butt, then Congrats! nice work! :king:
No I dont think it goes to Standard/Large argument, or kicking your butt; it took a week to get up to play the last 20 turns or so. I tried the gauntlet at first and since it wasnt feasible to use rainforest at normal, I thought I would try marathon and your game gave a good goal. I didnt count on getting so burned out with moving workers and dealing with repositioning cities and the whole thing slowing down. So I cranked up culture to 50% once I hit 600k and just slid by you.

Most of the AIs had just adopted Emancipation so that to keep a city happy I needed 57 happiness sources. Switching out of caste cost many thousands of gold and brought the economy to a halt, and switching back I had to switch citizens to merchants in every city, and still I didnt follow through on that. Im too lazy or burned out on the score thing to actually do it anymore but who knows that may change.

kivanc
Sep 01, 2008, 12:15 AM
I have played some Rainforest maps and they are nice if you can get enough workers built to clear the jungle. The pop limit seems to be around 3300-3400 for Huge, and the starting locations are amazing in terms of food. These are on your base Rainforest, but I discovered that if you play a non rainforest map where you can assign sealevel to high, and then switch to Rainforest, the results are large lakes that can have seafood in them for the health bonus (health is a problem on RF). Some of these are the problematic cant-build-lighthouses kind however.


i also noticed that on a highlands map.
on such map types, climate cannot be assigned. so it seems to take your previous selection of map type.
i started with a tectonics/rocky map and then went into highlands, i started near a large group of hills
and then started on a pangaea/temperate map and then returned to highlands, and then i started near a large flatland. hmm strange.
it seems not fair bec i noticed that HC started on a flatland while i started on a rocky land in the first try.
i thought i would be rocky and thus nearly equal positions for all but it seems not.
i like rocky climate but only when all AIs start on rocky climates as well.
i think the climate is only for your starting position, not AIs, right?
but however, the guys who are experienced on map types say if you select tropical climate, then the map will have more jungles.
i really am not sure. if it's for your starting locations or for the whole map.

oyzar
Sep 01, 2008, 12:18 AM
Highlands doesn't read the sealevel and temperature parameters at all.. It does generate some flatland, just not a whole lot...

kivanc
Sep 11, 2008, 08:29 AM
Highlands doesn't read the sealevel and temperature parameters at all.. It does generate some flatland, just not a whole lot...

sea level: well, first of all, you are able to select the sea level in highlands map. if you cannot see any selection about it then just click esc and go into map type again; that time you will see.
the game rememebrs some selections and doesn't ask much except world size (when you go thru play now)

climate: and about climate i believe what i wrote in previous post is true. but still i have to try a bit more to be sure.

yes, there isn't any selection for climate in highlands map. but the script seems to remember and selects (as default and cannot be changed) the climate type you have selected in your previous start on another map type, which allows climate selection.

as it's pale grey and there's no permission for selection of climate, a person cannot be sure if it will generate a world like it knows as default or if it will generate the world according to the climatre type you have previously selected.

i tried a few times to generate different map types with different climates and then go back to highlands and generate, each seemed to have different results. that way seemed to allow me to select the climate type, like a trick.

but as i said, i have to test it again. as killercane just told a similar thing to my own experience, i thought it was not a coinsidence. i think, it reads climate selection that way.

killercane
Oct 28, 2008, 04:07 PM
sea level: well, first of all, you are able to select the sea level in highlands map. if you cannot see any selection about it then just click esc and go into map type again; that time you will see.
the game rememebrs some selections and doesn't ask much except world size (when you go thru play now)

climate: and about climate i believe what i wrote in previous post is true. but still i have to try a bit more to be sure.

yes, there isn't any selection for climate in highlands map. but the script seems to remember and selects (as default and cannot be changed) the climate type you have selected in your previous start on another map type, which allows climate selection.

as it's pale grey and there's no permission for selection of climate, a person cannot be sure if it will generate a world like it knows as default or if it will generate the world according to the climatre type you have previously selected.

i tried a few times to generate different map types with different climates and then go back to highlands and generate, each seemed to have different results. that way seemed to allow me to select the climate type, like a trick.

but as i said, i have to test it again. as killercane just told a similar thing to my own experience, i thought it was not a coinsidence. i think, it reads climate selection that way.
Any interesting variations you have found on the climate strategy?

Icewise
Oct 29, 2008, 02:17 PM
Nice thread!

oyzar
Dec 27, 2008, 02:00 PM
I used much of this in my attempt at G-Major 40. However it is funny to note that even with my clearly suboptimal playing and luck i managed to beat the OP game :p by 50%. Judging by my game it certainly should be possible to get 2M if you milk sushi and the new colony bug in the hof mod... Despite the claims that darius rocks, playing julius is just sooo much easier...

Jrrd Tzu
May 15, 2009, 03:54 PM
...this seems pretty bad*ss. Thanks for all the info.

WastinTime
Dec 08, 2009, 08:29 PM
This discussion is a bit dusty, but I thought I'd add my update. There was a good debate going on whether Huge maps were the best size for high score games. Seems like the obvious choice, more land = more cities. I was strongly against this assumption, but didn't think I had the patience for high score quests to prove my theory.

I had a 2.1 million score game on Tiny. Tiny was clearly too small for the biggest scores.

My 3.17 million score was on Small. I still feel like I ran out of space a bit.

Standard size may be the sweet spot. It depends on if the slower tech rate or the likely later finish date (because you have to build many more cities) hurts your score too much. Anyone up for the challenge?

T-hawk
Nov 05, 2011, 08:19 PM
Bumping an old thread but I have a significant new contribution.

I've analyzed and reverse-engineered the victory score normalization formula. VirusMonster did this up earlier in this thread (http://forums.civfanatics.com/showthread.php?p=6552908#post6552908), but I've gotten it to come out to match in-game values exactly. Here's my article with detailed explanations of the formula, and a calculator.

http://www.dos486.com/civ4/index/score.shtml

http://www.dos486.com/civ4/index/calc.shtml

WastinTime
Dec 27, 2011, 10:56 AM
I like your article. I had to read it several times to grasp it. It is a good analysis of the score formula. However, I believe you have a faulty conclusion.

You claim "bigger maps are better for victory score". And then go on to explain how the LAND component of score is easier to maximize on bigger maps. While that's true, we all know Population is the key to score, not land. I still think Huge is too big for the highest scores.

T-hawk
Dec 27, 2011, 11:37 PM
I used land to illustrate the formula, because it's easier to see what's going on around that concept of "magnification" from a base value of 21 land tiles rather than from a base value of 1 population. Population is of course the heavyweight in practice, because of that tiny base, but I found it a little harder to wrap your head around at first. Maybe I could have connected those dots together better.

You are right that the most score potential is created not by bigger map size but by higher possible and actual population. But of course there's a strong correlation between map size and possible/actual population.

"Bigger maps are better" is a general conclusion in theory, and it's true that the potential is always higher for a bigger map on the same script. Whether the actual score comes out higher -- whether the higher possible magnification outweighs the reduced exponential bonus for taking longer -- is beyond the scope of my discussion there. I'm hoping the calculator can serve as a step in that direction, so you can play with the numbers to see what a large or huge Big and Small map with Sushi might do.

ZPV
Dec 29, 2011, 07:40 AM
You are right that the most score potential is created not by bigger map size but by higher possible and actual population. But of course there's a strong correlation between map size and possible/actual population.

"Bigger maps are better" is a general conclusion in theory, and it's true that the potential is always higher for a bigger map on the same script. Whether the actual score comes out higher -- whether the higher possible magnification outweighs the reduced exponential bonus for taking longer -- is beyond the scope of my discussion there. I'm hoping the calculator can serve as a step in that direction, so you can play with the numbers to see what a large or huge Big and Small map with Sushi might do.

There's another big factor in working out how much potential population score you can get on a map: how many Sushi (or Cereal Mills if you prefer) resources there are on the map.

In particular, the corporation yield modifier scales roughly with the number of tiles, but the number of seafood resources doesn't - those will scale with the length of the coast, (proportional to the square root of the number of tiles).
This means your cities can get more food in smaller map sizes.

(The "expected" population, and the number of cities you can squeeze in, also scale with the number of tiles.)

This competes against the exponential magnification, (which does favour larger maps), which makes it quite possible that something other than Huge gives the best scores.

Luigiii2
Aug 25, 2012, 11:05 PM
nice thread! Now if only I lived long enough to use some of the corporation tactics you guys are talking about...

WastinTime
Aug 26, 2012, 04:53 PM
nice thread! Now if only I lived long enough to use some of the corporation tactics you guys are talking about...

Yea, this thread is a good start, but we're an order of magnitude beyond this. Instead of 500,000, the target is 5,000,000 points here (http://forums.civfanatics.com/showthread.php?t=468582).