The Height of Level 9 - Songhai, Asia map, mod Marathon, mod difficulty

[maltz]

Spoiler :

075. Horseriding research is done and I burned my Great Scientist for Chivalry for serious ass kicking! But... there is no Chivalry to choose from the list!

I then realized the entire game is a huge screw-up since I thought Chivalry is as easy as Steel - NOOOO! I need Civil Service and Currency first. How am I going to survive with 4 warriors ~ I should have went down Iron Working and go for Steel instead. And I definitely don't need to pick animal resources and end up in the center of conflict...

And that just shows how meaningless Songhai is in a level-9 game. One really needs to have some powerful early military units (Persia and Greece comes to mind). And how many Barbarian camps I busted myself? 0. The AIs attacked me nonstop and I had no chance to even go out to see any camp myself.

I decide to leave the game here since there are too many mistakes. And I need to rethink my level-9 mod. Simply extrapolating AI's advantages is reasonable, but that pretty much force people pick only a few Civs, not others. I want to make a level-9 mod that people can still win with their favorable Civs. Not to be battered around in Ancient Era, seeing 10+ units flushing to their home. I am lucky to have survived this far.

So back to drawing board!

 

[maltz]

Spoiler :

Here comes drawing board. I wish to explain my design rationale for the next version. Now I already know my central theme - I want to make a mod difficult, but not sadistically intrusive on the players especially in early game.

The above picture shows a very simple concept. If we let a Civilization grow, it will take on this kind of growth curve when nutrition (happiness, etc.) is unlimited. Simple.

 

[maltz]

Spoiler :

Now we can imagine a strong and a weak civilization. Maybe the same player plays on two different Civs, one is simply better than the other. Maybe the same Civ is controlled once by a veteran and the other by a rookie. So we will have the same starting point, but different growth rate. Simple.

 

[maltz]

Spoiler :

Here comes what we have in the game - say the Deity setting. The AI has a huge initial advantage, so they start higher on the Y axis. But the AI has a poor growth rate compared to a human veteran, who usually manages to outgrow the AI at Turn 120-150, more or less. The point where the human and AI's powers cross is the tipping point. Before this point, the game is exciting. Beyond this point, the game is boring. You are already the front runner and you run faster than your opponent anyway. So what's to puruse?

Of course - the track record. That's why so many people submit their games to Hall of Fame, just to show that they can do it 1 turn better. I find this very tiresome (and I am not that good anyway). I play game for excitement, not to make my alias famous. And most of the time I feel good about myself - not to suggest otherwise for Hall of Famers.

The developer's job for veteran players is to delay the tipping point. So the player is constantly behind and feels the pressure, but still have the hope to win.

 

[maltz]

Spoiler :

There are two strategies for delaying the tipping point. The first one is very easy - just raise the AI's initial bonus. For example, instead of starting with 2 cities, level 9 AI will start with 3 cities. If that's not enough, they will start with 4 cities. The blue line shifts vertically upwards until the human can barely beat them just before the end of game.

 

[maltz]

Spoiler :

The second strategy is also simple. Instead of adjusting the initial bonus, now we adjust the growth rate of AI. As the AI grow faster, they will be harder to catch up - all the way to the end of the game.

 

[maltz]

Spoiler :

Personally I like the second approach better than the first. Why? Because the second approach creates an illusion of fairness. In fact, one can completely adapt the second approach, and abandon the first one. We can make the AIs' growth so mimicking the human veteran, that they do not need any starting bonus at all!

The human will feel that they are playing in multi-player. Everyone starts with a warrior and a settler. But somehow, the AIs do very well as time goes on, just like some human players are better than others on multiplayer! At turn 0, the game looks like a Prince level game. 30 turns later, it looks like King. 30 turns later, it looks like Emperor. After Turn 120-150 (the initial turning point for Deity), the game goes beyond. Level 9, level 10... all the way.

Now isn't this a fun thing to try out? So I am going to start a game with this design in mind.

 

[maltz]

Spoiler :

000. The game will be played with Persia on the official Mesopotamia DLC map. Still on the same modded Marathon. No ancient ruins. All other settings default.

I used Persia on this particular map once before, but I dropped out after seeing my Rifleman butchering warriors and archers. So here comes a revisit of this scenario with the new level 9 design. I am filling in a few local Civs: Arabia, Egypt, Greece, and Ottoman, and leave the rest. I didn't buy Babylon, so that is not on the list.

 

[maltz]

Spoiler :

001. There are two "golden" spots on the Mesopotamia map. One is of course the Mesopotamia, along the Tigris–Euphrates river system. The other is Egypt, with Nile. Rivers brings riches and population. But these two areas suffer from a great shortage of hammers. All other starting spot suffers the opposite problem - no river, no gold, no population!

I think the first option is better. So I have to pick a map with one of the rivers.

 

[maltz]

Here comes a complete change list of the 2nd version of level-9 MOD.

(1) AI's technology is enhanced to 3X of original.

I want to give AI a human-like growth rate. So they must be able to research a lot faster. In 1 vs. 7 the AI team researches 184% times faster than human and they were still lagging behind in military tech. So I made an educated guess of 300%!

But there is no such line of AI's research rate in XML files. And I don't want to make human's research 3 times slower, as that will break all the balances. So I came up with a workaround.

(1) sid meier's civilization v\assets\Gameplay\XML\Technologies\CIV5Technologies.xml
Chnage every technology to 1/3 flask requirement.

(2) \sid meier's civilization v\assets\Gameplay\XML\GameInfo\CIV5HandicapInfos.xml
<ResearchPercent>100</ResearchPercent> -- change to 300.
This means that human has 1/3 x 300% flask requirement (unchanged, maybe little discrepancy due to roundups). But the AIs are happily enjoying 67% off techs!

(2) Eliminating all AI's startup bonuses

\sid meier's civilization v\assets\Gameplay\XML\GameInfo\CIV5HandicapInfos.xml
<AIStartingUnitMultiplier>1</AIStartingUnitMultiplier> - deleted (no extra Settler)
<AIStartingDefenseUnits>2</AIStartingDefenseUnits> - deleted (no extra Warrior)
<AIStartingWorkerUnits>2</AIStartingWorkerUnits> - deleted (no extra Worker)
<AIStartingExploreUnits>1</AIStartingExploreUnits> - deleted (no extra Scout)

(3) Eliminating all AI's explicit bonuses

\sid meier's civilization v\assets\Gameplay\XML\GameInfo\CIV5HandicapInfos.xml
<AIWorkRateModifier>100</AIWorkRateModifier> - deleted. Now AI workers work as fast as human's.
<AIUnhappinessPercent>60</AIUnhappinessPercent> - change to 100. Now AI does not get ANY unhappiness discount.
<AIGrowthPercent>60</AIGrowthPercent> - change to 100. AI's city grows as fast as human's.
I have received a lot of concerns on whether removing AI's unhappiness cheat will cause stunt growth - my actual testing reveals it won't! Because now they have cities as small as humans', eliminating the greatest source of unhappiness.
<AIBarbarianBonus>60</AIBarbarianBonus> - deleted. AI does not enjoy bonus against barbarians.

Note: I forgot to remove Deity AI's 4 free techs in this game.

(4) Enhacing AI's implicit bonuses

\sid meier's civilization v\assets\Gameplay\XML\GameInfo\CIV5HandicapInfos.xml
<AITrainPercent>50</AITrainPercent> - enhanced to 33. The 9th level AI trains units 3 times faster, compared to 2 times faster in Deity.
<AIConstructPercent>50</AIConstructPercent> - enhanced to 33. The 9th level AI construct buildings 3 times faster.
<AICreatePercent>50</AICreatePercent> enhanced to 33. The 9th level AI complete projects 3 times faster. (Not 100% sure on this one.)

\sid meier's civilization v\assets\Gameplay\XML\GameInfo\CIV5HandicapInfos.xml
<AIBuildingCostPercent>50</AIBuildingCostPercent> - enhanced to 33. The 9th level AI purchases buildings with 1/3 of original price. (50% for Deity)
<AIUnitCostPercent>50</AIUnitCostPercent> - enhanced to 33. The 9th level AI purchases units with 1/3 of original price. (50% for Deity)
<AIUnitUpgradePercent>50</AIUnitUpgradePercent> - enhanced to 33. The 9th level AI upgrades units with, again, 1/3 of the original price (50% for Deity).

(5) Army size, from level 7-8-9 linear extrapolation

\sid meier's civilization v\assets\Gameplay\XML\GameInfo\CIV5HandicapInfos.xml
<AIUnitSupplyPercent>50</AIUnitSupplyPercent> - enhanced to 60. Slightly larger army.

(6) Barbarians, from level 7-8-9 linear extrapolation

\sid meier's civilization v\assets\Gameplay\XML\GameInfo\CIV5HandicapInfos.xml

<BarbarianLandTargetRange>8</BarbarianLandTargetRange> - enhanced to 9
<BarbarianSeaTargetRange>20</BarbarianSeaTargetRange> - enhanced to 22

That's it. Now let's watch the AI grow!

 

[maltz]

Spoiler :

001. The very first map gives me River! This must be Nile. No hammer, but lots of growth potential. I will gladly take it.

 

[maltz]

Spoiler :

002. Warrior explores south to reveal more Sugar, but I decided to forfeit some sugars and settle 1 tile to Northeast. Why? (1) I want to get to the riverside Wheat to boost early population growth. (2) There are some 1/1/1 riverside plain tiles to the north. Since I am so low in production, each additional hammer actually makes a good difference. (3) I probably can't find enogh buyers for 5 units of Sugar anyway - there are plenty of marshes and flooded plains on Mesopotamia, too.

 

[maltz]

Spoiler :

003. From my last game in Mesopotamia, I learned that there should be a Civ born on the shore of North Africa. So I sent my warrior to explore west (to know who I am going to defeat first). But their exploration revealed nothing but a City State Stockholm. No Civ on north Africa this time, then. North Africa is mine! But I still prefer to take AI's cities.

 

[maltz]

Spoiler :

004. My capital starts by building a warrior, so I can use two warriors to explore and properly defend against barbarian threat. Here comes one camp! And right in the middle of a close contest, the barbarian warrior underwent asexual reproduction and budded a full-health warrior just beside it. Darn! Under the 30HP mod, every early barbarian camp is an epic struggle.

 

[maltz]

Spoiler :

005. Got enough cash to purchase a Scout directly. So who is my closest neighbor and my first victim?

 

[maltz]

Spoiler :

006. The angry barbarian warrior gives chase after realizing that its camp was razed, and my dying (1/30, 2/30 HP) warrior fled home and went directly to hospital emergency. That was so close...

The AIs entered Classical pretty early (Turn 38 Marathon = Turn 13 Standard). So I realized that I forgot to remove AI's starting tech bonus. But I will play on to see how things develop.

 

[maltz]

Spoiler :

007. The Persian scout went ahead to see who is lucky enough to own Mesopotamia. And it is China! I see Wu Zetian a lot in recent games.

The AI now has completely normal growth rates in their cities. It is good to see the AI plays fair for once - but of course that's just an illusion. Each AI population is equivalent to 3! (Each gold = 3, Each hammer = 3, Each flask = 3).

 

[maltz]

Spoiler :

008. Got enough cash to purchase the 3rd warrior straight out. Since the AIs have no more free workers for me to kidnap, I have to train one the old-fashioned and honest way.

 

[maltz]

Spoiler :

009. Not really! I soon discovered a free worker for Grab in a nearby City State!
2 Workers is a lot better than 1 Worker. Why not?

 

[maltz]

Spoiler :

010. The Scout keeps exploring along the Persian Gulf, and hasn't seen any other capitals! The Mesopotamia map, although stands on 80x50 as a Standard-sized map, is mostly land. (While > 50% of the area is Ocean for a standard Pangaea and Continent) So this more like a large map.