OCC Always War

[oppy]

@nishant can you post some info on your build? If I understand right, you are saying we should take civil service with oracle? I can't even get oracle before math....not to mention getting to code of laws after that to open the civil service bulb...

 

[vanatteveldt]

did the map when the thread was first posted, did okay defense/tech wise but never managed to break out.

Tried again now and underbuilt defense and got badly hurt.

One thing I noticed is that the game would actually be easier if one of the metal-forests was removed so the worker can boost the run to BW after getting mining. Replacing two FP by a cow and a sheep (and/or a horse) might also help in the health dept.

How many workers do you guys build initially (eg before BW)? I found that with only one, (1) you have too many warriors and (2) worker improvements lag behind city growth. I've tried three but that was too many, so probably two is best?

[Edit re oppy] Same here. I actually missed the Oracle this game in 1700BC (on the plus side I got MC from a hut). Tech path was Hunting-mining-bw-masonry-agri-myst-poly-priest-writing

 

[Killroyan]

I got one worker out while teching and grew to size four. Then a second worker when BW came in to chop out all the mines (copper, silver and gold) while building some defenses and the great wall and pyramids. Then when I had nothing to build and enough defenses I build another worker so I could get the fortress up really quick. So I have 3 workers now (950 AD).

Civil service from the oracle was not possible in my case. The first two games it went 5 turns before I would complete it. The last game I just took MC from it.

 

[Kid R]

I think 3 workers ASAP is best as there is so much to do. Not only are all the specials covered with forest but once they're improved the hinterlands are all forested too. All 3 of my workers were chopping nearly non-stop right up till Oxford - all doubled hammers into wonders, plus math bonus for lots of them.

Working gems+gold+silver straight away after the deer I don't think any wonders should be missed except maybe stonehenge if an AI fancies it very early. Def masonry before agriculture - gwall is only 2 chops.

 

[nishant1911]

@nishant can you post some info on your build? If I understand right, you are saying we should take civil service with oracle? I can't even get oracle before math....not to mention getting to code of laws after that to open the civil service bulb...

my initial builds were something like this
worker,warrior,barracks,warrior,worker
use the initial warrior to pop huts and then go out of the pass.
just leave one on the pass.
kill those scouts for xp. got my first woodsman III this way

initial tech path was hunting,(popped agriculture),mining (5 turns in, then popped it from a hut),bronze(don't adopt slavery),pottery, writing and library's scientists asap,mysticism,meditation,priesthood,maths,code of laws.
you will have to micromanage your capital between production(you will run out of things to build early on) and research(max it out)

cs slingshot consists of using first gs to bulb maths while researching code of laws. but here i researched both manually and completed oracle 1575bc (yup you need some luck, could have completed it 1675bc by bulbing but decided to gamble)

i have attached the 300 ad save ,look at the event log for exact timings.

 

[nishant1911]

nevermind

 

[cripp7]

A little cheat that I did in WB was added gold to the floodplains, which is essentially when a mine is built it is a town. Which that really skyrocketed the research, but I am still trying to get down the CS Oracle slingshot in my own game. I have completed domination in the mid-1800s, but I am going to try a space bee lining Internet to pick up everything else

 

[lymond]

A little cheat that I did in WB was added gold to the floodplains, which is essentially when a mine is built it is a town. Which that really skyrocketed the research, but I am still trying to get down the CS Oracle slingshot in my own game. I have completed domination in the mid-1800s, but I am going to try a space bee lining Internet to pick up everything else

I would be interested in a report on your game, cripp, if you got a dom win. You appear to be the only one to do this so far.

It was fun for a while playing hill defense and I did alright in teching, but just couldnt break out. The kicker is that there is always one civ that gets out ahead and gets Infantry sooner than me. I just couldn't break out. I may try different tactics in replays.

Interestingly, I played twice - both to around Industrialism. In the first try, I did not tech as well but held out for a long time. Interstingly, in this game the opposing stack were generally pretty manageable. In the second game, in which I teched much better, I faced some seriously monstrous stacks.

 

[Kid R]

I broke out early with 12 super GG infantry because Hammy built the UN and robotics was miles off. Razed the UN and lots of other cities but gradually got sloppier because of how strong my guys seemed, and they eventually got split up and killed. 12 units weren't going to clear the whole map anyway so kind of gave up. Got about 1/4 the way round and I'm sure the AIs were just resettling the land behind!

Razing the buddhist and hindu holy cities made a big difference to the AI tech pace though, and next time I'll try and make a quick lightning strike much earlier and do that job, maybe just after the troops upgrade from samurai to rifles.

My general plan is to make warlords out of all GGs and level each one up to level 20-21 by rotating them in and out of the pass. Then when the time comes send a stack of 9-10 each way round the map, mopping up after them with much weaker fresh units. The worries are whether they can make it all the way round, and how effective the mop-up troops will be, but the premature breakout attempt with infantry was encouraging. 9-10 super units can take a city every 2 turns or so, and there aren't too many enemy units left to mop up because of how keen the AI is to suicide against the invincible stack. So with a 30 turn headstart on mech inf it might just be doable ...

 

[vanatteveldt]

Good idea on the shrines. Reminds me to research alphabet at some point so I can build spies to do some recon. Did you use commando units to get there in reasonable time?

What do you guys do with the floodplains? I cottaged them up, thinking that financial + cottages would be a good deal, but I think watermilling them might make more sense.

Let's think of three periods, early, mid, and late

Early: (assume metal casting, code of laws)

cottage/hamlet: +2/+3c (fin bonus plus
watermill: unavailable
workshop: -1f +2h
farm:+1f

Mid game (assume villages, guilds, no SP elect chem)

Cottage: +4c
Watermill: +1h
Workshop: -1f +3h
Farm: +1f

Late game (assume all)

Cottage: +5c (+1h if US, but costs a lot of beakers and we don't have much gold for buying)
watermill: +1f +2h +3c (elect + fin bonus)
Workshop: +0f +4h
Farm: +2f

Since health will be a problem unless we build the NatPark, a specialist will cost 3f for most of the game, meaning 1f=2c+1gpp

So it seems that in the early game, it might not make a lot of sense to use them at all. Cottages make little sense compared to farm/specialist taking the high value of early gpp into account. In the mid ganme, cottages and workshops more or less tie. In the late game, watermills win hands down. Cottages lose value becuase we won't be running free speech as it conflicts with buro.

My conclusion I think is to avoid cottages next time, and build farms early on and focus on getting many great people. Is that what others have been doing?


[Edit I guess replacing some of the specials might also make sense.

Copper mine: +4h
Copper windmill: +1f+1h+5c (4c plus fin bonus assuming environ)

beaver camp: +1h+3c (+1h for forest, +3c plus fin, -1 for river blocked)
ele camp: +2h
deer camp: +1h+2f +1/2health -1c
riverside lumbermill: +3h (beaver +1c since 1c beaver 1c river -> 1c fin)
watermill +1f +2h +3c -.5 health

So I think it makes sense to replace the copper mine with a windmill if we run environmentalism and we don't need the hammers badly. It makes sense replacing the beaver and elephant by a lumbermill or watermill depending on the value of hammers and health. The deer should probably not be replaced until health problems are solved.

I don;t think it will ever make sense to replace the cereal farms or the precious mines.

Anotherl consideration: assuming that we need both epics, oxford, and globe, we have room left for one nat wonder, presumably natpark or ironworks.

If we choose iron works, we can get health from cereals (+6), deer (+2), environmentalism (+6), pub trans (+4), hosp (+3) aqua (+2) starting (+5 fresh+forest+diff) = 28. Genetics (+3), gardens (+1), and experiments (+1/2) might give extra bonus but let's not count on it. Unhealth comes from forge (-1) factory (-1) ind. park (-2) ironworks (-2) lab (-1) airport (-1) coal (-3) oil (-3) = -14. So, we will be health-neutral at pop 14. Assuming a pop of around 24, thils will cost us 10f but give us +100% production, but also take away the option of running SP (since we will need envi for our health).

If we chooes the natpark, we don't have to run envi, so have +20 health, which is plenty to cover the buildings, so we can safely chop the forests and watermill the deer.

My conclusion is that natpark is superior to ironworks.

Another option would be dropping heroic epic and taking iron instead, but I don't think that makes much sense as iron will only give +50% prod with natpark (as coal is gone) and it comes too late to have many buildings left to build.

Final question: I found that I get a lot of axemen with high experience. If I upgrade them the XP > 10 is wasted, so I am trying to get great generals for the axemen with really good experience. This implies that it makes sense to focus XP on a few axement, ie don't rotate troops to divide the XP. Is that what other people are doing?

So, I am going to try a new game, avoiding cottages, gunning for early scientists (which migth make the CS slingshot possible) , and concentrating XP on a couple of axemen.

Edit (update): Some observations from the early game:
1) I think it is worth it to not collect the huts in your 'hinterland', and let them be collected by city growth. This gives better results (more chance at tech) and later techs are more valuable. In three games, this gave me two techs (metal casting and maths), both quite valuable.
2) In my last game, I didn't build a single warrior. The initial warrior collected the two nearby huts and killed scouts but stayed close to the pass (to prevent scouts from entering and stealing our huts). Build order was something like worker - barracks (grow) - worker - worker - warrior (turns into spearman) - axemen (plural) - library.
3) I think that you have to choose between great wall (more GG points) and oracle-CS, but if someone manages to snatch both I'd like to hear about it! Probably, oracle-CS will be a gamble in the best case...

Edit: update tried two games. In both cases cheated by WBing the forest from the gold to speed up early game. First game without GWall (ie writing before masonry. Got lucky and popped math from hut so didn't even need scientist), one with masonry GW -> writing Library. Had to WB a GSpy into a Gscientist (50/50 chance). Missed oracle in both cases, went in 2100 BC and 1675 BC respectively. I think you need until around 1500 BC to get all prereqs for CS. So, my conclusion is that the CS slingshot is a total gamble, even when ignoring the GWall.

 

[Kid R]

Good idea on the shrines. Reminds me to research alphabet at some point so I can build spies to do some recon. Did you use commando units to get there in reasonable time?

The strike party all had commando and morale, moving about 6 or 7 tiles a turn through enemy culture, which isn't bad for supposedly 1-move infantry! I tried a few spies earlier on but they got detected so soon I gave up.

What do you guys do with the floodplains? I cottaged them up, thinking that financial + cottages would be a good deal, but I think watermilling them might make more sense.

Interesting. For me the most important thing is research though and it's between cottages and farms. If the city has enough hammers to build all the wonders and 1 mech per turn at the end that's enough for me.

I was going to go cottages again next try but you've persuaded be on farms.

Say the cottage averages 5, that's 5 x 175%, making 8 or 9

The farm is just 1 for 1.75 but powers an extra 0.5 x representation scientist (4.5 base ) before biology, so thats a total of more like 10 which is better on the face of it.

After biology the farms get even better and as you said the GPP are yet another bonus.

I guess replacing some of the specials might also make sense

Good call. Copper windmill sounds good, elephant and beaver farm maybe. Not convinced about watermills myself.

...national wonders...

Last time I held a slot open for Ironworks, but as time went on it became increasingly clear it wasn't going to help and I wished I'd gone for the nat park and more lovely lovely specialists earlier on. Agree with your comparison completely.

 

[Killroyan]

Right now I have 75% of the floodplains cottaged and a few watermills while working on replacable parts. Cottages are early game a bit better for the research if you ask me since it takes a while to grow and get more specialists going then I am running already. Happiness is not yet a problem at size 21. When I get to biology I will propably farm up half of them and watermill the other half or maybe even workshop them.

National wonders I am aiming for are HE/NE/Oxford/NP and then propably still Globe. Crap I forgot if NP and ironworks can be combined? Since you can't use any coal with NP.

 

[vanatteveldt]

you can combine natpark and iron, but only get the 50% bonus for iron (as coal is removed).

The main thing mills have going for them is the fact that they give decent science (compared to non-FS towns) but also give hammers and a food. After natpark, health is not an issue so a food is half a specialist = 3science plus GPP. So, you give up 2c to get 3science (which is equal after buro) and get the 2 hammers and GPP for "free". Of course, it will take a while to get natpark and electricity, so early on it's different.

On farms vs watermills: You trade 1 food for 2 hammers and 3 commerce (=4.5 science). Since a food is half a specialist, you trade 3sc1.5GPP for 4.5sc2h. So, the question is entirely on the relative merit of hammers and GPP. So, don't use farms for the science (or gold), purely for the GPP.

Also, if you still have that copper mine:

copper mine plus farm = +2f +4h
copper windmill plus watermill: +2f+3h+6c (+8 if not envi; the +6 might be +7?)

So if 6/8c is worth one hammer, replace the mine and one farm by a windmill and watermill! The elephant and beaver have similar considerations as replacing them and a farm by two watermills is food neutral.

 

[White Dragon]

2nd attempt.
Emperor/Epic. Also, I turned off Events & Vassals as well as Espionage since in an OCC it is not at all fair.

This try I went with cottages on the flood plains instead of the farms since, as you can see, early health is non existent.

Ran Parthenon - Great Library - National Epic - Pacifism for most of the game.

Stayed away from Globe Theater and went Ironworks & Heroic Epic.

Used Market, Cathedrals, Temples, Incense & 2 Eiffel Tower enhanced "Hits" for the extra happiness up until the war.

Somewhere between 1600AD & 1800AD I swapped cottages for State Farm powered watermills and workshops.

Used the Chariot & Samurai hammer overflow to gold trick to raise my gold reserves sometime starting in the 1300's. Of course there are numerous wonders and improvements that interrupted the queue during the 1300-1700's so I did not get as many units or gold as I would have wished.

Here are some screen shots.
Missed Oracle all together. Did not pop a single tech from a hut.
1 turn from Liberalism in 900 AD. Had a Great Engineer saved for a quick National Park.

Spoiler :

1030AD

Spoiler :

1370AD. My Hammers were in pretty good shape at this point considering I didn't have a Factory or Shale Plat (which provides power regardless of coal).

Spoiler :

1515AD. Swapped from all Scientists to Angkor Wat Priests a little while earlier. Added another 100 hammers per turn total.

Spoiler :

1610AD. Made final swap to all Engineers & Priests and railroads on mines. Only 8600 gold at this point. I will need alot more to upgrade those samurai.

Spoiler :

1812AD. Didn't have a save of my pre-war empire but here are some shots just after the start or war. 4 Great General Mech's lead the charge with CRII, Drill IV Mech's on cleanup.

Spoiler :

1836AD. Infantry are showing up in force. Surprisingly they are not a problem. Machine Gunners are worse with the +50% Gunpowder and the canceled First Strikes. Have started to take heavy losses so I have slowed down substantially. Hopefully I can pillage improvements while waiting 3 turns for a single reinforcement, even having railroaded in enemy territory.

Spoiler :

 

[Killroyan]

Played a little further to tech towards rifling and thought I was save. Was switching my great generals and high level samurai a bit to give each one a good chunk of XP. Wow, bad mistake when I took 3 of my high level units out of the fight (still had 1 GG defending and several 60+ exp units). Defeated 40 units, then the dying started. The AI killed 20 of my units with some extra losses and then invaded me with still 36!!! units left. How about that stack of doom??? And they were all Viking for crying out loud.

 

[Kid R]

Hehe, I reloaded when that happened to me, it doesn't feel so much like cheating if you deliberately weakened your defence there, it's just a micromanagement issue. If the Civ 4 HUD was clearer, or the game didn't involve such huge stacks where the list goes right off the top of the screen, you could more easily see when huge stacks were approaching and move enough defenders into position to ensure safety but get some XP on the weaker units.

Incidentally one thing I noticed is that on the turn before each new defensive upgrade (samurai, rifle, infantry, mech infantry), the AIs sent huge attempts to get through the pass, and all defenders had to be in place that turn. Might be conicidence but it happened every time. Obviously deity AIs can see our research, but the AI is quite impressive if it takes that action as a result.

 

[Killroyan]

Wow, you are actually right about that. I had 1 or maybe 2 turns left before rifling when this mother of all stacks attacked. No problem however I will reload this evening again. I am having way too much fun with this.

 

[vanatteveldt]

Two funny observations:

- I got a scout from a hut and thought I'd go scouting. A number of times, I landed next to an enemy archer or vulture (if that's what they are), and wasn't attacked! I guess if they have an explore or other mission they ignore targets of opportunity...

- I played without espionage, and built the great wall. I got GP points, but no "100%" anything. So I guess there are some sort of 'colourless' great people points?

On the downside, I've missed the oracle every time I tried, and am now revising my strategy to go hunt - bw - masonry (GW) - agri - (pottery?) - priesthood (oracle metal casting). That should probably be in time to snatch it.

I've settled down on playing with no espionage and the forest removed from the gold mine, that feels fair but doable (late game espionage is really boring with constantly rebuilding sabotaged improvements...).

 

[Kid R]

- I played without espionage, and built the great wall. I got GP points, but no "100%" anything. So I guess there are some sort of 'colourless' great people points?

It's +100% great generals (which is crucial in this game), not GPP. You also have to wait till culture reaches the hill, which is a lot quicker the more wonders go into Kyoto. Without espionage, I don't know, perhaps it does make something strange happen.

forest removed from the gold mine

Waste of hammers!!!! Build more workers! My build order last game was:

worker
garrison warrior
part barracks
worker
worker
great wall
complete barracks
worker
stonehenge part and fail
axeman
oracle
...

4 workers and it didn't seem too many.

 

[Kid R]

Speaking of which, some thoughts about that game (second attempt - failed, but not so badly )

1. The bad

Spoiler :

1a)
The strategy of using all GGs for warlords means that after robotics the pass has to be defended by quite weak freshly-built mechs, with just 2 or 3 promotions. The AI can defeat these quite handily with artillery and regular infantry and I was overrun about 20 turns after robotics. I'm going to try a few reloads but I suspect defending in larger numbers, say starting with 20 such stock mechs and replenishing one per turn thereafter, might be OK.

1b)
I tried breaking out at rifling with 5x 300 XP warlords to raze the 3 holy cities nearby. No dice, riflemen are too vulnerable to cuirassiers. After reloading and playing on I tried the same mission with the 8 best warlord infantry when I got to asssembly line. Excellent results! In fact I got carried away razing cities and cut down Hammurabi far too much, and Shaka a bit to much. This effectively made their neighbors Toku and Boudica into monsters who attacked/vassaled shortly afterwards, and were much stronger in the end game. Next time self control is required - just hit the holy cities and return.

Another undesired result of the long mission outside the valley was that it distracted the AIs from attacking the pass, and my defending stack got nearly no XP for a long time.

1c)
The "mop-up" troops I imagined streaming out of the valley at the end to help the main stacks are not going to work if they're weak like this (compare to settling 20 GGs for 40 XP which gets drill 4 at least). They would get cut down almost immediately. Indeed if they could even get out at all - turns with no enemy units on the exit tile don't come that often. In summary this strategy is going to rely on the super-unit stacks operating alone, except maybe when there is just one or two AIs left.

2. The good

Spoiler :

Despite not managing the XP-gathering mini-game as well as I could have, at Robotics there were still 10x 300-450 XP warlords who could go out as the desired 6-move unbeatable stack. The AIs were still fielding cavalry, cannon and had just got infantry, but in any case they really have no idea how to handle a stack approaching cities from 4-6 tiles away. I guess not knowing which city will be attacked they seem to hedge their bets and just leave their units roaming about! This is great because cities generally then don't have many defenders, and the 10 warlords could usually clear out one city per turn. Or sometimes 2 turns killing a few easy infantry or cavalry on the way to a convenient resting hill.

Eliminating the most advanced AI, Boudica, was easy before Kyoto fell, and I think the more ******ed AIs could have been taken down after that before they reached a similar tech level. So in summary 16 super-units (2 stacks of 8, one going each way) would at least have a shot at clearing the whole map, and I believe with better XP management a force like that is achievable.

3. The indifferent

Spoiler :

This time I used all farms instead of all cottages, but apart from that I managed the city much the same, and I would be hard pressed to choose a winner. I was probably a bit more streamlined with my wonder path but built pretty much the same ones as before (that is, nearly all of them). Gut feeling is the cottages approach was better pre-biology (i.e. pre-national park) but lost ground after that. Hard to tell though because I built the globe theatre this time and didn't before.

Another big thing about the farms is it means more pop to draft, doubling unit production speed, in the 10-20 turns immediately after robotics. I'm not convinced this is crucial any more (see above), although it's defintely a handy ability to have in reserve just in case.

4. The unlucky

Spoiler :

By the end the AIs were in 3 vassal+master teams, which improved their tech and gave me less of a lead and therefore less time to do the job at the end. Two of the vassalizations were my fault (see above) but one happened just by chance when Toku and Monty got a random event (hindu wedding thing) which put them at war.

Random events destroying forges and libraries are a pain, but I guess the AI gets those too. Espionage I don't really mind. 6 workers can rebuild the copper mine or a farm in a single turn so nothing is lost there, but I agree losing sabotaged buildings is tedious. On the other hand you have to think if the AIs are investing in espionage points it takes away from their teching, which is good.