Population (worked tiles) diminishing returns?

[Bibor]

@Bibor Not settling on banana slows you down significantly so it's a bit strange you don't do it on your 2nd run.

It's hard for me, believe me, because it goes against my long established house rules regarding settling: if I can help it, I don't settle on food resources that yield 5 or more, dye, or non-desert metal.
I will look at Swnb's save more closely, although I already have an idea for a 6-city construction rush. The jungle can wait this time around. I finished my second run with a T250 conquest victory that could've been easily a T200 victory if not for my "empire building" and probably even faster if I would've not invested into the jungle.

 

[krikav]

It's a habit I think we all have had one time or another when we have learned this game.
It's some abstract thought of what is best "in the long run".

The problem is, that "in the long run", the game is already lost or won.
Getting one extra food in your initial capital, thereby getting your first worker 3 turns earlier lets you improve the corn 3 turns earlier which gives you at least 3 extra food per turn 3 turns earlier, etc etc.
This in itself gives you a better position in the long run, much more so than some slight benefit that you can get once you reach calendar, at the time where the extra 2 food from the banana won't make much differance at all.

Something that annoyes me, is that resources make the city look weird when you zoom in.

 

[coanda]

Calendar comes fairly late. You'll probably have 6+ cities (in your more recent save it was turn 137, 8 cities), and odds are something like 20-50 population in your empire by that point. A single extra food from a single tile (the difference between settling on a banana and working a farm, or setting on another grass tile and working a banana plantation) is *maybe* a 2% boost to your empire's productivity.

Meanwhile settling on top of the banana with your starting settler and working the corn means 4 surplus food instead of 3. 25% faster worker production on that first worker, then 33% faster growth while you wait for the corn tile to be farmed. I played the 4000 BC save up to the point of settling a second city, and Utretcht was founded on turn 30. In your 25 AD save, the event log shows it founded on turn 37; a 7-turn delay almost entirely due to choosing not to settle on the banana.

 

[sampsa]

Also, "wasting" a 5 tile is really sacrificing only one even long term. Settling on a grassland river, working banana is +2 city center, +3 banana, +5 total. Settling on banana, working grassland farm is +3 city center, +1 farm, +4 total. You can of course often build a better improvement than a grassland farm (grass river cottage). Also note that when there are many decent grassland tiles available, settling on sugar or dry rice is very attractive.

*cross-post with coanda

It's hard for me, believe me, because it goes against my long established house rules regarding settling: if I can help it, I don't settle on food resources that yield 5 or more, dye, or non-desert metal.

I think plains hill metal tiles are the best to be settled on, -2 tiles are so weak that you rarely want to work them (gold in the early game or when tons of food available is an exception). You need to appreciate food more, that was also apparent from your save as you build too many mines, when great tiles like financial grass river cottages are available.

I will look at Swnb's save more closely, although I already have an idea for a 6-city construction rush. The jungle can wait this time around. I finished my second run with a T250 conquest victory that could've been easily a T200 victory if not for my "empire building" and probably even faster if I would've not invested into the jungle.

Yes, certainly an early attack will nearly always lead to great results on emperor level. However, my opinion is that empire management is a more important skill to learn, as the nuances remain the same when you climb up the levels to immortal and finally to deity.

 

[Bibor]

Meanwhile settling on top of the banana with your starting settler and working the corn means 4 surplus food instead of 3. 25% faster worker production on that first worker, then 33% faster growth while you wait for the corn tile to be farmed. I played the 4000 BC save up to the point of settling a second city, and Utretcht was founded on turn 30. In your 25 AD save, the event log shows it founded on turn 37; a 7-turn delay almost entirely due to choosing not to settle on the banana.

It's when you lay it down like this that it start to cure my house rules This game is truly difficult - not because of the difficulty level - but because of the required brain processing power for hours on end.

Yes, certainly an early attack will nearly always lead to great results on emperor level. However, my opinion is that empire management is a more important skill to learn, as the nuances remain the same when you climb up the levels to immortal and finally to deity.

So, in your opinion, what timing would be good for an attack on this map, with decent time allocated for empire development practice? Perhaps Engineering?

Also, one more question - I'll probably need a merchant at one point to upgrade my troops from swords/axes to maces or horse archers to knights, since there's so much food, is it better to simply stay (on this map) in slavery and whip, and build a marketplace in one of the cities to run 2 merchants, or would it be better to go caste and get that merchant out faster (but lose 2 turns on disorder)?

 

[sampsa]

Also, one more question - I'll probably need a merchant at one point to upgrade my troops from swords/axes to maces or horse archers to knights, since there's so much food, is it better to simply stay (on this map) in slavery and whip, and build a marketplace in one of the cities to run 2 merchants, or would it be better to go caste and get that merchant out faster (but lose 2 turns on disorder)?

I think you don't need those upgrades at all, as siege is doing all the real job and other troops are just there to mop it up. Market is a big investment and it's rather inefficient to build it just for a GM. If you really really need a merchant, I'd start a golden age with a GS and dip into caste to get 2-3 out.

 

[Swordnboard]

Music artist-->golden age is also good. I'm planning to use cuirassiers to attack, since no need for siege and the distances are so large that the movement is a big deal. Coincidentally that's what I would use the scientists for (bulb philosophy and education). But as others have noted, elephants or medieval units would also do the trick against brennus's outdated military.

 

[elmurcis]

Out of topic about breaking habit - damm, it was so difficult today to get over fact that AI capital near me is settled on plain (not Plain Hill) Copper (Gandhi, managed very quick and high effiency Archer-rush as there was no other Copper anywhere near and he never got to BW at all). But that city (because it had "just" regular grassland hills for short term use but 4 food tiles turned into beast managing 4-pop settler whip every whip cycle with max overflow into building Oracle (for Math chops), ToA, HG, TGL and big failgold from Parthenon With old habit I would raze that city and resettle, costing huge amount of hammers, delaying everything and most likely missing every single wonder it got here. Maybe I would feel happy for moment (empire *must* make me happy - I will run US just for that instead of Rep with Pyramids) but after that - big big anger for lost options

 

[Swordnboard]

Hey Bibor!

I finally got around to playing until 1 AD. Final exams + Holiday Season really did a number on my civ time. I'll put some analysis here later, but have a plane to catch at the moment.

 

[Anysense]

Impressive bpt @Swordnboard ! What is the plan? Assuming you are aiming at Space, would not it be more effective to go for larger empire earlier, like 15-20 cities 1AD? You'd be teching slower until certain point, but then there would be explosive growth, once you've finished expansion.

 

[Bibor]

Out of topic about breaking habit - damm, it was so difficult today to get over fact that AI capital near me is settled on plain (not Plain Hill) Copper (Gandhi, managed very quick and high effiency Archer-rush as there was no other Copper anywhere near and he never got to BW at all). But that city (because it had "just" regular grassland hills for short term use but 4 food tiles turned into beast managing 4-pop settler whip every whip cycle with max overflow into building Oracle (for Math chops), ToA, HG, TGL and big failgold from Parthenon With old habit I would raze that city and resettle, costing huge amount of hammers, delaying everything and most likely missing every single wonder it got here. Maybe I would feel happy for moment (empire *must* make me happy - I will run US just for that instead of Rep with Pyramids) but after that - big big anger for lost options

I guess what's so terrifying and beautiful about Civ4 is that the combination of random maps and opponents are constantly shifting the ground you stand on, like the sea. Form comes from the consistency of the understanding on how you need to play, not on some arbitrary predetermined idea on how you will play. Trying to fit in your quirks that affect gameplay doesn't work (I'll build temples in every city no matter what), only descriptive quirks (I'll name each city after my cats names) do.
If you want to master it, this game deliberately asks you to break your OCD habits and see most of the apparent "choices" for what they are - that they won't pay themselves off, given the turn constraint.

 

[Swordnboard]

@Anysense: yeah, I think space is a decent idea from here. Could perhaps also shift gears to UN diplo if you wanted to. As for extra early expansion, I would say on deity it wouldn't be necessary, as the AI would develop things for me to take with cuirs. But on emperor this huge jungle area is still undeveloped and having a couple of cities earlier would probably help. It'll all get workshopped over once communism comes in anyways.

Anyways, on to comparisons/suggestions for OP. I think the most important changes would be a faster start (settle on bananas and found cities 2/3 in closer, more immediately productive locations), more aggressive trading for both techs (religious stuff/monarchy) and resources, and expanding in a way that doesn't interfere with key techs like currency and CS.

 

[Fish Man]

Played until t54

This is actually a pretty killer start, all things considered. First time I've had this much food in a while. And fishing's useful as a starting tech for once.

Spoiler :

My continued obsession with fish has led me to settle the first 2 cities in...pretty decent spots, actually. First is the ideal GP farm (FOUR food resources in BFC) and will get a library up ASAP; second is cottage heaven and has WAY more food than necessary. In both cases coastal fin fish is one of the strongest food tile in the game IMO, with a potential to yield 6f3c, as much commerce as a riverside cottage and as much food as a wet corn. The only drawback is work boat, but requiring a work boat is the same as needing 4 worker turns (to chop a single forest), which makes it about the same as other food tiles in terms of improvement speed once you have BW.

Brennus is pathetically behind in tech/cities, whatever he's doing. Come on emp AI...step it up.

Only issues are I haven't been able to meet any other AIs, and...barbs are beginning to worry me. There are 2 barb archers out there in the fog AT LEAST, and until I have chariots I can't deal with them, much less prevent new spawns.

I'm conflicted about what method to use to conquer. Cuirs are fast and will be such a stomp on emperor but it's a HUGE detour from communism. Knightrush...alright, but very iffy once they hit longbows. Trebs are slow but reliable and directly on the path to chemistry -> SM -> communism (I can probably lib communism no problem here). And cannons...way overkill, also steel is giant detour, though the advantage is absolutely nothing the AI can field at that date will be able to stand up to them. Not a big fan of elepaults, considering how we just have so much land to expand into still...

 

[sampsa]

@Undefeatable a bit strange choices already. Why granary with this -cap? Why stagnate capital @3 and not work cottages?

 

[Fish Man]

@Undefeatable a bit strange choices already. Why granary with this -cap? Why stagnate capital @3 and not work cottages?

Because if I wanna whip I don’t wanna spend 500 years growing back.

Because I need a worker NOW and if I can hard build them in 4 turns it’s better than whipping.

 

[Fish Man]

Played until 1AD

Thought it would be nice to try a knightrush for once .

Reasons - all the other AIs are a billion miles away. For some reason the generator gave us a third of the map to cross just to reach Brennus, much less anyone else. Even after roading that distance, if I use phants or other 1-move units it's gonna be ultra slow going. Over 200 years from when I build the army to when I declare, at least! Also, knights are brutally effective against almost all classical era units (praets and phants notwithstanding). They shred archers in even the most fortified cities, and make quick work of most melee units too. They also win 1v1s handily vs spears, the mounted "counter." In fact, knights vs classical/ancient is more of a stomp than even cuirs vs medieval, based on the strength ratios. Longbows are a problem but on this map, nobody's even close (c'mon, Brennus and Shaka, do something with those 8-9 cities of yours). Finally, it's much less of a tech detour to knights. If we're talking space, we want to avoid nationalism/MT and lib communism ASAP, followed with a direct path to either steam power for fast workers or electricity for supermills. The nat/MT path helps with none of that and is 6k beakers dumped into a unit that, at this level, comes too late to take advantage of the emperor AI's utter inability to develop newly settled cities (I mean, there's gotta be a reason that Brennus doesn't even have fricking IW or writing yet, right?), but too early to really take advantage of the tech snowball you ride on most science wins (I've gone tanks vs longbows once. It took a single galleon to capitulate a civ). Meanwhile guilds is needed for superworkshops anyways, machinery for engineering->chem->SM, and feudalism for capitulating people you don't want to destroy completely.

5-6 turns from guilds, also almost at edu because of a bulb. I can't for the life of me, though, figure out why my BPT is ~40 lower than @Swordnboard 's even after taking into account the golden age he's having?

Shaka's super behind, too, especially after getting stuck in a tangle with Vicky. If I'm lucky I can clean up 3-4 AIs with knights and then ride the land advantage to space sub-t250.

 

[sampsa]

5-6 turns from guilds, also almost at edu because of a bulb. I can't for the life of me, though, figure out why my BPT is ~40 lower than @Swordnboard 's even after taking into account the golden age he's having?

Shaka's super behind, too, especially after getting stuck in a tangle with Vicky. If I'm lucky I can clean up 3-4 AIs with knights and then ride the land advantage to space sub-t250.

No foreign trade routes? I guess you should have cleared up the fog in northern shore, also Brennus's area. I don't really understand why no trade routes though, as the symbol is there on the scoreboard.

Overall I like your strategy (expand using knights) but I think sub-optimal technical play (I guess many people call this micro) is slowing you down. Not only lack of trade routes, but also I think a lot could be improved pre-T50. The main points were already presented in Cookbook-thread, so it feels moot to have the same conversation again. So if you really want to improve your play, I'd suggest investigating the early game as scientifically as possible.

 

[Fish Man]

No foreign trade routes? I guess you should have cleared up the fog in northern shore, also Brennus's area. I don't really understand why no trade routes though, as the symbol is there on the scoreboard.

Overall I like your strategy (expand using knights) but I think sub-optimal technical play (I guess many people call this micro) is slowing you down. Not only lack of trade routes, but also I think a lot could be improved pre-T50. The main points were already presented in Cookbook-thread, so it feels moot to have the same conversation again. So if you really want to improve your play, I'd suggest investigating the early game as scientifically as possible.

After some snooping I noticed the difference was simply 2 non-riverside towns that I didn't bother to build, and also that he was running like 7-8 more scientists to pop some GS during a golden age.

I think my early game can use some improvement, yes. I've been winging it and going on instinct for the most part, following my gut on how to get out settlers/workers fastest. Usually I follow a "whip cycle" for my capital where I 4-2 or 5-3 whip settlers, use overflow to finish worker (if there's enough production), build warrior to grow back/get rid of unhappy, repeat. Put a few chops in there when needed. Is that not how I should do things?

I could look back in the cookbook, but it's such a long thread, I'll have to do some digging. Not averse to that, though it might be a while, and I'm known to be very impatient.

In the meantime...t167 update - knightssssssss...

I think I'm on track for a sub-t250 space win here; got 28 cities with room for about 2-3 more. Capped 3 civs with knights in just about 30 turns and somehow didn't encounter a single longbow. Spears, swords, axes, archers were all easy pickings for a 10-strength 2-move unit with shock. In other news, it's been a while since I've messed with fin colossus but...I've forgotten how much I missed literal bags of money floating on the coast waiting to be picked up. Giggles is plotting but I could gift him philo and then beg map or 1 gold to stop that nonsense (or he could be going after Asoka, y'know, his only valid land target). Or he'll attack with vultures and be shredded by my veteran shocknights. Any of that works.

Just got communism and switched into endgame civics (SP and all that jazz). Workshops are suddenly OP now. To savor Colossus as long as I can, I'll be going economics, corporation, then steam power/AL. Electricity can wait; I only have a few riverside windmills and watermills take so long that the opportunity cost has thus far prevented me from building any. In the meantime there's somewhat of a worker shortage but I have about 1 every city so it should be fine once I build a couple more, maybe.

Anyone else playing through this to space?

 

[sampsa]

I could look back in the cookbook, but it's such a long thread, I'll have to do some digging. Not averse to that, though it might be a while, and I'm known to be very impatient.

Mostly I mean the following responses you got to getting the fish-city up quickly.

Especially in the early game you want to settle cities that become productive as soon as possible. In your example you mention fish takes 12 turns to improve, while pig takes 4. If we only consider those resources (haven't studied this particular map closely yet), then settling the pig first gives you a productive city 8 turns earlier, meaning that city can give produce a worker or a settler 8 turns earlier. 8 turns earlier worker means it will always be 2 improvements ahead, 8 turns earlier settler is a huge deal. As with everything in this game, early gains are way more valuable than later gains. Besides, if we look at total food produced, fish first will probably never be ahead of pigs first, so there isn't even a later gain to be had.

I think you are making some mistake in what you classify as productive. You say that the fish city does something productive when it is producing a monument. I have a hard time accepting that.
That monument is built only to bring that fish city to the initial state of pigcity. It's a necessary evil, an investment sink that you have to go through to eventually make the city productive. Alot similar to clearing jungle to make jungle cities productive. Clearing jungle isn't something that I see as productive, and I'm not rushing into the jungle asap after IW either.

OK, here it is different since no border pop is required (no wasted 30, then 10T waiting period). However, the point remains that improving seafood costs 30 (and you need to have a way to get them quickly) while improving a -resource with a worker costs nothing but worker turns. In your fish city you spent those turns building mines, which admittedly are somewhat useful later. But that's later, and right now we shouldn't care much about later. So while putting 30 to a work boat is productive, maybe some other site that is less draining to your empire is superior, since setting it up costs less and worker turns?

I think my early game can use some improvement, yes. I've been winging it and going on instinct for the most part, following my gut on how to get out settlers/workers fastest. Usually I follow a "whip cycle" for my capital where I 4-2 or 5-3 whip settlers, use overflow to finish worker (if there's enough production), build warrior to grow back/get rid of unhappy, repeat. Put a few chops in there when needed. Is that not how I should do things?

Sounds OK, if you whip settlers asap when available. Waiting for bigger overflow certainly just slows you down. I'm also assuming you have no granaries, because they are a big investment thus slowing down early expansion a lot, not to mention that the -cap is low, mitigating value of granaries. In cities with many grass cottages available, just growing to -cap and slow-building settlers/workers is a good option. Also depends how much chops available of course, you are correct in that a lot of attention needs to be paid on getting settlers/workers out.

This is also why I dislike your handling of northern seafood-city in your T54 save, it has drained a lot of worker turns (roads, mines), (workboats) and now needs to build a granary, only then it starts to contribute to your settler whips (whipping with a granary also makes those mines rather marginal)... There must be some other way. After putting all those resources into this city, I think the best is now to slow-build settlers/workers @4, because you will get them faster that way. I think even with a higher -cap (say we were CHA) I'd prefer this to whipping. EXP would change things certainly.

So, since it's a city that needs a lot of attention, I'm not sure if I would have prioritized it and certainly not wouldn't handle it the way you did. Helper 3E1N of cap could be better IMO (can also borrow corn, can work cottages, is connected), cow-horse could be prioritized... I know I'm ruthlessly bashing your save but I only dare to do that because I know you are a really good player. So while I'm at it, I dislike the road, the unworked cottages and the unchopped forests. 1.improve food 2.chop 3.cottage 4.connect in this order gets you really far.

In the meantime...t167 update - knightssssssss...

I think I'm on track for a sub-t250 space win here; got 28 cities with room for about 2-3 more. Capped 3 civs with knights in just about 30 turns and somehow didn't encounter a single longbow. Spears, swords, axes, archers were all easy pickings for a 10-strength 2-move unit with shock. In other news, it's been a while since I've messed with fin colossus but...I've forgotten how much I missed literal bags of money floating on the coast waiting to be picked up. Giggles is plotting but I could gift him philo and then beg map or 1 gold to stop that nonsense (or he could be going after Asoka, y'know, his only valid land target). Or he'll attack with vultures and be shredded by my veteran shocknights. Any of that works.

Just got communism and switched into endgame civics (SP and all that jazz). Workshops are suddenly OP now. To savor Colossus as long as I can, I'll be going economics, corporation, then steam power/AL. Electricity can wait; I only have a few riverside windmills and watermills take so long that the opportunity cost has thus far prevented me from building any. In the meantime there's somewhat of a worker shortage but I have about 1 every city so it should be fine once I build a couple more, maybe.

Anyone else playing through this to space?

Since I'm not a specialist on this, I'll just say that things are looking good. Knights need more love.

 

[Fish Man]

Mostly I mean the following responses you got to getting the fish-city up quickly.



OK, here it is different since no border pop is required (no wasted 30, then 10T waiting period). However, the point remains that improving seafood costs 30 (and you need to have a way to get them quickly) while improving a -resource with a worker costs nothing but worker turns. In your fish city you spent those turns building mines, which admittedly are somewhat useful later. But that's later, and right now we shouldn't care much about later. So while putting 30 to a work boat is productive, maybe some other site that is less draining to your empire is superior, since setting it up costs less and worker turns?

Sounds OK, if you whip settlers asap when available. Waiting for bigger overflow certainly just slows you down. I'm also assuming you have no granaries, because they are a big investment thus slowing down early expansion a lot, not to mention that the -cap is low, mitigating value of granaries. In cities with many grass cottages available, just growing to -cap and slow-building settlers/workers is a good option. Also depends how much chops available of course, you are correct in that a lot of attention needs to be paid on getting settlers/workers out.

This is also why I dislike your handling of northern seafood-city in your T54 save, it has drained a lot of worker turns (roads, mines), (workboats) and now needs to build a granary, only then it starts to contribute to your settler whips (whipping with a granary also makes those mines rather marginal)... There must be some other way. After putting all those resources into this city, I think the best is now to slow-build settlers/workers @4, because you will get them faster that way. I think even with a higher -cap (say we were CHA) I'd prefer this to whipping. EXP would change things certainly.

So, since it's a city that needs a lot of attention, I'm not sure if I would have prioritized it and certainly not wouldn't handle it the way you did. Helper 3E1N of cap could be better IMO (can also borrow corn, can work cottages, is connected), cow-horse could be prioritized... I know I'm ruthlessly bashing your save but I only dare to do that because I know you are a really good player. So while I'm at it, I dislike the road, the unworked cottages and the unchopped forests. 1.improve food 2.chop 3.cottage 4.connect in this order gets you really far.


Since I'm not a specialist on this, I'll just say that things are looking good. Knights need more love.

Thanks for the direct links, saved me a lot of trouble . Thanks for the advice too. And the critiques are no problem, it really got me thinking. Now that I ponder upon a bit more...city #2 did consume an ungodly amount of worker turns. But isn't fin coastal seafood and an extremely start to GP generation worth it? Don't virtually all other viable 2nd city sites require a WB too? And AFAIK, @Swordnboard settled the same way I did for the first 3-4 cities, and got away with it just fine?

The other thing I have doubts about is - slow-building settlers??? Really? I thought the rationale was to get cities up as fast as possible, and the benefits of that trump even working fin riverside cottages for a bit more in the early turns, only potentially being surpassed if you're really working a bunch of power tiles like gold, wet corn, or mined copper? And no granary? But food is king and granaries basically double it...

Anyways...t256 science victory. Only 7 turns off my goal ! That makes me sooo irrationally angry; there was nothing I could've done to shave off so much and yet so little time (still remain skeptical that even a better early game would've put me over 10k beakers and hammers ahead, and besides AFAIK I did everything perfectly after 1AD). If only I had MoM and was able to pop another golden age...but, ah well. Did all I could. Guess it still wasn't enough, though.