Need for Large City Games

[Öjevind Lång]

Unless you're a fan of the Heroic Epic.

Yep. In that case, build the barracks in all your cities while they are few, build the HE and then sell most of the baracks.

 

[Gaizokubanou]

Why? Am I saying something so totally unbelievable? All you have to do is play the game like pi-r8 here has. I told him the production rates were reasonable if he played this way, and he has seen it himself. What I say is obvious, if you play with Large Cities.

A screenshot doesn't help people play this way, though. Too, I'm having problem with my sched. I hardly have enough time to even play. There is no onus on you to believe what I say. If you prefer not to believe, feel free.

By the way, I didn't say that I didn't warmonger. I specifically said otherwise - that it would be necessary to take land from the AI at some point to get enough land to grow 10 size 20 cities in a reasonable time frame.

Yes.

There is no efficient way of conducting large city buildup in a game that you are still trying to win. People posted both theories and actual test games of why that is so, then you just either twist their wording or ramble on with your own theories that have no basis on actual game mechanics. For example, you held the view that science/production balance is fine, then when pi-r8 showed that science is indeed too fast, your argument was that the right way of playing would be to basically shoot ourselves in the foot by ignoring scientist specialists and other big game mechanics.

If you are going to essentially tell people to play inefficiently, what exactly are you asking here? How to efficiently play the inefficient style?

 

[ukrneal]

Yes.

There is no efficient way of conducting large city buildup in a game that you are still trying to win. People posted both theories and actual test games of why that is so, then you just either twist their wording or ramble on with your own theories that have no basis on actual game mechanics. For example, you held the view that science/production balance is fine, then when pi-r8 showed that science is indeed too fast, your argument was that the right way of playing would be to basically shoot ourselves in the foot by ignoring scientist specialists and other big game mechanics.

If you are going to essentially tell people to play inefficiently, what exactly are you asking here? How to efficiently play the inefficient style?

But I don't think this thread is about efficient play. Rather, it is how to play withand create big cities (efficient or not). Managing/Balancing science, gold, land, happiness, etc. is part of that, and if it means depressing something temporarily to play to the game style, well that sort of tradeoff is ok.

 

[aimlessgun]

Exactly. There is no way playing this way is currently optimal. We're treating it like a win condition in itself. Started a game trying this out.

 

[ropobop81]

I've been lurking as I'm more of a builder myself. Newer to the Civ series, I didn't play civ previous to the month prior to V being released, as Steam had a killer sale on a bundled IV w/ BtS. I'm up to Emp so far (158 hours played, some in multi/coop with friends), but haven't tried to push any further yet. Not sure I enjoy being behind in resources being the challenge, but that's another thread, back on topic...

Since we're trying to play a certain way that it is clear the game is not completely designed for, why don't we adjust the rules to accommodate things? I was thinking something like...

Conts/Pang/Small Conts
Huge Map
Turn off 2-4 AIs
Standard/Marathon Speed

I'm not a re-loader and I don't do the whole save at certain points thing, but if you start exploring and find you got boned with AIs too close or an unbalanced number then I guess restarting 20 turns in is better then 300-400 turns of frustration.

Last multi game I played gave me this idea as we played a huge/small continents map with all the civs turned on, but I was lucky enough to get a semi-isolated (same continent, but remote location) start and had the room I needed to get my three cores cities out and deep into the teens pop wise before I DOWed Monty for building within my 3 tile workable area. I didn't keep a turn count, but this was some time shortly after my Wu's UU came up as I had 2 out and defending my choke.

My theory is that we may not necessarily need to sacrifice anything, we may just honestly need more space between us and the AIs to make this work.

Orrrrr.....I've been drinking too much lately.

 

[Bandobras Took]

If you enjoy playing in a different way, that's fine, but you should keep in mind that a strategy forum for a new game is going to be full of players looking to share information towards the common goal of discovering the optimal way to play the game.

As for the Civ 4-Civ 5 comparisons, Soren Johnson came right out in the introduction to the manual and indicated that his goal was to maximize the number of interesting choices the player had to make.

Er . . . you do see the contradiction there, right? Now, tell me honestly, have you heard of these things in Civ IV:

The Liberalism Beeline
"The AI always goes for Engineering, so avoid that path and trade for it."
State Property in Vanilla
Slavery

The higher the level, the more your choices were constricted. In every Civ game.

EDIT: It should be obvious to you that making Science harder to get will make 1UPT a problem on high levels of difficulty.

Isn't that the point of high levels of difficulty? Making things harder?

Right now, you can outtech the AI and run a small force of units for defense.

Or undertech the AI and run Social Policies which increase combat strength in your borders.

Remove that advantage, and the player will have to build many more units in order to stay alive, soaking up the Hammers you intended to free by gimping Science.

Wait. So higher difficulty levels will be harder than lower difficulty levels?

This is the beauty of theory - it saves us time running experiments.

And is the lazy man's evidence.

The reason that they are useless right now is that they don't provide sufficient return on investment. It's too easy to shred through the end of the tech tree with Great Scientists and Research Agreements to achieve your win condition and start a new game, and a lot of players only find the segment of the game where the outcome is still in doubt interesting.

Nobody's been contesting that focusing on science (i.e. generating great scientists and bulbing techs) instead of hammers will get you science faster than you can build stuff. But that's equivalent to saying "I focused on wearing red today and now none of my clothing is blue!" It's totally true, but so what?

Besides, theoretically the outcome of the game is never in doubt, because it's too easy to shred through the end of the tech tree. See, I just proved that you should never play the game at all, because the outcome is never in doubt.

Alternately, you can make it much harder to burn through the end of the tech tree. Capping GS research by number of turns elapsed would help. This feature was in Civ 4 at launch; why is it not in Civ 5?

I agree. You can make it much harder to burn through the end of the tech tree if you want to. Don't generate a bunch of Great Scientists. Put down some manufactories on grassland. Keep your Civ going through multiple Golden Ages with Great Artists, or steal land from those City-States. The feature to make it harder to research is in Civ 5 if you want it. Remember that whole "increase the options for the player" thing you were talking about earlier?

You're nominally interested in playing optimally, but currently feel that the most optimal way to play is to use Great Scientists to make sure that your technology far outstrips your production? Isn't that imbalanced rather than optimal play?

 

[alpaca]

Agree with the "this is an end rather than a means" crowd because getting large cities is going to be less efficient and effective.

I've played a game a while ago with Persia (Emperor standard pangaea) where I, admittedly, didn't have tons of huge cities but I had something like 5 cities with 12ish population and capital at 20 when wrapping up the game at something like turn 220? (with a diplo victory beeline). I did conquer about two or three enemies and puppeted them but still had an opponent or two left, who were, however, quite behind in tech - and the later conquests were just for fun because I got bored of just hitting end turn.

Had I gone for a spaceship victory, maybe 50-70 turns longer, I would have gotten my capital to something like size 25 with a base production of 66 and a 90% modifier (factory + windmill + solar plant + hydro plant) in the turn 267 savegame I kept (after my win with most of the tech tree researched). I built mostly trade posts, not farms, so you might be able to grow it to size 30

The basic idea is to specialise your cities: Your big cities will be your science cities. They build and rush-buy science buildings. Your capital will be a hybrid science/production city, and you will need a couple of gold cities (for example puppets) and a few production cities to help wonder production and churn out units. In my case, I had two of these building units and sometimes wonders. Persepolis was building wonders practically through the game because the high-level buildings are much cheaper to rush-buy.

This is probably not quite what you're looking for because it was only on Emperor, Immortal would have been a fair bit harder and deity a lot (I was in a war against Alex & Monty at a time and lost one of my initial cities) and it was only a test game to see how well a strategy with a medium number of my own cities would work.

I think if I tried for large cities, I would go for a coverage-strategy: I would build my cities so they only have to claim their second hex at the maximum, helping each other out to avoid the huge culture need for the third ring. I would designate specialist cities and adjust tiles between the cities accordingly. I would have my science cities grow and groom and hatch them with rush-bought buildings. And I would grab every maritime CS I could get my dirty fingers on.

Edit: Pasargadae was actually a gold city but in the late-game it got some more science love, too. This was before the patch so the puppets did a fair bit of science.

Edit2: This was also one of the funny games where I rigged up a city state (Rio de Janeiro in this case) to take down a big nation civ. All Hail Brazil

 

[Martin Alvito]

The Liberalism Beeline
"The AI always goes for Engineering, so avoid that path and trade for it."
State Property in Vanilla
Slavery

As I already noted, your tech line was pretty well-defined. State Property was a choice. It was not a necessary one, and it was suboptimal if you wanted to push Science. Slavery was the single best mechanic of Civ 4, because it compelled interesting decisions. Which buildings, and when? Should you focus on Food or Hammers given this set of resources? You actually had to reason out production decisions. You wanted to whip every ten turns, except when you didn't, because Slavery was not a long-term Hammer maximizer. It just got the ball rolling faster.

This time around, the early game is well-defined. I know exactly what I want to build when and why. Altering the luxuries I get alters my tech path slightly. That's it.

The higher the level, the more your choices were constricted. In every Civ game.

This is called difficulty. As the game gets harder, you are compelled to play more and more optimally. Your choices should get restricted as the game gets harder. It would be preferable if a number of local optima existed, but that's very tough to design.

Or undertech the AI and run Social Policies which increase combat strength in your borders.

Why would you do that? That tree is terrible otherwise. You're strictly better off running a lot of cities and running SPs that synergize with lots of cities. And yes, that's a clear balance problem.

And is the lazy man's evidence.

I get a finite amount of time on earth. I prefer to conserve it, so that I can spend it on activities I enjoy. There's a cost to using theory, but even if I could know everything with perfect certainty (which I can't), I haven't got enough time to learn everything or run every possible experiment. Hence, theory.

Nobody's been contesting that focusing on science (i.e. generating great scientists and bulbing techs) instead of hammers will get you science faster than you can build stuff. But that's equivalent to saying "I focused on wearing red today and now none of my clothing is blue!" It's totally true, but so what?

You're missing the point, which is that the win conditions are further up the tech tree. If you want a Domination win, the best way to do it is outtech the AI and upgrade your existing units. Culture? Cristo Redentor is critical. Science? Strictly necessary. Ditto Diplomatic.

You have to get up the tech tree to achieve win conditions, and getting there faster is always more efficient than building stuff in the middle of the tech tree. It's a simple balance problem. If the buildings in the middle of the tree were good enough, it would be optimal to ease off the gas on Science and build them. But they aren't, so you shouldn't.

Besides, theoretically the outcome of the game is never in doubt, because it's too easy to shred through the end of the tech tree. See, I just proved that you should never play the game at all, because the outcome is never in doubt.

You can lose early on. If I get chain DoW'd by three AIs before I can kill one, I'm going down. It happens. But you're correct about the implicit balance problem - if I can achieve a clear winning position before turn 100 on Deity, something is wrong.

I agree. You can make it much harder to burn through the end of the tech tree if you want to.

Why would I gimp myself? You may find it fun, but I don't.

You're nominally interested in playing optimally, but currently feel that the most optimal way to play is to use Great Scientists to make sure that your technology far outstrips your production? Isn't that imbalanced rather than optimal play?

It's clearly optimal under the present rules set. Whether it's imbalanced is up to the observer, but it appears that we agree that it is imbalanced.

Last point - aimlessgun is right on target about the fact that we're having to treat this as a win condition. We have plenty of clear evidence that large cities are suboptimal, but that does sum up the problem neatly.

 

[Stanislaw]

Since the OP is looking for savegames/screenshots of 'Builder' type games in which one builds tons of buildings/large cities, I'll just dump these screenshots here and hope it's appropriate.

Huge Map, Marathon game. I think I have something akin to 40 cities and all of them should have at least 20 population. Cities building wealth have already finished building all buildings except Harbor/Walls/Barracks. The screenshots are pre-1.0.0.62 patch so wealth is only generating 10%.

Spoiler :

Spoiler :

 

[learner gamer]

Did you try to save up several great scientists, so that you can bulb the last 3 techs in quick succession? That's the key in this game to getting any specific tech quickly. For what it's worth, 1300AD isn't even all that fast- Paenblack posted a screenshot where he launched a spaceship in the 1300s, so I'm sure it's possible to get biology much sooner than I did.

Yeah, using GSs to bulb rather than build academies is now standard practise for me. Instead, in retrospect, one of my main handicaps has likely been that I’ve not been beelining the upper half of the tech tree quite enough. Ironically (given the subsequent posts) it looks like I’ve been doing something similar to Roxlimn: in my case, incorporating the occasional detour toward the bottom half of the tech tree so that I can get access to some hammer based improvements and hammer improving buildings. The problem I was finding though was that even this approach left me painfully short of hammers – by contrast I found teching to be much easier, partly because I’d been using GSs to catapult me along the tech tree. So when I saw Roxlimn (and others, since my kindly copied OP was actually my second request of this kind) mention that it was in fact possible to build a network of large, infrastructure heavy cities, I genuinely thought I was missing something; I thought that someone had at last managed to find a balance between hammers and beakers in the Civ 5 world that had so far eluded me.

Of course, this all takes me back to the very point that I made to Bandobras Took re: the alternatives that Civ 5 seems to put before the civver in my experience. The fact that Civ 5 makes me choose between alternatives is not the issue; the key IMHO is that Civ 5 seems currently to be making me choose purely between what I view (so this is purely my own opinion) as highly skewed (or extreme) outcomes: an economy with (i) lots of research and very little production or (ii) sufficient production, but very little research. What I was hoping this thread would demonstrate was that there was in fact a middle ground that I was missing – that it was in fact possible to make a more graduated choice and grow my civ along a less extreme path by trading away some of my research potential in exchange for some hammers (along the lines that Martin Alvito refers to). (Please note BTW that in saying this, I’ve not at any stage wanted an abundance of hammers and beakers in Civ 5...I understand completely that a great strategy game will ration both IMHO, the severity of the constraint increasing with game difficulty, lowering the amount of infrastructure that can be built accordingly.) What I’d hoped would result would be a demonstration of just how it was in fact possible to win a game of Civ 5 whilst constructing a network of large cities, building no more than appropriate infrastructure - as opposed to what I was seeing when I employed an ICS style game: winning by building a plethora of mid pop cities consisting of a skeletal infrastructure.

@Stanislaw: That’s exactly what I was hoping to see! Brilliant!

As you can probably imagine given my (kindly copied) OP, I have more than a few questions, given my failure to replicate your achievement to date. The most obvious one of course is what difficulty are you playing on? FWIW, I completely accept that my desire to construct infrastructure impacts the level at which I can play Civ 5; it’s a trade-off like any other. (That said though, playing at marathon may be a stretch too far for me...so I may well have to accept building less infrastructure if this was key to your game's outcome.)

In addition to your game’s difficulty, I obviously have questions re: what your early tech path and city builds were. Did you for example prioritise biology early as has been discussed in this thread – and then backfill the lower half of the tech tree? Were early city builds focussed around the food or tech enhancing buildings for instance? Or did you spam the gold enhancing buildings and mass buy your infrastructure? Did you go to war early? Did you beeline a particular set of social policies (eg. the order tree)? Did you adhere to the advice earlier in this thread re: farming riverside tiles - and then trade posting other (non-resource) flatland tiles? And so on. I presume BTW that you’re playing Civ 5 vanilla – or has one of the mods (that I’ve recently been contemplating using in cfc’s Civ 5: Creation and customization folder) helped you on your way?

Rather than go on listing questions, is there any chance that you could find the time to outline what you believe were the keys to your game please? Even if they were only set out in something like bullet point form (like they were in pi-r8’s excellent walkthrough), it would help me immensely. Moreover, judging by the ongoing debate that’s been taking place in this and many other threads re: the feasibility of a builder style approach to Civ 5, I really think it would help loads of other civvers too.

and once again, very, very well done!

 

[Bandobras Took]

As I already noted, your tech line was pretty well-defined. State Property was a choice. It was not a necessary one, and it was suboptimal if you wanted to push Science.

Ah, so playing optimally is a matter of first defining your goals.

Slavery was the single best mechanic of Civ 4, because it compelled interesting decisions. Which buildings, and when? Should you focus on Food or Hammers given this set of resources?

See, that's the problem. It wasn't which labor civic to use. It was "how can I best use Slavery?" That's not choice, that's number-crunching. "Is Slavery or Serfdom better in this situation?" Would be decisions and choices, not:

You actually had to reason out production decisions. You wanted to whip every ten turns, except when you didn't, because Slavery was not a long-term Hammer maximizer. It just got the ball rolling faster.

This is called difficulty. As the game gets harder, you are compelled to play more and more optimally. Your choices should get restricted as the game gets harder.

Then why are you complaining that large production cities get less viable on harder difficulties?

Right now, you can outtech the AI and run a small force of units for defense.

Why would you do that? That tree is terrible otherwise. You're strictly better off running a lot of cities and running SPs that synergize with lots of cities. And yes, that's a clear balance problem.

In other words, there is another option, but you don't use it.

I get a finite amount of time on earth.

May I suggest that if you're going to blow your finite amount of time on earth commenting on a video game, you at least run the experiments so you can comment well?

You're missing the point, which is that the win conditions are further up the tech tree. If you want a Domination win, the best way to do it is outtech the AI and upgrade your existing units. Culture? Cristo Redentor is critical. Science? Strictly necessary. Ditto Diplomatic.

You have to get up the tech tree to achieve win conditions, and getting there faster is always more efficient than building stuff in the middle of the tech tree. It's a simple balance problem. If the buildings in the middle of the tree were good enough, it would be optimal to ease off the gas on Science and build them. But they aren't, so you shouldn't.

No, you're missing the point. I'll clarify it for you as soon as you tell me what defines optimal play. Is it winning ASAP? Then you have no business talking about optimal play unless you play Settler duel maps. But if it isn't winning ASAP, then what is it?

Hence:

Why would I gimp myself? You may find it fun, but I don't.

Why is it a gimp to have good production? Especially if I still end up winning? You said that it was too easy to burn through the end of the tech tree, but nobody's forcing you to. In fact, you have the decision, which you were complaining about earlier.

It's clearly optimal under the present rules set.

It's clearly optimal under an artificial definition of optimal that doesn't actually square with the way you play the game.

 

[Stanislaw]

@Stanislaw: That’s exactly what I was hoping to see! Brilliant!

As you can probably imagine given my (kindly copied) OP, I have more than a few questions, given my failure to replicate your achievement to date. The most obvious one of course is what difficulty are you playing on? FWIW, I completely accept that my desire to construct infrastructure impacts the level at which I can play Civ 5; it’s a trade-off like any other. (That said though, playing at marathon may be a stretch too far for me...so I may well have to accept building less infrastructure if this was key to your game's outcome.)

This game was played on Prince difficulty, as I'm the type of Civ player that simply hates the AI 'cheating' in the sense they get unfair bonuses. In the end I don't think it mattered much since Persia and Egypt over on the other continent is just as strong as I am...

In addition to your game’s difficulty, I obviously have questions re: what your early tech path and city builds were. Did you for example prioritise biology early as has been discussed in this thread – and then backfill the lower half of the tech tree?

I don't quite remember what my tech paths were in this game... I think I typically filled the tech tree in order (cheapest first) and didn't focus on anything in particular unless there was something I wanted in the immediate vicinity. The only thing I do remember is teching straight for Metal Casting so I can unlock Workshops so I could get Great Engineer points faster.

Were early city builds focussed around the food or tech enhancing buildings for instance? Or did you spam the gold enhancing buildings and mass buy your infrastructure?

My early builds were in fact focused around Culture. There were several reasons for this. One, as the Songhai, the Civ I played in this game, I had access to the Mud Pyramid Mosque that replaces the standard Temple, only that it provides more culture AND doesn't cost any maintenance (and doesn't need the Monument as a requirement). Two, my cities were placed further apart in this game, and thus I needed my borders to expand as fast as possible so I could get access to important tiles faster. Purchasing tiles in a Marathon game gets expensive extremely quickly and I couldn't count on that too much. Henceforth, whenever I built a new city I'd typically start off with the Mud Pyramid Mosque, and if I had enough gold in my treasury I'd rush buy a Monument to boost it even further.

Did you go to war early?

Yes. Only because I was pretty much forced to, though. The Germans built a city right next to my Capital, and followed it up with a bunch of other cities to my immediate south, effectively boxing me in and cut me off from expanding. Got a bunch of Longswordsmen and put them out of their misery. Although it was an unwanted war, in the end I think it served me well. I captured about 10 cities, effectively doubling the number of my cities in my possession at that time (I didn't raze any).

Did you beeline a particular set of social policies (eg. the order tree)?

The first policy I unlocked was actually Honor. As the Songhai, and playing on a Huge map on Marathon speed, this turned out to be a superb choice. With Barbarian encampments revealed on map, and with the Songhai special ability of getting Triple gold per encampment, on Marathon speed each encampment I eliminated gave me a whooping 225 gold. And because there was plenty of free land available due to it being a Huge map game, every time I eliminated one another one would typically spring up, supplying me with a constant source of Barbarian gold. This allowed me to ally up with multiple city states extremely fast and spawned me a few Great Generals while at it too, giving me free Golden Ages that propelled the growth of my empire further.

After Honor, I think I went for Policies in Liberty, Piety and Patronage. Freedom was unlocked once I reached the Renaissance. Order is current one I'm working on but it wasn't made a priority.

Did you adhere to the advice earlier in this thread re: farming riverside tiles - and then trade posting other (non-resource) flatland tiles?

Sort of. I farmed every single available riverside tile, but on every non-riverside Hill I placed a Mine, on every Plain I placed a Farm, and on every non-riverside Forest I placed a lumbermill. I placed Trading Posts only on Tundras, Deserts, Jungles and non-riverside Grasslands. Worked well enough for me.

And so on. I presume BTW that you’re playing Civ 5 vanilla – or has one of the mods (that I’ve recently been contemplating using in cfc’s Civ 5: Creation and customization folder) helped you on your way?

It's Vanilla. No mods installed.

Rather than go on listing questions, is there any chance that you could find the time to outline what you believe were the keys to your game please? Even if they were only set out in something like bullet point form (like they were in pi-r8’s excellent walkthrough), it would help me immensely. Moreover, judging by the ongoing debate that’s been taking place in this and many other threads re: the feasibility of a builder style approach to Civ 5, I really think it would help loads of other civvers too.

and once again, very, very well done!

Like I said earlier, I don't quite remember what I did exactly in this game so I can't really give you a detailed play-by-play, but I can tell you what I think I did right. The key to managing this game I think was that I managed to unlock a lot of Social Policies because I had tons of Culture - which was accomplished in a better fashion than a Standard map size game because when a game is played under a Huge map size, each additional city increases Culture costs by only 15% as opposed to 30% by Standard size. Because of this, despite my Civilization having multiple cities, I managed to unlock Liberty (complete), Honor (Discipline, Military Caste), Piety (complete), Patronage (Philanthropy, Scholasticism, Cultural Diplomacy), Order (Socialism), and Freedom (Constitution, Free Speech, Civil Society, Democracy), which helped tremendously in managing my empire, especially Freedom when it comes to managing happiness. And yes, I realize I could have gone for a Culture victory, but this was a builder's game.

Another thing I did in this game different from others is that I tried to maximize Great Engineers as opposed to Great Scientists, and using Great Engineers I rush-bought a whole ton of wonders, saving me tons of hammers that allowed me to build other buildings I wanted to build (I only missed out on building the Colossus, the Hanging Gardens, the Great Lighthouse and the Great Wall). Another thing I did - which I still do in my other games is after I built the Monument (or the Mud Pyramid Mosques in this game's case), I immediately built the Workshop, and later the Windmills (if applicable) and Factories before I built anything else. Basically, I focused on Production whenever I could, and whenever my gold output started to look bad I used the city governors on well-developed cities to focus on Gold. Happiness never became a problem after unlocking Freedom.

Another 'trick' I employed during this game was that I accumulated 14 types of luxuries but purposely left out the last remaining resource within my borders. By doing this, I made all my cities want that single remaining resource to trigger the 'we love the king!' days and once they all wanted that single resource, I build the required improvement to trigger the 'we love the king!' days in all my cities. Once all my cities enter the 'we love the king!' day period, I send a military unit to raze the improvement, bringing me back to 14 resources. Once the 'we love the king' day wears off, all my cities will want that resource again because I no longer have access to said resource, only that I'll have access to it again once I rebuild said improvement. With Civil Society (Freedom) unlocked and the Hospitals and Medical Labs built, using this trick I triggered insane population booms that allowed my civilization to grow to high populations extremely quickly.

I know it wasn't much, but hope this helps.

 

[andrewlt]

I've done games with many large cities before on Chieftain and Warlord difficulty. It's not really hard, especially after I've knocked out most of the AI civs already.

Even without hospitals, having 5+ maritime cities in my pocket grows cities eventually. Add in granaries, windmills and such and cities will get to a decent size (around 30).

The funny thing with this approach is that almost all of my wonders are concentrated in my capital city and 1-2 cities that have relatively smaller population compared to my other mega-cities. I also stopped expanding at some point in the game so my teching ability didn't up unlike what ICS gives. Everything was puppets and a few annexes after some point. With factories and railroads, it helped me catch up on infrastructure.

Obviously, I don't play this way on harder difficulties. Global happiness and infrastructure costs just kill this type of gameplay.

 

[Martin Alvito]

Ah, so playing optimally is a matter of first defining your goals.

Sure. The game gives you four. The problem here is that all four are optimized by pushing Science aggressively. The returns on pushing Science are invariably better than the uninspiring returns you can get on Hammers in the midgame once buildings and units get expensive. You're better off pushing Science and upgrading what you already have militarily rather than building more units, you're better off getting up to Cristo Redentor before you spend Culture, and the other two wins require completing most of the tech tree.

I think that part of the problem is that you don't realize how fast you can get through the tree. In the last Deity game I finished, I had Globalization on turn 156.

See, that's the problem. It wasn't which labor civic to use. It was "how can I best use Slavery?" That's not choice, that's number-crunching. "Is Slavery or Serfdom better in this situation?" Would be decisions and choices, not:

Yes, Slavery introduced some micro, but there were also decisions over which building/unit to produce when, where to settle, and so forth. When to go to war with whom was also a question. The answers varied with your start. None of that is present this time. There is a roughly optimal building sequence, the only thing that should vary is which you build when, and the only thing that causes the build order to vary is the speed at which you tech. You should always go to war immediately to steal Workers, you should always use the same units to prosecute that war, and there aren't any meaningful choices about where to put your cities.

In other words, there is another option, but you don't use it.

Elementary game theory: options which are dominated by another option are not really options. Tradition and gimping Science will produce nothing that could not otherwise be produced just as quickly by pushing Science. Further, you get other tangible, positive benefits from expanding and pushing Science. It follows that you should always expand and push Science, and that to do otherwise is to deliberately play suboptimally. The only exception is an OCC, but you want lots of large cities, not one.

May I suggest that if you're going to blow your finite amount of time on earth commenting on a video game, you at least run the experiments so you can comment well?

I run the profitable experiments. There is no reason to go up to the top of the Leaning Tower of Pisa and drop balls of different sizes in order to discover whether two objects of the same shape fall at the same rate irrespective of mass. Someone already came up with a theory to explain why, and ran the experiment to confirm it. Unless you're doing it for publicity or personal amusement, replication is a waste of time at this point.

Transporting theory from one domain to another, when applicable, saves time and enables you to focus on the worthwhile questions. To paraphrase Newton, we can see farther when we choose to stand on the shoulders of giants.

No, you're missing the point. I'll clarify it for you as soon as you tell me what defines optimal play. Is it winning ASAP? Then you have no business talking about optimal play unless you play Settler duel maps. But if it isn't winning ASAP, then what is it?

Have you visited the Hall of Fame for Civ 4? It establishes a question to be answered for every possible combination of settings - how fast can you win? This is why the norm is speed, but speed is judged in the context of the settings. If you want to define an inefficient goal, that's fine. I've already suggested what I believe to be two possible approaches for achieving Roxlinn's goal (which won't lead to an efficient win, but is a meaningful goal because it is challenging).

Why is it a gimp to have good production? Especially if I still end up winning? You said that it was too easy to burn through the end of the tech tree, but nobody's forcing you to. In fact, you have the decision, which you were complaining about earlier.

This illustrates the earlier point - decisions which are dominated are not options. I cannot imagine a set of goals that is not aided by burning through the tech tree aggressively at the expense of immediate term production. Have you noticed that every major strategy out there right now revolves around a tech? Pure ICS wants the FP. The ICS variant that runs the Order tree wants Biology. I'll tell you that Education is the key tech in the game. The Rifle rush is pretty clearly the optimal pure warmonger path right now on a map of decent size, since Horses have a bad upgrade path.

The only constraint on pushing Science is that you do not get killed by an AI. But all that you need to do to ensure survival is pump out a bunch of cheap units early on when your Science options stink, then upgrade them as Science permits, so the constraint doesn't bind.

In any event, we're pretty far afield and learner gamer appears to be satisfied. If you want to talk further about how to achieve the goal Roxlinn posed, fine. If you want to start a balance thread in the appropriate subforum, we can move the discussion there. But this isn't the place to continue the line we're on.

 

[E66man]

Since the OP is looking for savegames/screenshots of 'Builder' type games in which one builds tons of buildings/large cities, I'll just dump these screenshots here and hope it's appropriate.

Huge Map, Marathon game.

So the big question here is how much of your city growth was spurned on by the Marathon setting. Is the scaling for city growth at that speed such that you think somebody could replicate your results in a game that has less than 1000 turns?

 

[E66man]

Another 'trick' I employed during this game was that I accumulated 14 types of luxuries but purposely left out the last remaining resource within my borders. By doing this, I made all my cities want that single remaining resource to trigger the 'we love the king!' days and once they all wanted that single resource, I build the required improvement to trigger the 'we love the king!' days in all my cities. Once all my cities enter the 'we love the king!' day period, I send a military unit to raze the improvement, bringing me back to 14 resources. Once the 'we love the king' day wears off, all my cities will want that resource again because I no longer have access to said resource, only that I'll have access to it again once I rebuild said improvement. With Civil Society (Freedom) unlocked and the Hospitals and Medical Labs built, using this trick I triggered insane population booms that allowed my civilization to grow to high populations extremely quickly.

I'm not sure if this type of gimmickry is even needed. In games where I possessed every luxury, I would get the notification that city X demands Y and the announcement that city X started WLtKD because I provided Y in the same turn. So a city's desire for a luxury is not predicated on the existence of ones outside your empire, they just get picked preferentially if they exist.

 

[aimlessgun]

Alright, first try of this is not going so great. Part of that is my pacifist rules (building all your cities and workers yourself sucks!). Anways, got to Hospitals without too much trouble, but I can't build them! Wonder if it's just being on crappy hammer land, but these babies cost a ridiculous amount of hammers and/or gold.

Anyone have any ideas on that? Should I hold off on a hospital beeline and try to set up workshops in my cities first? I saved SPs for Rationalism>Secularism and Freedom>Civil Society, would it be better to wait until Order?

I guess my other problem is that I severely lowballed the amount of workers I needed to build since I'm so used to stealing them, and now I have a bunch of un-improved land.

 

[pi-r8]

Alright, first try of this is not going so great. Part of that is my pacifist rules (building all your cities and workers yourself sucks!). Anways, got to Hospitals without too much trouble, but I can't build them! Wonder if it's just being on crappy hammer land, but these babies cost a ridiculous amount of hammers and/or gold.

Anyone have any ideas on that? Should I hold off on a hospital beeline and try to set up workshops in my cities first? I saved SPs for Rationalism>Secularism and Freedom>Civil Society, would it be better to wait until Order?

I guess my other problem is that I severely lowballed the amount of workers I needed to build since I'm so used to stealing them, and now I have a bunch of un-improved land.

I would definitely get workshops first, since metal casting comes so much earlier. Basically there's not much you can do though- hospitals cost more than most wonders of the world and medical labs cost even more. That's one of the big bottlenecks of playing this way. All you can do I think is get biology very early, and allow 20 turns to build them. Might be worth saving one SP so you can get order once you tech biology.

 

[Stanislaw]

So the big question here is how much of your city growth was spurned on by the Marathon setting. Is the scaling for city growth at that speed such that you think somebody could replicate your results in a game that has less than 1000 turns?

According to CIV5GameSpeedsXML, the game settings for Marathon setting is the following:

	<GameSpeeds>
		<Row>
			<ID>0</ID>
			<Type>GAMESPEED_MARATHON</Type>
			<Description>TXT_KEY_GAMESPEED_MARATHON</Description>
			<Help>TXT_KEY_GAMESPEED_MARATHON_HELP</Help>
			<DealDuration>90</DealDuration>
			[B][U]<GrowthPercent>300</GrowthPercent>[/U][/B]
			<TrainPercent>300</TrainPercent>
			<ConstructPercent>300</ConstructPercent>
			<CreatePercent>300</CreatePercent>
			<ResearchPercent>300</ResearchPercent>
			<GoldPercent>300</GoldPercent>
			<GoldGiftMod>67</GoldGiftMod>
			<BuildPercent>300</BuildPercent>
			<ImprovementPercent>300</ImprovementPercent>
			<GreatPeoplePercent>300</GreatPeoplePercent>
			<CulturePercent>300</CulturePercent>
			<BarbPercent>400</BarbPercent>
			<FeatureProductionPercent>300</FeatureProductionPercent>
			<UnitDiscoverPercent>300</UnitDiscoverPercent>
			<UnitHurryPercent>300</UnitHurryPercent>
			<UnitTradePercent>300</UnitTradePercent>
			<GoldenAgePercent>200</GoldenAgePercent>
			<HurryPercent>100</HurryPercent>
			<InflationPercent>10</InflationPercent>
			<InflationOffset>-270</InflationOffset>
			<VictoryDelayPercent>300</VictoryDelayPercent>
			<IconAtlas>GAMESPEED_ATLAS</IconAtlas>
			<PortraitIndex>0</PortraitIndex>

So seeing as Growth for cities being tripled, I don't think it being Marathon really mattered, although of course being able to not worry about beaker overflows as much on Standard probably helped my teching speed. What mattered more, in fact, was that I was playing under Huge map settings, prompting me to play against more Civs (more cities to capture) and also had more City States to work with. I think I was allied to something like 10 Maritime City states in the screenshotted game.

Also, I was simply 'playing' the game and wasn't really going for an optimization strategy, nor was I expanding really aggressively. Had one done so, I'm fairly positive another could replicate my results in around 700~800 turns easily.

I'm not sure if this type of gimmickry is even needed. In games where I possessed every luxury, I would get the notification that city X demands Y and the announcement that city X started WLtKD because I provided Y in the same turn. So a city's desire for a luxury is not predicated on the existence of ones outside your empire, they just get picked preferentially if they exist.

Oh... I wasn't aware of that. I didn't know how 'we love the king days!' worked after getting all 15 luxuries, so I had played it safe. But are you absolutely positive it works this way? Seems kind of strange to me.

 

[pi-r8]

Can you post more screenshots or a save from earlier in the game? Or just describe the general course of the game? I'm a little skeptical that those cities all grew big early enough to matter. I mean, in that screenshot it's 1937, you're getting 5000 beakers/turn and researching future tech. There's no reason for the game to still be continuing at that point, except to just play around in a sandbox with building cities.