GPP City Specialization Strategies

[gentle]

How would you guys split the early and midgame wonders up ? Lets say...erm.up to scientific method ?
I think this GP point splitting is only important for this early/mid period anyway as I assume that once having reached Indutrial/Modern era the GP points needed to pop a new GP are too high anyway to rely on any GP based strategy from that on.Might be wrong here as I usually dont reach late game but if GP generation cannot be hold around 10-15 turns I lose interest in GP prodution at all.
So, what I would like to know is how many cities would you use and which city should hold which wonder (actually coming down to the question what type of GP should be focused on...hmmm...coming down to the question why to bother with other wonders-maybe except 1-2 useful ones).

[Aeson]

Originally posted by Generaldoktor Yeah, that's a case study for the premise from the CF thread I've been quoting. The conclusion being, which I don't disagree with, that if you want to get best utilization of the GP points generated by Wonders, build the Wonders in just one, or at most two cites; the second city presumably to siphon off GA and Prophet points as we propose here.

I don't agree that the best use of GP points from Wonders comes by sticking them all together. That results in "random" GP generation. Do you want your strategy driven by dice, or by intelligent planning?

(Massive) GP production is best when Specialists are in control. Specialists can be turned on and off, or reassigned as necessary. Wonders can factor in to give a boost, but if they mess up the GP type due to polluting the pools, they aren't worth it.

[Rancidlunchmeat]

Originally posted by Generaldoktor

Drama, as I mentioned above, I got with an otherwise superfluous GA. Quick, efficient, might not have done it otherwise.

I'm actually leaning towards drama for theatres to gain happiness. Theatres allow you to adjust the culture rate for increased happiness, and in order to get a religion you have to make it a primary goal early on, which this current strat founds upon.

Also.. religions give the benefit of only 1 happiness, same as theatres, but without the ability to increase that happiness later through spending. In order to increase happiness with religion, you'd have to use specialists or build cathedrals which will pollute the GPP, which I don't want to do.

Read a CivFanatics thread on this subject today and that insisted that National Epic, with its accompanying "pollution" was critical for a GPP city and there wasn't much sense in having more than one, or at most two.
<..snip..>

Yep, I read that thread and it's discussed earlier here in this one. In fact, the conclusions of that thread are a large part of the reason behind this one. Is it better to have MORE GP or the ones you WANT?

I agree with the Civfanatic thread and their conclusions, you will get more GPPs and more GP if you build everything in one place.

However, is that what is actually important? I'd rather have ONE scientist than TWO artists. So I make half the GPP and generate half the GP, but I that one GS is of far more value to me than two GAs.

Also, their mathematics on the National Epic are quite befuddled because they don't take into consideration the very question that we are asking: MORE but mixed or LESS but specific?

The National Epic will boost GPP tremendously. But it does so at a significant cost. And that cost is a huge dilution of the GPP by generating GAs.

With the strategy that I'm trying to formulate, the generation of even a SINGLE GA instead of the planned GS or GE would cause a tremendous setback. It's a gamble that can't be accepted. Like building a nuclear power plant... Everything's all well and good until the .5% chance of meltdown actually happens and then you realize the costs really do outweigh the benefits.

The CF thread did some kind of math analysis and predicted that mostly your primary GPP city is going to put out all the GP's because of the escalating point scale required for each subsequent GP and the other cities would never catch up, making them useless in that function. Of course this is with National Epic in the primary GPP city.

The last sentence is key. The rest of the conclusion is based upon that assumption. The other part of the conclusion is that each GP is equal to the others and just as important, regardless of strategy.. which is just simply false.

Also, they don't conclude that whatever city builds National Epic in it will always out-produce the other cities and essentially build 'worthless' GPP. But they did demonstrate that there is certainly a point where a city can generate all the GPP it wants and it will never produce a GP because they will never hit the breakeven point.

I don't know, it seems for every season, there's an opinion.

Actually, I'd say the Civfanatic thread is actually lacking of opinion. It does a great job of stating some slightly difficult mathematical concepts, but it offers nothing in the way of opinions. Unless you count their basic assumption that 'more of any type is better than less of a specific type' as an opinion.

(And it's really not. In fact, I think in that thread somebody brought up that very idea and it was never addressed.)

Maybe too complex a task for the Civfanatics, eh?

[Rancidlunchmeat]

Originally posted by couerdelion
GPP specialisation cities tend to rely on you having a non-capital GP farm. The simple reason for this is that you probably don't want to be appointing many specialists in your capital. For a start, your capital suffers the same health and happiness limits as your other cities and if you're running bureaucracy, the 50% bonus from tile working almost obliges you to use the tiles to the maximum. ......

I agree with most everything you've said here and later (I snipped for brevity's sake), but you are overlooking one of the key components of my particular focus which is GEs.

Because specialists come very late (the forge only adds one), you don't need to 'free up' alot of land for extra population to become specialists until the later end of the mid game.

I guess you could say that my second city, the GS city is really my 'GPP farm', but by midgame, I've got two.. one producing GS and one producing GEs.

[Rancidlunchmeat]

Originally posted by Generaldoktor

The religion came early by default of using a spiritual leader, who did not have to research Mysticism.

I just wanted to add here that having a spiritual leader doesn't guarantee you start with mysticism. Mansa Musa and the scrawny Egyptician chick are spiritual and don't start with mysticism

[Rancidlunchmeat]

Originally posted by Generaldoktor


Yeah, that's a case study for the premise from the CF thread I've been quoting. The conclusion being, which I don't disagree with, that if you want to get best utilization of the GP points generated by Wonders, build the Wonders in just one, or at most two cites; the second city presumably to siphon off GA and Prophet points as we propose here.

Actually, not quite.

Preferably, a THIRD city would siphon off any GA or GP or GM GPP that we didn't really want. That third city would really never generate any GP because it would never be able to catch up to the capital or the second ("true" GPP farm). So in that city, you are building wonders because the Inherent Wonder Benefit is large enough that you can justify their build without the added expectation of any contribution toward GPP production.

I can think of a couple of these... The Parthenon, the Great Lighthouse, Colossus, Notre Dame..

[Rancidlunchmeat]

Originally posted by Aeson


I don't agree that the best use of GP points from Wonders comes by sticking them all together. That results in "random" GP generation. Do you want your strategy driven by dice, or by intelligent planning?

(Massive) GP production is best when Specialists are in control. Specialists can be turned on and off, or reassigned as necessary. Wonders can factor in to give a boost, but if they mess up the GP type due to polluting the pools, they aren't worth it.

First.. let me say I'm not trying to spam the message board with posts, the multiple posts are due to the fact that I want to try and address everybody's points and I'd miss some if I didn't respond to each message individually.

Back to the point at hand.... Exactly!

In fact, in my last test game, my second city which was going to be my GS farm had a library and two scientist specialists. It was going to pop a great scientist in two turns. My GE farm (capital) was going to pop a GE in four turns. I simply removed the two scientist specialists, the city sat there with it's GPP intact, until my GE popped. Then, I re-assigned the specialists.

This is actually what has happened, typically, in my last two test games (except the first one I didn't remove my specialists and got an extra GS and missed on my GE).

I can usually get three GEs out before the GS farm starts kicking in and flooding the GPP pool with scientists. At that point, I won't get any more GEs until I get to factory or Ironworks or the GL goes obsolete (which ever comes first).

My real test right now is what to do with all those GSs... I actually end up getting "too many". Of course, the only reason I get "too many" is because my tech path is flawed and the techs they offer are either too far BACK down the tech path to make it a worthless use of a GS or too far UP the tech path to make them appear worthwhile.

But.. again, that's the fault of my tech path. It's not a flaw in the GS/GE or GPP city specialization strategy.

[fed1943]

Aeson,please:
Sure,to be able to chose the type of GP is an advantage;but,the weaker GPP cities will not produce GP.
So,there will be a ratio(maximum/minimum)between the cities produting GPP,to make it useful(the obvious,and may be ideal,would be 2 or more cities with the same number of GPP).
Do you want to teach us about?
Best regards,

[Hermann the Lombard]

Based on the strong differences of opinion here, competing GP strategies would make for an interesting AU course, maybe more so than making it a Devel's Workshop idea.

[couerdelion]

Originally posted by Generaldoktor

Well, opposing views are always welcome; you are, of course, disputing about everything that's been said in the thread so far

I rather thought that I had been raising a few problems with the question about using one GP farm alone to generate the GPs.

The simple reason why I don't think one farm is enough is that I rarely have enough food to support the specialists necessary to generate sufficient GPs. I do recall one game when I had four or five food specials so was able to generate 108 GPP/turn in my capital and I guess that would be what I would consider to be enough. But that was just one game and these days I'm lucky if I ever manage to find three food specials with a city radius.

The mathematics work if you compare, say, one city generating 25 GPP and another generating 15 GPP to just a single city producing 40 GPP. In this case, you'll get more GPs with the 40 GPP city. But would will also get more GPs in the first case that in a civ with just one city generating 25 GPP/turn. The argument that one city produces more GPs works if you are dealing with the same total number of GPP/turn. Put another way, I generate more GPs with several farms simply because I have more GPP/turn in total.

Regarding dilution I can certainly understand the case you are making against Great Artists. But I find that each of these specialists have their uses and personally have little problem with some dilution since I can usually find a respectable use for whatever the dice gives me. I may not get the best one but I do not consider the others to be worthless.

I actually think that you might even be overvaluing GEs. But I also think that most gamers here tend to place too much enphasis on pure production. For me the GE simply accelerates a wonder or allows me to build a wonder that I would not otherwise build. Few of these generate any great value since my key advantage for building a wonder is that I will have grabbed the necessary tech so far ahead that I can complete it before the other civs have a chance to rush build it with their GE. Let them waste the GE on Sistine Chapels and the like while I'm putting up my third Academy or second Shrine.

[Generaldoktor]

Originally posted by couerdelion


I rather thought that I had been raising a few problems with the question about using one GP farm alone to generate the GPs.

The simple reason why I don't think one farm is enough is that I rarely have enough food to support the specialists necessary to generate sufficient GPs. I do recall one game when I had four or five food specials so was able to generate 108 GPP/turn in my capital and I guess that would be what I would consider to be enough. But that was just one game and these days I'm lucky if I ever manage to find three food specials with a city radius.

The mathematics work if you compare, say, one city generating 25 GPP and another generating 15 GPP to just a single city producing 40 GPP. In this case, you'll get more GPs with the 40 GPP city. But would will also get more GPs in the first case that in a civ with just one city generating 25 GPP/turn. The argument that one city produces more GPs works if you are dealing with the same total number of GPP/turn. Put another way, I generate more GPs with several farms simply because I have more GPP/turn in total.

Regarding dilution I can certainly understand the case you are making against Great Artists. But I find that each of these specialists have their uses and personally have little problem with some dilution since I can usually find a respectable use for whatever the dice gives me. I may not get the best one but I do not consider the others to be worthless.

I actually think that you might even be overvaluing GEs. But I also think that most gamers here tend to place too much enphasis on pure production. For me the GE simply accelerates a wonder or allows me to build a wonder that I would not otherwise build. Few of these generate any great value since my key advantage for building a wonder is that I will have grabbed the necessary tech so far ahead that I can complete it before the other civs have a chance to rush build it with their GE. Let them waste the GE on Sistine Chapels and the like while I'm putting up my third Academy or second Shrine.

Whew, there's a lot to respond to here; I was playing my test game when everybody chimed in. I'm going to comment a little on this last one, but I think I'll defer mostly to RancidLunchMeat, who started the thread and is more than capable of answering everybody, to defend the GPP "specialization," which is the GP "strategy" that is being mainly pushed here.

I agree, Coeur, that there's use for the other specialists, though you and I might be alone in this. In my test game, which I may restart because of errors I made achieving "pure" specialization, I did pop a Great Merchant at a "polluted" city. That allowed me to send him successfully on his transcontinental trip and got me 5700 gold to upgrade my army to late Middle Ages, which will also make the transition to gunpowder much easier and more complete. I also now have a border/culture tug with Cyrus and so my next "superfluous" GA will be dropping a "culture bomb." I managed to get Divine Right (after half-researching it) with another one, got Drama with another and used a third on a Golden Age, so they're okay, sometimes.

That being said, I think some of the critics here are missing that you can really pump up GP points toward GP with these Wonders, to a far greater extent than what you could do with specialists from extra food alone, though this helps. This is a major side benefit of Wonders and I can't believe Coeur said they are unnecessary or icing on the cake. Try playing with none while the AI or MP are racking them all up.

Also, paying attention to which Wonders are going where, definitely gives you some considerable control over what kind of GP you pop and, as observed above, it appears evident to me you can pop a lot more, whether in two cities or three, with Wonder GPP. You may not have the same level of control as you do with your own selected (or de-selected) specialists, but the two techniques are designed to "supplement" each other , within this strategy. This does require a bit of micromanagement , which I know some people don't like and ran to Civ4 to escape. Well, you don't have to do this, but you just might win.

I'm a little lost with Coeur's argument about the two cities with the 25/15 GPP, versus one with 40. The research that's been done with this by the CF people or whatever, suggests you would seldom, if ever pop GP with the second city due to the escalating point scale requirement and the dominance of the first city. I think what RLM is trying to do with three; and this is why I have to start a new game, to try it; is get significantly more GPP out of BOTH of two cities and then have a third, accumulating useful Wonders that generate GA/Prophet points, as essentially a throwaway that would never produce.

Food specials are useful, but I don's see where you need three to generate enough loaves to get a healthy group of specialists, backed up by aforementioned Wonders, provided it is understood this is predominantly a GPP "farm, that will do little else. I'll have to go back and do a tile breakdown of my capital in the current game, but I've got the requisite three forests for health, with lumbermills; and a couple mines; this gives me adequate production for early Wonders, though I did finish some with early GE's, which is an advantage you have with this. Forge is there, of course, (built cheap as I was Industrious.) Everything else is river farms to promote excess pop to turn into specialists and I have one food special and built Hanging Gardens. It works for me. Of course, your Engineer specialists also generate production of themselves.

Oh, by the way, RLM; if you get too many GS's, just scatter Academies everywhere. You'll like it.

[Aeson]

Originally posted by fed1943
Sure,to be able to chose the type of GP is an advantage;but,the weaker GPP cities will not produce GP.

When you spread things out evenly, you don't really have "weaker" GP cities. They all will be more equal in GP p/t because of it.

Even "weaker" GP cities will produce some GP, especially if you can turn off your "stronger" GP cities when it is efficient to do so. If you have a city with X GP points in it's pool, and it just won't ever produce a GP because the limit is going up too fast... it's worth it to delay up to X GP points from other sources if it will allow those GP points to "count". (Especially if it's going to allow a pure pool of the GP type you want to count and is delaying a polluted one.) A good trick that Specialist economies can pull in this regard is to "save" GP points by building up excess Food while waiting for the "weaker" city, then starve the excess Food off by running extra Specialists for a similar number of turns. In some cases there is no tradeoff at all in doing so.

Third, Specialist based economies can be efficient even when the Specialists in most of the cities will not produce many (if any) GP. So even if you only get a couple GP out of a city, it can still have paid off because the Specialists used are competitive economically speaking.

If you have a "weaker" GP city that will never produce a GP, and you don't think the Specialists on their own are worth it, you are simply doing something wrong. Either you shouldn't be running the Specialists (thus it shouldn't be a GP city at all...), you should be setting the city up so the Specialists it used are worth it on their own, or you aren't managing the city and it's GP competitors right to take advantage of the extra GP p/t that are being produced.

So,there will be a ratio(maximum/minimum)between the cities produting GPP,to make it useful(the obvious,and may be ideal,would be 2 or more cities with the same number of GPP).
Do you want to teach us about?

It's not a zero sum game. The amount of GP p/t from a single supercity is not going to be (anywhere near) the same as the amount of GP p/t from an entire empire focused on a Specialist economy. (Which could also have a single Wonder supercity as well.) Saying you can put all the GP p/t in one city, vs 1/2 the GP p/t in each of 2 cities, is missing the point entirely. Depending on the map you may be able to have 5-50 cities at ~1/2 the GP p/t as the supercity.

The highest GP p/t city I've ever had was something around 300. Virtually all the Wonders, PHI, Pacifism, National Epic, The Parthenon, and several Specialists (though obviously not too many, as it was a major Production site, not Food site).

That's quite a few GP. It's also quite a few GP of the "wrong" type. Don't discount this factor. The things you can do with specific GP types at specific times is quite astounding. Getting 3 Artists and 2 Priests instead of 3 Scientists and 2 Merchants can mean using Ancient units instead of Grenadiers for 50+ turns on one path. A few Merchants, Priests, or Engineers instead of Artists at the wrong time in a Culture game can cost you just as much on another. The possibilities for targetted GP use are very interesting, and usually very fragile. I've personally had a game where Arists and Priests cost me 80 turns on a spaceship launch due to ~5% pollution from each. Luck can be a real bastard sometimes. I've also had a game where I had a city with most of the Wonders and over 250GP p/t, and I would have razed if I could have due to how much frustration it was causing.

Targetted GP for deep beelines into the tech tree is one of the most useful tools in SP. (It may just be me, but in timed MP I never had enough time to properly manage a Specialist economy to full use.) Wonder spamming will usually remove this option.

The fastest I've ever produced GP was with 30+ cities being run in a Specialist economy, all between 50-150 GP p/t. (It's not hard at all with Caste System, Biology, Mercantilism, and SoL to get 5-10 Specialists per city. That's 15-30 base GP p/t. +250% for PHI, Pacifism, and The Parthenon, for 52-105 GP p/t for a "basic" late game city. A really good city can get 15+ Specialists, and have specialized Wonders too.) Every single GP I got was the one I wanted. Including the ones used to get the necessary tech requirements ASAP. And I got more GP than a single city could hope to produce. (A very similar game, where I had built most of the Wonders in one city, was the one I referenced earlier. I wished I could have just razed the Wonder supercity because of the Artists and Priests it kept producing that didn't fit into my plans. It caused me to take longer to get to the tech requirements to run an maxed out Specialist economy, and then dilluted the benefits of doing so.)

[Generaldoktor]

I guess I'll give feedback about Aeson's comment without quoting, since his latest isn't far above; and he commented extensively.

1. I am concerned in a "full, specialist economy" that military and infrastructure are being neglected. It would seem that mines, mills, workshops, etc would still outproduce specialists, so if 30+ cities, or even 10+, are all food-heavy and relying on specialists, I would think one would be dwarfed militarily unless you drew a player personality mix that was universally benign.

2. How do you stockpile food and "stop" specialist (and associated GPP production,) for a time? Maybe I'm showing my inexperience here; I know there's a "stop growth" button, but does it stockpile food?

3. Since we just had an inconclusive discussion about how much food is enough to generate healthy amount of specialists, perhaps some instruction would be helpful on how one gets 15 or so specialists out of a typical fat cross. Some people are having problems with this; I thought I was doing good with six or so per.

4. At one point you dismiss high GPP concentration as automatically generating pollution, though I don't see how you'd get much of that if you watch the specialists and/or Wonders generated in a given location. At another point, you say in your empire-wide "specialist-economy," Wonder concentration and GP specialization, is easily achieved. I don't understand why this would be more so in 30 cities than in 3?

I guess we can expect our "emperors" to take a discussion to a higher level, I just want to make sure I don't get lost.

[Aeson]

Originally posted by Generaldoktor
This is a major side benefit of Wonders and I can't believe Coeur said they are unnecessary or icing on the cake. Try playing with none while the AI or MP are racking them all up.

In SP a lot depends on the difficulty level. On Deity for instance, you simply aren't going to get all the Wonders in one city. You can only grab a couple early Wonders, so it makes a lot more sense to make sure they are the "right" ones. Specialists will make up a much larger % of the GP p/t, so there's less of a tradeoff in splitting the Wonders up (by type, if necessary).

My experience with MP is much the same. You can't count on building too many Wonders, and will likely make yourself very vulnerable if you try for too many anyways. Specific Wonders that are central to your strategy should be targetted, and so again, Specialists can make up a much larger % of the GP p/t. As I said in my last post, Specialist economies are tough to get right in MP due to time constraints. Limiting the scope is definitely in order (and often forced by smaller mapsizes). I think that makes it more important to eliminate randomness in GP production though. With a smaller sample size, the GP pollution is more likely to account for large swings.

[quote]I'm a little lost with Coeur's argument about the two cities with the 25/15 GPP, versus one with 40.

The research that's been done with this by the CF people or whatever, suggests you would seldom, if ever pop GP with the second city due to the escalating point scale requirement and the dominance of the first city.

The second city gives a decent number of GP still. Don't scoff at an extra 2-3 early GP... that can mean 2-3 tech levels in some cases.

If you're starting at 0 with just a 25:

GP: Cost, Turn

1: 100, turn 4
2: 200, turn 12
3: 300, turn 20
4: 400, turn 36
5: 500, turn 56
6: 600, turn 80
7: 700, turn 108
8: 800, turn 140
....

If you're starting at 0 with a 25 and a 15:

GP: Cost, Turn, City, GP points (25:15)

1: 100, turn 4, to 25, 0:60
2: 200, turn 12, to 25, 0:180
3: 300, turn 20, to 15, 200:0
4: 400, turn 28, to 25, 0:120
5: 500, turn 53, to 25, 0:495*
6: 600, turn 60, to 15, 175:0
7: 700, turn 85, to 25, 0:375
8: 800, turn 114, to 15, 725:10
....

As you can see, by GP 8 the 15 city has cut off 26 turns in the GP generation, and that is accellerating. It has accounted for 3 out of the 8 GP. This is a little scewed towards whichever city would be producing GP points second (which could be either of them), but often the ability to start producing GP points coincides with having multiple cities available to start producing GP points. By the time a Wonder is completed, a Specialist housing building generally can be too. Caste System opens things up for all. And if you're relying on a certain building (Forge, Library) it will be available to start building in all your cities at the same time.

*This may be a good example of how "waiting" a turn could pay off, if you want GP 6 3 turns sooner, with a turn delay for 5 and possibly a turn delay for 7 onward. If this were a situation where the "weak" city would never be able to produce a GP later on it would be a no-brainer to get those GP points then probably take the "weak" city off of GP production unless it's best use is Specialists anyways, as it's trading a few for several hundred. If it would be as close at some later point though, you might still wait. But at some point you definitely want to "count" those accumulated GP points that are "close but not quite".

[Hermann the Lombard]

Aeson: Do you have an example saved game of a mature specialist economy with high GP p/t? I underuse specialists and it seems I can rarely support more than a couple per city. Needless to say, feel free to say "no."

Generaldoktor, I think the pollution primarily comes from Wonders produced in a convenient city rather than a planned specialized city, or from carelessly-assigned specialists...such as when generating "one free specialist in each city" and then forgetting to (re)assign them.