How do you reckon the GP Farm versus empire-wide GPP?

[Tristan_C]

It is technically more effective to produce Great People in one city and have all the other cities working tiles. The GP Farm is the linchpin of city-specialization. However, some of this game's popular (though debated) assets beg us to amass GPP in other cities. Of course I'm primarily postulating p-words,

Phi
Pacifism
Pyramids
Parthenon
​

And some big things are rendered more powerful if you choose to accrue GPP widely:

Mercantilism
Caste
The SE (in general)
SoL
​

Also, other cookie-cutter cities, which can produce "pure" GPP of a certain type, tempt us to farm GPP empire-wide:

Shrine - run 3 priests, bulb something or settle the prophet, then ditch the priests and go back to working tiles.
Ironworks - a wonderful GE pool, and a complete waste inside your GP farm. Am I forgetting something in saying this is your very first chance to have more than one Engineer in a city?
Conquered Wonders - we capture AI cities loaded with wonders sometimes. These will yield you a very cheap GP if you take appropriate measures.
NP - has the potential to give us a huge fricking BANG out of 20 tundra-forest tiles. So much potential here, it even tempts us to delay the National epic until we can stick it in this city. But why not? Too late in the game, of course.
​

GP farm, or wide-area GPP... we pretty much need to chase one or the other. If we flounder between them, or change strides, we're either not going to get many great people in our game, and/or the ones we get will be at extra-large opportunity costs. With a lot of waste. But just how strong wide-area GPP is *probably* depends on your geographical situation and the difficulty, so I was wondering what people's thoughts on for when they'd prefer it to a GP farm.

 

[Cam_H]

Tristan,

You make several good points, and it reminded me of the thread; Great People Points: Focus in one city, or distribute across many? that was started back in 2005 and discussion continued into this year.

With the introduction of Corporations (founded with GPs, Sid's Sushi / Cereal Mills boosting a city's food input) and the power of the National Park, the game expansions have new variables to the debate.

 

[DigitalBoy]

When I'm playing my favorite philosophical leader, I like to spread out GPPs between two (or maybe three) cities with scientists in the early game. It generates some great people in the period of game where spreading out GPP can realistically result in great people, instead of nearly getting to the threshold before the dedicated farm pushes it back some more.

By the time a dedicated GP farm is set up, I take specialists away from these improvised farms to go about their normal business. Even with Philosophical, half-assing GPP production in multiple cities doesn't stack well against dedicated farms.

As for how to leverage late game cross-empire GP bonuses, like Mercantilism, I've had no luck to speak of. Just count the free specialist in dedicated farms as a blessing and ignore the GPP accumulation in other cities.

 

[Artichoker]

I use multiple GP Farms regularly. It is quite possible to combine the "single GP Farm approach" with a "multiple GP Farm approach". Here are a few important elements of this approach:

1) Secondary GP Farms only serve a temporary role as a GP Farm, whereas the role of the Primary GP Farm is permanent. Changing game dynamics including switching into/out of Pacifism, and obsolescence of the Parthenon, are the main factors determining the lifespan of temporary GP Farms.

2) When in multiple GP farm mode, the secondary GP Farms should take a slow GPP flow approach that is accelerated at the very end of the stretch with a Golden Age and corresponding increase in the number of specialists in the city. They should not try to grab multiple GPs prior to the Golden Age, as this will make subsequent GP harder to get.

3) Careful planning should be made to trigger a Golden Age at the time when the greatest number of GP Farms (both secondary and primary) are ready to make a last sprint toward a GP. With correct timing, every secondary and primary GP farm should pop a GP by the time the Golden Age finishes.

4) National Epic is optional, depending on your setup. With the Parthenon, it becomes less attractive because the primary GP Farm already gains +50% GPP from the wonder. On the other hand, with weaker sources of GP in secondary GP Farms, it may be well worth it to build the National Epic, even without Marble.

5) Using anywhere between 2 to 4 GP Farms (including 1 primary and 1-3 secondary) is feasible in most cases.

 

[Wreck]

If you are philosophical, you should always consider multiple GP farms, especially in the midgame (when running Caste and Pacifism). But you should consider that even if you are not Phi.

In all cases, though, I would not advocate really "wide-area" GPP production. Rather, you always should have a GP farm (where the NE is), and then you selectively add other cities. In the long run, you'll have two GP farms: the old one, and the National Park. At various times on the way, you may manipulate a particular city to one-off a GP (usually after it generates a fair number of GPPs on its own, i.e. from wonders or a few scientists). Or you may run one or more cities as extra GP farms, for a prolonged period (if Phi and/or in Pacifism); but then will end after a point (when you move on from Pacifism, usually).

 

[mariogreymist]

I prefer to use multiple cities to produce my Great People in the early/mid game, but will usually build the NE in the most appropriate place thereafter. If possible, I choose a good NP site. But if I am behind in tech early, and unlikely to catch up without producing a great number of scientists, I aim for the GL/NE psychotic GPP pushing place.

 

[Woodreaux]

The only time I have a singular farm is when I'm playing an Industrious leader and I'm wonder spamming my capitol. If I'm not doing that, my normal practice is to run a bunch of scientists in my Oxford city, merchants in Wall St and engineers in the Ironworks. If I want a Great Artist for some reason, I run a bunch of artists in the Globe Theatre.

Those are late Renaissance / Industrial constructions so, before that I'll run a few scientists and priests here and there when the land and situation doesn't offer superior payoffs. I hardly ever found early religions, so I don't worry about amassing Great Prophets points for shrines, I just use the priests for the before merchants are available.

Now, later on, once the National Park comes into play, I use its specialist power to fill a gap. This is usually either a backup GE or a spy farm. Scientists and merchants are redundant because of Oxford & Wall St, while the max artists slots is 4 because National Park & Globe Theatre are a wasteful pairing (NP has built-in ). For these reasons, I usually focus it on spies or engineers.

I like multi-polar, or spread arrangements because allows me to control which type of . It also puts the specialists to work in a very useful fashion... not just accumulating points. The obvious drawback, is are more powerful the earlier they're born and this approach defers that advantage.

 

[slobberinbear]

I typically will only focus on GPP generation in 1-3 cities. Beyond that is a point of diminishing returns (not any/many more GPs), unless you are simply running specialists for the production.

I don't go out of my way to get/avoid certain types of GPs, either. I am happy to get what I get and find them all useful.

 

[iggymnrr]

Corps did some interesting things to GP strategies. I once built the ironworks into the main GP farm and used sushi to power 9 engineers to get mining inc. Of course 9 engineers in the ironworks city is very good too. Add a fair hammer base and mining inc and only a buro capitol would equal it. (Though another city might have been better for the ironworks per se.)

 

[Monsterzuma]

Scientists are a great way to make cities viable in the earliest stages of the game when forrests, culture needs and happiness restrictions make cottaging difficult. Whether I go with a GP-Farm approach or with a distributed scientist mini-farm approach depends largely on how many Libraries I "need" to keep my expansion going early on. When during expansion I end up seeding scientist farms throughout my empire, GP farm additions become almost redundant and great opportunities to skip a few techs (aes. lit.) arise.

 

[Tristan_C]

So, there is really strong support for running several GPP cities, a primary and some secondaries, swapping them at different eras. I've tried this before and was not very good at it, but I will grant that I need practice specializing my cities correctly to pull it off well. In my last several attempts, no city was ever able to catch up to the Epic without halting that city for a long time, which is wasteful. In my last game, I made an error with two golden ages; I didn't use the chance to catch up 2-3 GPP-producing cities to my NE city, and I reckon I could have done so. This meant I was set to get one GP shortly after the golden ages, rather than three.

The corporations sort of slipped my mind, since they come around even later than the NP. I guess the food really could come in handy for your GP farm if it's set up to get the type of GP you want, or otherwise, to bring one of the secondary cities into play. Modern GP's are different in that getting an unexpectedly-wrong type can make for a near-useless unit. I guess we could use the food corps to ward against that possibility.

In the best reasonable scenario, I could envision the Epic overtaking alternative GP farms continually if it had forests. In such a case we could employ this:

Capital has 2+ seafoods, forests, and 0-1 land resources. (Or something vaguely similar to that.) Use it as a whip city early, but do NOT chop. Under caste, place the NE there. Endeavor to move palace to a more useful locale-- such as to a conquered enemy's capital in the geo center of your civ.
At biology, build the NP in your ancient capital. Put sushi/cereal there when it becomes available. Zero-; forest-; plus a food corp; means that this city could have an incredible specialist population, though unfortunately, outside of caste I think it's really gonna hurt for hammers for the necessary buildings.

​

On the whole, though, I have to agree with all of you that the civics really want us to gather GPP's in multiple cities. 1-3 extras, as suggested. Besides geography, the civics really account for a great deal of this decision.

 

[TheWilltoAct]

--this city could have an incredible specialist population, though unfortunately, outside of caste I think it's really gonna hurt for hammers for the necessary buildings.

Well I suppose you would have the entire 'early' game to prepare the city for that setup.
Forests are good for production, you could run lumbermills then switch over to Forest Preserves later (even pre-build them I suppose, to be ready once NP is done).

 

[Artichoker]

So, there is really strong support for running several GPP cities, a primary and some secondaries, swapping them at different eras. I've tried this before and was not very good at it, but I will grant that I need practice specializing my cities correctly to pull it off well. In my last several attempts, no city was ever able to catch up to the Epic without halting that city for a long time, which is wasteful. In my last game, I made an error with two golden ages; I didn't use the chance to catch up 2-3 GPP-producing cities to my NE city, and I reckon I could have done so. This meant I was set to get one GP shortly after the golden ages, rather than three.

During a Golden Age, you can follow this rule of thumb: The GP Farms with the least accumulated GPP should get highest priority in grabbing the next GP, as a rule of thumb. This is because if you let a different city get it first, then it becomes more expensive for the city that deferred the GP, making it harder to get the GP in time.

 

[mariogreymist]

During a Golden Age, you can follow this rule of thumb: The GP Farms with the least accumulated GPP should get highest priority in grabbing the next GP, as a rule of thumb. This is because if you let a different city get it first, then it becomes more expensive for the city that deferred the GP, making it harder to get the GP in time.

I disagree. The weakest non GA gpp producers which can get one during the GA should be the priority. Those cities will never produce one unless they are given a clear shot to do so, usually by growing population or building infra in the better gpp cities. It can work (particularly if you have the MoM) to string a couple GAs together by popping the GPs which come from the weaker cities and still get at least one from all the best GPP cities as well, which will produce at a prodigious enough rate to get a GP fast after the weaker cities pick off the cheapest ones.

Or the shorthand: The best cities can afford the higher prices, the weaker cannot.

 

[Artichoker]

I disagree. The weakest non GA gpp producers which can get one during the GA should be the priority. Those cities will never produce one unless they are given a clear shot to do so, usually by growing population or building infra in the better gpp cities. It can work (particularly if you have the MoM) to string a couple GAs together by popping the GPs which come from the weaker cities and still get at least one from all the best GPP cities as well, which will produce at a prodigious enough rate to get a GP fast after the weaker cities pick off the cheapest ones.

Or the shorthand: The best cities can afford the higher prices, the weaker cannot.

Essentially the same as my answer, because I assumed that you won't be starting the GA unless you know that the weakest cities will be able to pop a GP.

As far as giving the weaker cities a shot, it helps a lot to run 2 scientists from a Library and use Pacifism. Then wait until the weakest city has about 50% of its GPP pool filled (or whatever it takes to make the GPP from the GA enough to pop a GP).

In the current game I'm on, the only specialist-related buildings in each city are a Library (for all GP farms) and a Temple (for the primary GP farm). At the end of the GA, the main GP farm (with the NE and Temple) is able to complete the 3rd GP of the GA, which is the 6th GP in the game. And this is with a single, ordinary 8-turn GA.

 

[mariogreymist]

Usually, the cities that get priority at the beginning of a GA for me are the ones that have been working two scientists from the library for some time, and are about half full, like you say. They don't have wonders, nor will they likely get any. Ideally they have a full granary, allowing me to starve them aggressively for the duration of the GA. When a GA starts, I go through my cities, and hire specialists until the time to shrink is one specialist short of what would cause lost population before the GA ends. If the GP will be produced before then, I keep that setting. If not, I work tiles. (I sometimes set it up to be done with growth the turn after the GA, so if I string a double, I can then aggressively starve that city in the second GA)

This is different from pushing the ones with the least accumulated GPP, as the one with the least GPP will usually be the one which generated the last GP (the one which started the GA), frequently a farm. Since the GP farm will continue to produce great people after the GAs are over, I don't overstress GPP generation during a GA, but instead work on infrastructure if it is needed, taking advantage of extra .

 

[Wreck]

During a Golden Age, you can follow this rule of thumb: The GP Farms with the least accumulated GPP should get highest priority...

I think it is much simpler than that. During a golden age, you should run as many specialists as food allows (and indeed, starve cities), to generate GPP in many cities where you can eventually get a great person. How many GPPs a city currently has is immaterial: only that you will eventually get a GP there. And when you get the GP is also immaterial, although as with any other resource, getting it done earlier is better.

Put another way: all GPPs converted into a GP are worthwhile and (especially early) very, very valuable. All GPPs which don't get turned into a GP are worthless, and (especially early) a significant waste of resources. Some worthless GPPs are unavoidable in any serious game of Civ4, but you control most GPP generation via specialist assignments, and therefore should have as a goal to get your ratio of GPPs not converted to GPPs converted as low as possible. (Another goal being, of course, to get many GPs.)

 

[Artichoker]

I think it is much simpler than that. During a golden age, you should run as many specialists as food allows (and indeed, starve cities), to generate GPP in many cities where you can eventually get a great person. How many GPPs a city currently has is immaterial: only that you will eventually get a GP there. And when you get the GP is also immaterial, although as with any other resource, getting it done earlier is better.

No...it's not that simple. Because aside from the +100% GPP bonus, a GA also provides a means to make free civics switches. I regularly use the GA to switch out of Pacifism without anarchy. This means that the cities with unfinished GP better have some kind of bonus leaving the GA (such as the National Epic), or else those GPP accumulated during the GA have a risk of being wasted.

With this approach, it is indeed highly desirable to arrange the GPP rate of every city before and during the GA so that all GP Farms (typically from 2-4 in number) are able to generate a GP by the time the GA ends.

After my 1st GA, I typically transition from a muliti-city GP approach to a single-city approach, unless I have the Parthenon or am using a Philosophical leader.

 

[Artichoker]

Usually, the cities that get priority at the beginning of a GA for me are the ones that have been working two scientists from the library for some time, and are about half full, like you say. They don't have wonders, nor will they likely get any. Ideally they have a full granary, allowing me to starve them aggressively for the duration of the GA. When a GA starts, I go through my cities, and hire specialists until the time to shrink is one specialist short of what would cause lost population before the GA ends. If the GP will be produced before then, I keep that setting. If not, I work tiles. (I sometimes set it up to be done with growth the turn after the GA, so if I string a double, I can then aggressively starve that city in the second GA)

This is different from pushing the ones with the least accumulated GPP, as the one with the least GPP will usually be the one which generated the last GP (the one which started the GA), frequently a farm. Since the GP farm will continue to produce great people after the GAs are over, I don't overstress GPP generation during a GA, but instead work on infrastructure if it is needed, taking advantage of extra .

Yes, and I should have clarified myself better.

When considering the city that has the least accumulated GPP, you should not consider those GP farms which have an exceptional GP rate, such as your primary GP farm with the NE. In these exceptional cases, you can start from 0 GPP at the beginning of the GA and still have enough GPP flow to pop a GP by the end of the GA.

My strategy for assigning specialists is a little different from yours, though. I also pay attention to the amount of available food, but I'm also careful not to make any GP farm fast enough to interfere with the GP progress of a different GP farm.

Say, for example, GP farm #1 is 40% full and GP farm #2 is 60% full, but they have approximately the same GPP rate. I'll make sure that GP farm #1 is able to get its GP as soon as possible, and I assign specialists to GP farm #2 so that it's scheduled to pop a GP one turn after GP farm #1 (while also considering high-yield tiles like special resources that can warrant delaying one of the GP farms by a turn or two.)

However, once GP farm #1 pops its GP, then GP farm #2 suddenly requires a higher price to pop its GP. So I crank up the specialists in GP farm #2 and all other GP farms after it, including my primary one. As each subsequent GP is popped in the GA, I follow the same process with the remaining GP farms.

 

[mariogreymist]

My strategy for assigning specialists is a little different from yours, though. I also pay attention to the amount of available food, but I'm also careful not to make any GP farm fast enough to interfere with the GP progress of a different GP farm.

My theory is to get as many GP points in as many cities that will produce GPs as I possibly can during a GA...and not just dedicated farms (of which I rarely have more than two). That said, I am willing to let significant GP progress in a city go to waste if I will have to slow GP production too much in my true farms, both because of the fact that 3 GPs from 2 cities in the next 20 turns is better than 4 from 4 in the next 40, but also because I don't want to stop working cottages in a GA just to make use of GPP earned during economic recovery (the main reason I accumulate gpp in cottage cities). There is a balance between making all the GPP count and maximizing GP potential, and working 2 farms aggressively is often superior to working 3-4 quasi-farms cautiously.