Great People Points: Focus in one city, or distribute across many?

[SniperRedFox]

Wow, is there a short summary for those of us who just joined this forum? (and cannot make sense of the graphs and vocabulary)

 

[Gumbolt]

I think the gist is have one specialist Great people farm (city) using lots of irrigation to allow more specialist such as artists etc. Using the caste civic to allow multiple specialists in your cities. Plus Pacifism civic to add 100% to your base rate Great people point per turn assuming you have the religion in your city that your currently using. other wonders too. Its been pointed out elsewhere in forum that adding great merchants to city give you +1 food and extra trade allowing a bigger GP farm. (320 points a turn is highest i have seen on forum so far)


I find if you can have a city of 16+ in size with a huge excess of food this will allow 7-8+ specialist. End of day this post seemed to be about the value of having a second/third great people farm. In my opinion its only of value if the second/ third or even fourth city producing great people points stands a chance of catching up to your specialist GP farm.

If city number 1 produces 100 Great people points a turn and city 2 only produces 20 points a turn it will take many turns to ever produce a great leader. If it ever does.
4 turns
100+100+100+100=400
20+20+20+20=80
After 9 turns
100+100+100+100+100=500
80+20+20+20+20+20=180

By time city 1 resets to 0 after producing a great leader and the required rate rise 100 to 500 city 2 now has 420 points to gain to reach the 500 mark and is not looking likely of ever producing a great leader.
If the second city produces 70 points a turn it increase chances of great leaders arriving. It appears the charts try to look at the value of additional cities producing more great leaders and what the optimum solution is.

I figure a city must produce at least around 50% or so of the leader great people farm you have to stand a decent chance of producing a few leaders in game. The 2nd GP farm needs to cover in excess of what the incremental rise in points required for the next leader is to stand a chance of ever producing a leader.

I hope that makes sense. Im no expert.

 

[SniperRedFox]

wow thx...

 

[floppymoose]

I have a problem with city specialization on civ in general.

I tend to take industrious leaders and build wonders. Wonders (Great and National) are one of the cornerstones of city specialization.

But Wonders are also a large part of the GP point supply, at least for folks like me who build lots of wonders.

And GP supply is maximized by putting all the GP points in one city, and building the national wonder there (which I already knew before I found this thread, but this thread reveals it objectively).

Hence there is a tension between building a wonder where it would be useful for city specialization, and building it where it will maximize GP points.

In practice I find that maximizing GP points performs better. At least for the frst 3 quarters of the game or so, which is usually what decides a game.

So the net effect of this is that I don't end up doing much city specialization, which rather detracts from the fun idea that the civ4 designers had for spicing up the game. Or to be more specific, I do one special city (my capital) that has gobs of wonders, the National Epic, and usually the Oxford University. The rest of my cities are fairly generic, although there will usually be a partly specialized city for cranking out military (Herioc Epic, West Point) and another for production (Ironworks) if my capital city isn't in a good production spot.

One possible way to fix this would be to simply remove the national epic altogether. If you do that, you aren't giving up so much when you spread your GP points around, as the first graph in this thread demonstrates.

 

[DaviddesJ]

So the net effect of this is that I don't end up doing much city specialization, which rather detracts from the fun idea that the civ4 designers had for spicing up the game.

I don't believe the designers especially wanted to encourage city specialization. I think they wanted to create a variety of viable approaches---which I think they did. You may find that specialization or non-specialization fits your goals better---it's fine either way.

 

[KrikkitTwo]

Well Rarely does a World Wonder contribute anything to city specialization. The only City Specific effects of a World Wonder are Culture and GPs
(the Great Library also gives some science to the city its in) so for a non Cultural Win, for max GPs, all World Wonders should be in one city. (and you don't give up anything for that.)

Now the National Wonders... well they contribute very little GP so they don't need to be in the GP city (although having one of the National Wonders that allows more specialists is nice there, and trhe one that boostsa GP is mandatory)

The thing is what specializes a city is the National Wonders, the Buildings, and the Terain Improvements, World Wonders really don't do anything.

 

[Shigga]

*pull*

VERY interesting read and confirming a gut feeling I had some time now. But has anybody checked the math for later patches and Warlords?

 

[Agramon]

After reading this thread, the philosophical trait seems pretty weak. If you should collect your GPP in one City, the "major" advantage of philosophical is reduced to one city. Not very much of my (usually not very large) empire.

 

[a4phantom]

I don't believe the designers especially wanted to encourage city specialization. I think they wanted to create a variety of viable approaches---which I think they did. You may find that specialization or non-specialization fits your goals better---it's fine either way.

The cap on National Wonders per city suggests to me that they wanted to strongly encourage city specialization.

 

[DaviddesJ]

The cap on National Wonders per city suggests to me that they wanted to strongly encourage city specialization.

I don't see that at all. The National Wonders are extremely powerful when combined, so, if you let them all be built in one city, people would tend to do that. Forcing them to be separated prevents them from being too absurdly powerful. This has nothing really to do with whether different cities are specialized or all do much the same thing.

 

[malekithe]

I would argue that specialization arrises naturally whenever there exists some degree of comparitive advantage in the realization of a goal. The simple fact that, in CIV, you settle your cities across diverse terrain leads to certain cities being better suited to certain tasks. The quantity of cities which you devote to any given task will be determined by the desired quanity of output of that task. The singular purpose of a national wonder is simply to reduce the number of cities devoted to a task required to meet a specific output goal (or, put another way, to increase output if resource devotion remains constant), allowing the marginal cities to pursue other goals. So, in a way, the simple inclusion of the national wonders (regardless of the 2 per city limitation) seems to indicate a desire to drastically increase the comparitive advantage of certain cities, reinforcing the idea of specialization. At the same time, the 2 per city limitation has a moderating effect, limiting the ability of any city to gain wide ranging comparative advantages across a number of tasks and forcing interesting decisions regarding the spreading of specialization among numerous cities.

 

[malekithe]

After reading this thread, the philosophical trait seems pretty weak. If you should collect your GPP in one City, the "major" advantage of philosophical is reduced to one city. Not very much of my (usually not very large) empire.

If Great People only benefitted a single city, you might have a point. But, as it stands, your whole civilization stands to gain from an increase in the birth rate of great people.

For instance, wonder building is often best confined to one or two very high production cities. Does this mean that the industrious trait is only adding any benefit to those two cities? Is industrious automatically a weaker trait than creative based on the simple fact that creative give a passive benefit to every city across the empire? Certainly not, as every city is able to reap the rewards from the additional wonders the industrious trait allows you to complete. In the exact same way, philosphical lets you transform a limited single-city bonus, into a civ-wide bonus.

This doesn't even address the fact that, in the early game, it is quite reasonable for a philophical civ to be running specialists in a number of cities, while the cost of great people is relatively low, speeding even further the rate of great person acquisition.

 

[frob2900]

From experience running dedicated production of great people (with a specialist economy, many large cities) I have to say that the assumptions of the above analysis are very weak, i.e. no skilled player should run cities at constant ratios, and therefore one should never find the situation described above..

One has to monitor the GPP in each city and assign specialists as necessary, sometimes even to the point of starving a city, so that great people appear in all cities..

I'd say that with a specialist economy ca 60% of the great people come from the GP farm in the long run, and having specialists in many other cities noticeably increases the total number of great people.

The analysis simply shows how a badly run civ would fare...

 

[malekithe]

The analysis simply shows how a badly run civ would fare...

While I agree with you that the original analysis was full of holes (it is quite old at this point), I disagree with the conclusion I believe you're attempting to draw.

I'd say that with a specialist economy ca 60% of the great people come from the GP farm in the long run, and having specialists in many other cities noticeably increases the total number of great people.

You mention the long run, but CIV is not about the long run. Why should I care that if I run 3 scientists in this city over here, in which I could be working cottages instead, 200 turns down the road, I'll get one extra great person a few turns sooner than I would have. That's a high short term cost for something that won't pay off till the very long term. (Note: If, instead, those 3 scientists are going to get me an extra great person soon, I see less reason not to run them. This is why specialists are much more powerful in the early game.)

No one will disagree that running extra scientists in extra cities will produce more great people. However, you have to look at the cost of those extra great people. A natural great person farm will have an easy time running lots of specialists due to numerous food resources and generally favorable terrain (high food concentration onto a smaller number of tiles). This, combined with the output-increasing national epic, results in a very high output/cost ratio for a basic stream of great people. Now, if I choose to augment this stream with a second city (even if that city is just as well suited to great person generation), it must be accepted that this will not double the rate of great person births. Hence, the output/cost ratio of your great people has been lowered across the empire. You are spending more resources, but not as efficiently as before.

A more realistic scenario is that any other city chosen to augment your national epic city will be inferior in it's ability to generate great people. At this point you have to look at what that city could be doing instead of producing great people. It's generally accepted that, absent representation and past the very beginning of the game, cottages produce more beakers than specialists (when Great People are excluded from the equation). With that in mind, it becomes a matter of cost-benefit analysis. Is the extra great person(s) sometime down the road worth a reduction in my present research capacity? This will depend upon a number of factors: the number of cities already devoted to GP generation, the quality of the land, the current stage of the game (cost of additional great people), and some others.

In the end, it's an interesting balancing act. I support a play style that spreads specialists out during the early portions of the game, but consolidates into a single great person farm sometime shortly after completing the national epic. The other cities, by that time, have transitioned back to cottages, as I prefer the increased research output (and other benefits) to the distant opportunity to get an extra great scientist.

You say, in the long run, 60% of Great People will come from the Great Person Farm. I say: A) What does that look like in the short term? and B) Wow, the whole of your empire outside your great person farm could only account for a measly 40% of total great people? Seems like a misappropriation of resources to me. Maybe they should have been doing something else with their time...?

 

[svv]

I used to advocate putting wall street and national epic together in a GP farm, preferably in a city with a shirne(s), where you build a ton of farms and maintain a huge bunch of merchant and priest specialists. You get huge GP points and huge gold.

The problem is, as I've increased in difficulty levels, I've found it difficult to keep all those citizens happy - unless you have globe theatre. Since I can't put all three of those national wonders (globe, national epic, wall street) in the same city, I have to choose how to split them up.

In my current game, my original capital was a pretty good gp city (decent food resources, some early wonders). I managed to start a religion in a city with good food potential. So, I ended up with two GP farms, my capital with national epic and wonders, and my money/religion city with globe and wall street. I think the capital ended up generating about 144 gpps, and the science/money city about 110.

I built a new palace in a commerce city, and this new capital/science city also ended up generating enough gpps to produce a gp every now and then as well.

So, I suppose if you were ONLY interested in GP points, you'd combine globe and national epic in your GP farm. But, it seems worthwhile to incorporate a money city.

 

[pigswill]

It's also worth considering that it takes time and resources to create a good GP Farm; relevant techs have to be discovered, wonders have to be built, the city has to produce food and grow to support specialists (and it can't run as many specialists while growing), once GP farm reaches health/happy caps buildings/wonders have to be built to raise caps (or additional units shipped in while running HR).

Therefore while the maths indicate a GP farm running eg 40gppt will outpace five cities running eg 10gppt you need to work out how long it will take your GP Farm to create that kind of output and whether its better to be running multiple GP producers in the meantime.

Whilist I'm not clever (or keen) enough to do the maths this suggests to me that early game (before GP farm matures) it will be stronger to run multiple GP producers (aka specialist economy), late game (with mature GP farm) its better to run single GP producer (and cottage over the rest of the cities) and somewhere in the middle it will become very confusing.

 

[Wodan]

What I do quite often is run GPP in parallel early game (i.e., SE), and as each city produces a GP, effectively "turn off" the GPP by sending in the workers and cottage over most of the farms. Early on, I choose which city will be my GP Farm, and I don't cottage that one over.

Very nice synchronicity in this strategy, and it allows complete customization as to what type of GP you get early and how many of each. e.g., you could run priests in just one city because you were able to found only one religion (with scientists in all the rest), or whatever is needed.

Interestingly, this whole strategy is more difficult to manage on lower levels. On high levels, you can't get a significant number of Wonders, if any at all. Wonders make the GPP income vary from city to city, can't have their GPP turned on/off and thus you have less control over maximizing your GPP resource production into GP, and also muddy up your GP types. So, on high levels (Emperor and above) it's easier to do this (takes less management).

Wodan

 

[oyzar]

This is obviously only if your not running an economy heavy based on specilists. One big problem for actually graphing up stuff like this is that the gpp generation doesnt start at the same point for all the cities nor do it remain constant over an extended period of time. Is it possible to make a threedimensional graph with one beeing how much gpp the primary city is generatng oin average then seeing how much GPP would be needed in the 2nd city to be benificial and in the 3rd city and the 4th city and so on? For any given value of city 1(say 40) you need an amount of GPP in the second city to produce at least 1 great person and then another amount to get the 2nd city to produce 2 and the 3rd city to get 1 and then 3,2,1 etc with how much reqruired for the extra cities depending on the average output in the maincity. Is this even remotly possibly to graph? From your analysis it seems that you never want to have more than 2 cities producing gpps but with philosophical pantheon and pasicifism is this true? Say you can run 4 specialists in up to 5 cities would this produce enough great persons or do you need more to be worth it? This analyses feature only very low amounts of GPP. This is what i mostly use in my games but i would still be very much interested to know at what point its worth to get up the 3rd city? or the 7th? or the 10th and so forth. Eventually the city wont produce any great persons and it is likely better to use the city for something else.

I hope i made sense and didnt come across as too harsh and it was indeed helpfull for somehwat understanding the mechanics for generating GP assuming low gpp generation.

 

[aryah]

This case assumed that there were no boosts to Great People Point production. Most boosts are empire-wide (Parthenon, Philosophical, Pacificist), and thus the d value is unchanged (all numbers get bigger, but the ratios stay the same, you have 60 points to distribute instead of 30).

However, the National Epic boosts Great People Points by 100% in just one city. Let's see how that impacts the distribution of points:

Perhaps Im misunderstanding something, but pacifism, philosophical are not 'ratio-neutral' at all if one is considering the impact of National Epic, but infact diminish its importance significantly; all these bonuses multiply the base 3 GPPs, so for instance someone non-philosophical & outside pacifism would be getting 3GPPs per scientist in other cities, and 6GPPs in National Epic city; ratio of 1/2. Someone both philosophical and in pacifism would be getting 9 GPPs in other cities, and 12 GPPs in National Epic city; a much better ratio of 3/4.

Another problem I see with this analysis if I understand it correctly, is, if it were done, for instance for beakers instead, it seems to me it would likewise come to the conclusion that any beakers produced outside the oxford city are just as wasted; for instance library, academy, university and observatory combined give 125% (or little less due to rounding down on each 25%) boost to base beakers, and oxford adds another 100% - so non-oxford cities would be getting just a slightly better ratio than, in the case of GPPs, non-national-epic cities of someone EITHER philosophical OR in pacifism have for them, and far worse than someone who is both philosophical and pacifist.

Yet, avoiding having any commerce cities besides the oxford one is hardly a good advice on how to play this game, for working many more tiles (i.e. with more cities, population) obviously more than compensates those not nearly as bad ratios..

 

[Roland Johansen]

Perhaps Im misunderstanding something, but pacifism, philosophical are not 'ratio-neutral' at all if one is considering the impact of National Epic, but infact diminish its importance significantly; all these bonuses multiply the base 3 GPPs, so for instance someone non-philosophical & outside pacifism would be getting 3GPPs per scientist in other cities, and 6GPPs in National Epic city; ratio of 1/2. Someone both philosophical and in pacifism would be getting 9 GPPs in other cities, and 12 GPPs in National Epic city; a much better ratio of 3/4.

This is correct.

Another problem I see with this analysis if I understand it correctly, is, if it were done, for instance for beakers instead, it seems to me it would likewise come to the conclusion that any beakers produced outside the oxford city are just as wasted; for instance library, academy, university and observatory combined give 125% (or little less due to rounding down on each 25%) boost to base beakers, and oxford adds another 100% - so non-oxford cities would be getting just a slightly better ratio than, in the case of GPPs, non-national-epic cities of someone EITHER philosophical OR in pacifism have for them, and far worse than someone who is both philosophical and pacifist.

Yet, avoiding having any commerce cities besides the oxford one is hardly a good advice on how to play this game, for working many more tiles (i.e. with more cities, population) obviously more than compensates those not nearly as bad ratios..

This isn't a correct analogy.

The difference between the great person generation and the beaker generation is that beakers are accumulated over your entire empire and then used to research a technology while great person points are accumulated locally and used for the local generation of a great person.
The amount of Great Person Points (GPP's) needed to generate a great person is a moving target. After the generation of a great person, the second one costs double the amount of the first and the third costs three times as much as the first. This means that a city slowly generating great person points might never reach the moving target. Each time another city is the first to reach the target and all of the great person points generated in this city are essentially wasted. They aren't reset to zero or something but they just aren't accumulated fast enough to reach the moving target. Of course, great persons also contribute production, gold, science and culture.

Oh, and welcome to civfanatics, by the way.