Positioning of wonders and GP generation

[pangu]

I have read that it is often better to have 2-3 GP farms than just 1, because 2 cities with 6 GP points per turn each will generate more GPs during the game than 1 city with 12 GP points per turn.

However, I wonder how that analysis will be affected if we take into account the NE. Is it best to concentrate all wonders / specialists in 1 city as far as possible?

Thanks.

 

[MrCynical]

I have read that it is often better to have 2-3 GP farms than just 1, because 2 cities with 6 GP points per turn each will generate more GPs during the game than 1 city with 12 GP points per turn.

This isn't the case - you'd get exactly the same number of GP in those situations. You might have slightly more control of what type of GP you get with GP generation spread over multiple cities.

However, I wonder how that analysis will be affected if we take into account the NE. Is it best to concentrate all wonders / specialists in 1 city as far as possible?

The NE tends to mean (particularly for non-philosophical civs), that many cities generating small amounts of great people points will not ever catch up with your fastest GP producer. They won't produce a GP in the time frame of the game, and so these points are wasted. Better to focus your efforts in the National Epic city, and maybe one or two others which can run enough specialists to occasionally give great people.

There is an argument for not stacking all your wonders in one city, as it makes it unpredictable which type of GP you'll get. Similarly you have a little more control if you spread GP production between two or three good GP farms.

 

[Leventis]

MrCynical said it quite nicely and points out the pros and cons to each argument. One other method I like to use sometimes is kind of a hybrid method; basically I have a main wonder building city where I can pretty much grab any wonder I like. However, instead of just building every wonder in this city, I am selective an will only build those wonders that give me specific GPPs. Usually, this means building wonders that give scientist, engineer, and merchant GPPs (like GL and pyramids) and avoiding others that give artist pollution in particular (parthenon etc). In this way, I am guaranteeing myself either a Great Scientist, Engineer, or Merchant (with ~1% chance of an artist once the NE goes there), which are probably the most useful Great People to get on a consistent basis. Any other wonder I want to get without polluting the GPP pool goes in some other city. I know other people around here also use this method very effectively. It gives a constant flow of GPs at a nice speed, whilst still maintaining a certain degree of control as to the type of GP produced.

 

[Gliese 581]

When you don't have the ability to run as many specialists as you would want, you will get a faster generation by using more cities. A typical example would be having libraries in several cities early game and running two scientists in each.
When/if you get caste system or when you've built the NE, this changes and it's usually better to use one city as a GP farm imo.
(This does not mean that other cities should completely stop using specialists, just that you should think of GP generation more than anything else in your GP farm and take other factors under consideration more in other cities). One of the strengths of running specialists is the great people generation and you need to consider that the gpp is potentially wasted if you have a strong GP farm that will "hog" all the GPs, you need to decide if it's worth running specialists in other cities based on their direct output aside from gpp.
Of course if you're running Representation under Pyramids you'll look for any excuse to run as many specialists as possible everywhere.