1. We have added a Gift Upgrades feature that allows you to gift an account upgrade to another member, just in time for the holiday season. You can see the gift option when going to the Account Upgrades screen, or on any user profile screen.
    Dismiss Notice

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

Discussion in 'Civ4 Strategy Articles' started by Vol, Nov 22, 2005.

  1. TabascoBob

    TabascoBob Chieftain

    Joined:
    Sep 23, 2005
    Messages:
    90
    Location:
    Morris County New Jersey
    The consensus in the thread seems to be that the optimal strategy for generating great people is to concentrate on one city. Over the course of a game, this certainly makes sense. What about the first few though? For example, if you are like me, you reach a point where the happy caps and your economy require you to put the brakes on city growth and the research slider, and you can usually whip a couple forges/libraries/courthouses/temples (or, if you're lucky, a wonder or two.) At that point, using my current game as an example, I have the GL and a library in one city, a forge, the Great Wall and Museum of Maus in another, and a random-event-generated-settled-GS in a third city, who now has company from the library there. While I am running HR, I have more food than happiness in those three cities. [I am alone on my landmass and was first to CoL, but the nearest civs are Hindu, who are warring with the Buddhists so I am not ready to alienate the world by picking a religon.]

    After reading this and other threads like it, I am wondering if I am heading down the wrong road by having three cities generating similar amounts of GP points? One city is (roughly) 75+% GS (GL and a "working" GS, plus the GP gpp from the Moai Statues); one is a mixed bag (has generated a GSpy already from the Great Wall; now with GW and MoM, and a "working" GE (forge); the third is 100% GS (two GS from a library and a third a settled scientist from a random event.)

    I am going to pop my 2nd and 3rd great people shortly, after which I hope to be able to choose a religon and lift the caps a bit. From the math, on a purely great person generating standpoint, it does not seem to make sense to continue running those specialists in all three cities.

    Does it matter that I'm playing a Phil civ? I know the ratios don't change, but I will get to great people 4 and 5 sooner than otherwise. My original thinking was that given my fairly isolated start and my characteristics and circumstances (decent food, nearby calendar resources), having more GP farms rather than fewer would garner me more GP. Since that clearly is not the case, when do you close down GP farms 2 & 3?
     
  2. TabascoBob

    TabascoBob Chieftain

    Joined:
    Sep 23, 2005
    Messages:
    90
    Location:
    Morris County New Jersey
    OK, after a couple hours of play, I've generated a total of 6 great people (one GSpy, four GS and one GE, plus a GA from Music). Playing as England/Victoria, I'm generating 30 GPpts in one city, 22GPpts in a second and 18 GPpts in a third, plus running a spy specialist in a border city. I have the Parthenon, so I've got that and the Phil bonus working. The GSpy settled in my capital, each scientist built an academy, the GE rushed Notre Dame (good-bye happy caps!!) and the GA was used for my first Golden Age (allowing me to be first to Philosophy, Civil Service and Engineering- but I got pipped by a turn on the Apostolic Palace).

    The point here is, despite the math showing that the spreading of GPpts over multiple cities is "sub-optimal" in the long run, the fact is I've actually popped 2 each from three sources, right when I could benefit the most from getting that many. I also have three more in the pipeline, and with a little micromanaging, I'll be able to ensure that the weakest producer is next before I "shut it down" (that is, I'll only run specialists that provide ancillary benefits, not because they're going to produce another GP). That will leave me with two strong GP farms: one producing beakers from Representation and scientists (2 GL and 2 hired), the other (my capital) producing some beakers but also producing GPpts from Wonders and specialists (Spy, Engineer and Priest).

    I just can't bring myself to abandon either just now- maybe after Scientific Method, the GL city will switch over to production and I'll concentrate GP in the capital.
     
  3. fed1943

    fed1943 Emperor

    Joined:
    Nov 28, 2005
    Messages:
    1,185
    Location:
    Lisbon
    GPP should be considered as a part of your strategy.
    Are Wonders and/or specialists profits the important and GPP subsidiary?
    Or as many GP as possible?
    Or just the required GP type?
    Best regards,
     
  4. Roland Johansen

    Roland Johansen Deity

    Joined:
    Apr 29, 2003
    Messages:
    4,292
    Location:
    the Netherlands
    The idea presented in the first post is that it is quite useless to have a city try to produce a great person with 2 scientist while you also have a city producing great persons with 2 engineers, 3 scientists, 2 merchants, 10 GPP from wonders and the National Epic bonus to GPP. One city would be producing 6GPP and the other (7*3+10)*2=62 GPP. The one producing 6 GPP would never during a game produce a great person.

    If you have 3 cities producing a similar number of GPP's, then it is likely that all three of them will regularly produce great persons and you will also produce slightly more great persons in total than if you would only have one GPP producing city. Of course, it would be even better to have all of those GPP's in one city boosted by the national epic, but it is often not possible to feed all of the specialists in one city.

    As you add more and more cities that produce great person points, then it becomes more and more likely that several of them will never create a great person. Some of them will be chasing a moving target number of GPP's and will never reach that target as other cities producing great persons move the target faster than the rate at which the city produces great person points. Still, regularly the raw gold, science, production, culture or espionage points can be worth the effort. These are often less than what would be offered by improved terrain, but if you're out of improved terrain and still have food left, then it is a good option.
     
  5. popejubal

    popejubal Emperor

    Joined:
    Jan 13, 2006
    Messages:
    1,034
    Location:
    Allentown, PA
    @TabascoBob:
    For me, it depends entirely on the game. Chances are, though, I will try very hard to pop a couple of great people from different cities for my first few. Once I'm up and running with a significant GP farm, I'm going to push the GP threshold higher with new GP in that city faster than my smaller GP generating cities can keep pace with, but for the first few, I can push out 4-5 Great People pretty quickly across a decent sized empire and that can get me quickly to Liberalism/Astronomy/Chemistry/whatever. I find that a quick dash to Chemistry and then Liberalism is often possible so that I can grab Steel and go marauding across the world with Cannons when some of the backwards Civs will still have Longbows for quite a while.

    I find that it's not worth a major push for Great People unless I have a specific purpose for them since it's usually only 1 or 2 cities that are well situated for specialists without sacrificing too much else. Other cities can work a couple of specialists, but not too many because they don't have enough appropriate buildings. Also, most cities don't have enough food and population to run specialists and still work those cottages + mines.

    Having the list of what gets bulbed in what order is very helpful for that so that you can strategically avoid certain techs and make sure that your bulbs are appropriately directed. If you are not bulbing, then cottages and mines are probably better than specialists in most cities.
     
  6. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    This is basically the difference between a "parallel" SE and a "serial" SE. Whether you realistically expect to have multiple cities generate GP, or have one GP Farm that is going to outproduce most if not all your other cities.

    In a parallel SE, for example, you specifically do NOT want to build the NE.

    A parallel SE is particularly useful in the early game. Most SEs start as a parallel SE even if they intend to make a GP farm later. The only question there is how well you manage your GP generation so that you maximize your GP production without overly delaying the NE and other bonuses (and thus losing out # turns of benefiting from their multipliers).
     
  7. TabascoBob

    TabascoBob Chieftain

    Joined:
    Sep 23, 2005
    Messages:
    90
    Location:
    Morris County New Jersey
    Thanks folks. I appreciate the further feedback. @popejubal: point taken. I am at a crossroads on electing the strategy you describe, or going spaceward, and taking my (sweet ol' turn-based game) time doing it. Wodan, never had the opportunity to thank you for your thoughts but I have learned a lot from your posts; the distinction between parallell and serial is a turn of phrase that helps crystallize the debate in my head: thanks. I have specifically been holding off on NE until I had reaped all the "low hanging" GPs. (Because I was racing for MoM, I ended up starting the GL in my coastal MS city; now, I am really not sure I want the NE there... but I digress.)

    I'm tempted to do some math to see where my specific break point is but... maybe I'll just lay down until the feeling goes away and play later. ;)
     
  8. PieceOfMind

    PieceOfMind Drill IV Defender Retired Moderator

    Joined:
    Jan 15, 2006
    Messages:
    9,319
    Location:
    Australia
    Especially before I have the National Epic built, it's not too unusual for me to pop GP from 2 or 3 different cities. There is nothing wrong with this. But one must make a distinction between making multiple GPP producers via wonders or via specialists. The great thing about using a small city to generate a GP from specialists is that as soon as the great person is born, you can "turn off" those specialists and never try to produce another GP.

    The problem of having multiple cities generate GP disappears if you make sure each city stops producing GPP when they have built what you expect to be the last GP it will need/ever make. As soon as you get a proper GP Farm up, with NE and all, it quickly becomes obvious that your less-capable cities are not going to be able to produce a GP so they should already have been turned off.

    With GPP from wonders, it's quite possible their GP points will be wasted forever. But with wonders, one would hope you were building the wonder primarily for obvious benefit rather than the GPP producing quality. Obviously the wonder GPP is not going to hurt if it's going to be your GP farm or if you just need that first great prophet or whatever.


    I'm not sure I understand why you would specifically NOT want to build the NE in any type of SE game.:confused: Do you mean you should not go out of your way to get the tech that allows it? Because if you can build it, I see no real reason not to. Having the NE in one city is not going to slow the overall rate of GP acquisition. It can only increase that rate.
     
  9. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    (By the way, to remind everybody why this is an important issue... GP generation requirements goes up with each GP produced. So, if one city makes a Great Engineer for 200 GP, then the next GP produced by ANY of your cities will now cost 300 GPP. Thus, if a single city is making more GPP than another of your cities, there is some possibility that the first city will "outstrip" the second and make 2 GP before the second city will make 1. In addition, if you have multiple cities making GPPs, which is what a "parallel SE" is doing, then effectively you're spreading out GPP production so that you get more GP earlier. Having a GP earlier in the game means you get its benefit earlier... that Academy is providing it's bonus to science over more turns (and to boot earlier beakers are better because you get techs earlier), you get that wonder earlier, etc. So if you have one city outstripping multiple cities, this is basically removing your "parallel" benefit, which while it might get you more GP over the total course of the game is giving you them later in the game which is a considerable negative.)

    Consider that you might have 7 cities each generating GPP. Take one of those cities, probably the one which already has the highest GPP/turn (since it's your best candidate for GP Farm, so it's either a high food or high production city), and build the NE.

    What have you done? You have now added +100% GPP production in that city. Except... your 6 remaining cities (except those which were just about to generate a GP anyway) are now likely to never produce a GP at all.

    Let's take the worst case scenario. Say all 6 will not now generate a GP. So we have lost all those GPP and also we have changed their GPP production to an effective non-benefit. Effectively they now have 0 GPP production.

    As you state, which I agree with 100%, the "trick" here is to use some basic resource management skill and "turn off" specialists in cities which will not produce a GP. In fact, the real "trick" is to use a modicrum of foresight and planning, so that you recognize this way beforehand, and never bother to run specialists in them in the first place.

    (Unless of course you value them for another benefit, such as the beakers they generate, which is a different story. And most SEs will indeed value them for this, since scientists are probably the main source of beaker generation. But, tell yourself they are effectively making 0 GPP and you can make the correct cost/benefit decision for yourself.)

    Back to the comparison. We have now lost:
    1) X accumulated GPP across 6 cities
    2) Y GPP / turn across 6 cities

    While we have gained:
    3) Z GPP / turn in 1 city

    Is the loss of X and Y worth it to have Z? Depends on the exact numbers in question obviously.

    But your question was, "I don't understand why you would specifically NOT want to build the NE in any type of SE game." Clearly, if X and Y are not insignificant numbers (which they will not be in a good parallel SE), then the decision is anything but a sinecure. Certainly, it would be lacking to assume that it is ALWAYS a good idea, which seems to be what you're doing.

    In addition, we should recognize that a good player may well be generating GPP of different types. If that player switches from parallel SE to serial SE (with NE), then the ability to perform this is diminished. While a GP farm (especially in Caste System) can indeed switch from generating one type of GP to another (e.g., from scientists to prophets), this can drastically change empire-wide beaker generation, and also it doesn't work nearly as well for engineers (because only Forge allows early engineers so you can really only run 1 specialist and because caste doesn't work on engineers, so either way it doesn't really work).

    Anyway that's why.
     
  10. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    ps I think a lot of players aren't even really aware of this issue, so they don't really plan for it.

    It's kind of like how the Colossus can be a negative to a CE player, because if the CE player is working coast tiles, he's obviously not advancing his cottages. It's not a clear benefit and when deciding to build the Colossus some sort of cost/benefit decision should be made. But seriously, most players simply build the thing, thinking of all the added commerce they'll be getting. Just click to the city screen and you can see the governor switch from working cottages, hamlets, etc, and working the coast. Yes, you are now getting a tad more commerce from the higher raw commerce income of the coast tiles. But, you're delaying town advancement which is the core strength of a CE. Same thing with whipping... you're removing guys from working cottages. Are you SURE you want to do that? It's really not a clear-cut decision.

    Good planning in a SE will make a conscious decision to run a "parallel SE" or a "serial SE". And, good planning will decide long beforehand whether a parallel->serial switch will be performed later in the game (or whether to stick with one or the other from the get-go), and in fact, pretty much exactly when that switch will occur. Once you do this a couple of times, the math isn't even that hard.

    While this is somewhat of an advanced skill, it's interesting to me that many players aren't even aware of it. (At least that's my observation.) Especially players who are clearly skilled ones. Because, contrarily, those same players are the ones who recognize that GP generation is one of the biggest strengths of a SE.

    SEs are not simply an alternative to CEs. SEs have clear strengths over CEs, particularly in GP generation, ability to do early lightbulbs and get way ahead, etc.

    anyway I have to get to work. :)
     
  11. TabascoBob

    TabascoBob Chieftain

    Joined:
    Sep 23, 2005
    Messages:
    90
    Location:
    Morris County New Jersey
    Speaking of "lightbulbs", this thread helped one flicker dimly/briefly over my head. Spending some conscious thought and an iota of planning (not the norm for me), I have embraced the idea of grabbing the early cluster of GP from three or four cities, all of which are running specialists to control population growth until I can lift the happy cap, and/or generate beakers while the economy recovers and I can raise the slider again. Whipping those libraries, running scientists, and popping three or four GS in close succession-- I've never had the "problem" of deciding whether or not to settle a GS, build the third academy, or bulb PHIL before. And, once I get calendar resources or Representation/Nationhood, or get the military city online and can move obsolete units back to the core cities for HR-happiness duty, those specialists can go back to the cottages or mines or (horrors!!) under the whip for a forge or market.

    A very helpful thread to me, I appreciate the thoughtful posts.
     
  12. PieceOfMind

    PieceOfMind Drill IV Defender Retired Moderator

    Joined:
    Jan 15, 2006
    Messages:
    9,319
    Location:
    Australia
    This is what I'm trying to get at. You ask is the loss of X and Y worth it for the gain of Z? What I'm proposing (I'm a bit tired at the minute so I may be way off), is that in almost all but the most unusual of cases X+Y is going to be less than Z. It could be said that the value X - the loss of accumulated GPP across the 6 cities is not a true factor in the decision, because if you plan for a GP farm you could avoid the accumulation of GPP that won't produce a GP. The main one I was thinking about before was Y - the loss of the GPP per turn across the 6 cities.

    In the case where the 6 cities are producing GPP that will never produce a GP due to 1 city producing so much GPP as to outstrip the others, then the number of GPP in the 1 city must be huge. I haven't done the numbers yet but I'm guessing the GPP per turn rate must have to be significantly higher for it to mean it will keep overtaking the weaker city in GPP.


    I look at it like this. Suppose we have 7 cities all generating the same number of GPP per turn. What happens if you give a 100% boost to one of them (let's assume no other modifiers). Well, without the NE, a city would produce the GP on turn x. With the NE, that GP could only possibly have been produced on turn x or earlier. For the second GP, without the NE, it will be produced on turn y. With the NE, the second GP would be produced on turn y or earlier. And so on.

    So I think looking again at the numbers X, Y and Z, it's going to be typical that the value of Z is greater than X and Y.

    I had a feeling this might be the reason. I was hesitant to mention it because I thought it might be a fairly minor point. Mainly because with a GP farm it is often very realistic to be able to influence the type of GP you want to at least 50% or so. I find the production of engineers to be a very small benefit to sacrificing faster overall GP births empire wide. I guess you could say I'd rather have 1.3 scientists than 1 engineer, generally speaking anyway. Obviousy because of the randomness in GP births, this is not an exact science. :D
     
  13. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    Well of course you can do that. That would be planning for and implementing a serial SE.

    Your reasoning is circular. "I don't have to choose between X and Y because if I don't spend the cost for X then Y is always better."

    A serial SE has an opportunity cost, just as a parallel SE does. The serial SE has to give up the opportunity of having multiple cities produce GPP.

    It has to be roughly twice higher. Which, by the way, is the bonus provided by NE.

    Also recognize that we're not simply talking about simply outstripping the others. If the NE generates ANY great people at all then it is delaying the generation of gp from multiple other cities.

    It's not "the" reason, it's an secondary or even tertiary reason. I just threw it out there because it will be a concern in some games.

    Did you just make an assumption that a GP farm automatically results in "faster overall GP births empire wide"???

    That's exactly what we're talking about here. I contend that such is NOT always the case.
     
  14. fed1943

    fed1943 Emperor

    Joined:
    Nov 28, 2005
    Messages:
    1,185
    Location:
    Lisbon
    If one city avoids the others to pop GP, that just means said single city
    outputs more than all the others together,i.e., no loss of GP.
    Remember that when the cities output the same GPP the first one to pop
    also delays the others.
    The problem is when you want that type of GP first,then another type,and
    so on,let's say, a Prophet to the Shrine, followed by 2 scientists to bulb,
    a merchant to the food or corporation. Here, the parallel way shines.
    Best regards,
     
  15. PieceOfMind

    PieceOfMind Drill IV Defender Retired Moderator

    Joined:
    Jan 15, 2006
    Messages:
    9,319
    Location:
    Australia
    I agree - I think the reasoning is sort of circular. I guess my point was that there should be no need to have X as a factor for any SE economy if sufficient planning (and predicting the future) could be used. As it is, it would be tedious to try and predict at what points in the future each city would increase its GPP count so the parallel SE would be simpler.

    But in any case it's not the main point I'm trying to drive here.
    IMO the opportunity cost is better described as the specific GP the other cities can genereate with higher odds. If we assumed all GPs were of equal value, this cost would be 0. If I get all 5 GPs from one city in 300 turns, it's no worse than getting 5GPs from 5 cities in 300 turns. The lost GPP from the non-NE cities are completely made up by the gained GPP in the NE city.

    I disagree. For the NE city to completely outstrip all the other cities forever, it needs a very very large amount of GPP. There will almost always come a point where the amount of GPP catch up the NE city has to do is so great, another city without the NE will get to a threshold first.

    I tried out a four city model (spreadsheet) as follows: At normal speed, the thresholds are 100,200,...,900,1000,1200,1400,1600,... (effectively two joined linear functions).

    I'll label the cities A, B, C and NE. The cities start on the the following GPP on turn 1 (to make it easier to determine the GP birth for the three first cities we stagger the initial GPP a bit).

    Turn 1:
    A: 18GPP
    B: 12GPP
    C: 6GPP
    NE: 0GPP

    Each of the four cities earn 6GPP, except the NE city which earns 12GPP.

    With this setup, on the 482nd turn the 14th GP is generated. Over those turns, city A had 3 births, B 2 births, C 2 births and NE 7 births.

    If we now remove the NE so that the four cities now are identical apart from the starting GPP on turn 1, we get...

    On the 435th turn, the 12th GP is generated (13th would be turn 515). Over the 435 turns, each city had 3 births.




    The generation of GP from the other cities may be delayed technically but it's (IMO) the wrong way of looking at it. Those GP that are "delayed" are being born earlier because of the NE city. Having a NE city can only increase GPP generation.

    This means I would usually oppose the idea of avoiding the NE just to ensure these the GP production in these other cities is not delayed. As I was saying earlier, only the value of being able to choose which GPs are born can possibly outweigh the potential for extra GP births.


    As far as my reasoning at the moment will allow, it seems the only good reason.

    I didn't make the assumption, but a pseudo-proof of what would be called the assumption. If we are allowed to ignore the cost of researching the tech Aesthetics and the cost of building the NE, and even the fact we'll pollute the pool with the great artist(?) (they are small factors in the decision though) then there is no reason not to build the NE to increase empire-wide GP generation. Buidling the NE can only possibly increase the GP generation or leave it constant. That last sentence I feel can be proven and I did that reasonably explicitly in my previous post:



    This surprises me because I feel it's a fairly straight forward mathematical argument.

    Let's suppose that for the nth GP to be born empire-wide, one of our cities has to reach X GPP. Let's also assume that city A would have been the city to reach the X GPP and so produce the nth GP for our empire if it weren't for the NE. By adding the NE to our empire, it cannot possibly take longer to genereate that nth GP. Adding the NE to a city other than city A can only mean the nth GP may instead by generated earlier and by a different city.

    I cannot for the life of me think of any analogy, and at the moment, the precise form of a proof eludes me. If you're still not convinced, I'll try again to think of a way to prove it. I'm almost certain there must be a way to prove it by induction.
     
  16. Roland Johansen

    Roland Johansen Deity

    Joined:
    Apr 29, 2003
    Messages:
    4,292
    Location:
    the Netherlands
    While the NE will of course increase the GPP production empire wide, it will not in every single case increase the short term GP-birth rate. It will normally do this, but not always.

    Counterexample for the proposition: By adding the NE to our empire, it cannot possibly take longer to generate the nth GP.

    Next GP: 200 GPP

    City A: 192 GPP + 3GPP per turn (GP in 3 turns)
    City B: 165 GPP + 9GPP per turn (GP in 4 turns)

    Without the NE, city A will create a GP first, with the NE in city B in this turn, city B will create a GP first. To produce the great person at 300GPP the fastest it is better to let City A create the next one and then speed up the GPP rate in city B. This would happen if for instance city B got the NE in three turns instead of this turn. It's pretty clear that City B will get to the required 300 GPP much faster than city A, but by delaying the NE it got an approximately 200 GPP head start compared to not delaying.

    Of course, the same effect can also be achieved by switching off some specialists in city B, so it is not really necessary to delay the NE to get this effect. You just don't want to delay using the 200GPP produced in city A until the say 7th great person. (I didn't calculate when city A would get a GP when city B would get the NE now, it's an estimate.)

    The main idea is to use the slow cities earlier than the fast cities so that the conversion of GPP to actual Great Persons in these cities is not delayed. The slow cities will be ineffective/inefficient for producing great persons later in the game. You don't want to keep using large numbers of great persons spread over many cities for a long period of time without much result. We all agree on that. (except for the direct hammer/gold/science/culture/EP gain of these specialists).

    Note that the counterexample shown here is contrived. While it may happen once in a game, it will typically be better to get the NE out earlier and maximise the GPP generation in that city. It's a contrived counterexample, not the typical situation. However, you may encounter it in some games.
     
  17. PieceOfMind

    PieceOfMind Drill IV Defender Retired Moderator

    Joined:
    Jan 15, 2006
    Messages:
    9,319
    Location:
    Australia
    Ah yes you are correct Roland. I really should have added to my proposition (because I was sort of assuming it in my head :lol:) that the GP bars start at empty (or very near empty like in my example).

    So in the short term, in certain situations delaying the NE will allow for more GPs but in the long term getting the NE sooner is going to work out better.

    One thing I did in at least one game, was run a secondary GP producer into a food deficit for a few turns while I tried to get a GP out just before the NE got its next one.

    With the model I made above, I was also assuming a constant GPP rate for the 4 cities - this is another added limitation that is not necessarily realistic in-game.

    I'm still really struggling to see how a "parallel SE" could possibly encourage you to delay the NE deliberately, at least for any more than a few turns anyway.

    I could imagine the NE being much less valuable if there are several GP producing cities of similar output (eg. 7 cities all producing 6GPP). But usually it is possible to find a good site for a GP farm where running many more than 2 specialists is easy, and in this city delaying the NE more than necessary seems wasteful.
     
  18. Roland Johansen

    Roland Johansen Deity

    Joined:
    Apr 29, 2003
    Messages:
    4,292
    Location:
    the Netherlands
    Yes, that could prove valuable.

    I don't think I'd delay the National Epic often. The only reason would be to get a Great Engineer out somewhere. Even then, I might just build the NE and lower the number of specialists and use the extra food in the city for other purposes. You can view the NE as a way to double the efficiency of the food->GPP ratio. Why delay that?
     
  19. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    I agree.

    And, the converse is equally true: There is no need to have Y as a factor for any SE economy if sufficient planning (and predicting the future) is used.

    Either way, this reads to me as a non argument. If I plan for one thing, then there's no need to plan for the other. Well, duh. :)

    (Noting the assumption)

    Yes, agreed. That's the question here, though, isn't it. Can the NE city truly "make up for it" and how fast? The NE isn't even available until considerably into the game.

    I'll grant from your example that my estimation of "double" wasn't accurate. However, I think the example is not a good one for drawing other conclusions. Several reasons.

    For one thing, it doesn't track WHEN each GP is produced. Second, it assumes stagnant GPP production capacity. Third, it ignores the impact of civic choices and wonders.

    I ran some simultations on my own.

    I'll agree with your statement "having a ne city can only increase GPP production". I'll repeat my own which was "having a NE delays the generation of gp from multiple other cities."

    Also, I'll point out that building the NE has a hammer cost. At that point the game, a hammer cost that might be better spent elsewhere.

    (One strat I do quite a bit is run a parallel SE, and, once I get close to Emancipation, as each city generates a GP, I do a SE->CE switch. I keep a NE city as GP farm but otherwise am wholly CE.)

    Which was one of my points. As was the hammer cost of NE, and also the effect of civic choices.
     
  20. PieceOfMind

    PieceOfMind Drill IV Defender Retired Moderator

    Joined:
    Jan 15, 2006
    Messages:
    9,319
    Location:
    Australia
    Agreed. The reason I mentioned it though was that it seems both X and Y usually go together. It's possible to make both of them a non-cost with the right planning, excluding the GPP from wonders.

    I agree it's a good point that the NE is only opened up a fair way into the game. I would estimate that as long as it's coming in before about half of your GPs are born, it would be worth it. I've usually found that by the time I want the NE I often have marble available, and even without that, the NE is not too expensive a wonder. To be able to build the wonder, it's obviously wise to have a source of hammers (2 hills maybe) for when the building needs to be done.

    Well I had tracked when each GPP was produced but I didn't include it in the summary. If I get time tomorrow I"ll upload the spreadsheet, as there are a number of inputs that can be fiddled with.

    It assumes stagnant GPP production capacity, but I think that's not too big a concern if the GPP production capacity across the cities increases roughly at the same rate. I have a feeling that a gradually increasing rate of GPP would give more reason to build the NE, rather than less reason.

    I have to agree that things like the Parthenon and Pacifism and Phi leaders are going to make the NE bonus less game-changing. So the idea of a single GP farm I guess is more important in economies that have none (or few) of those other GPP modifiers.

    Ok ok but what is the point? I feel it's like saying that if two stores have a sale on - one has 30% off store-wide and the other has 50% off store-wide - and you go spend all your money at the 50% off store, then you are sacrificing the potential for any savings at the 30% off store. I know analogies like that can be annoying, but it's about the same level as how I imagine delaying a NE is.

    Right. This is something I do loosely as well. The decision to go for a single GP farm will happen around the time the NE becomes available. IMO this is not an argument for delaying the NE though. I absolutely agree that before the NE it's worth running a "parallel" SE but my original point was I couldn't see why you'd think too much about keeping a parallel SE going after the NE was available at the cost overall GPP production. I've seen a few reasons now, but some of them, like the hammer cost of the NE seem very minor. If you're going to build the NE eventually anyway, I'd almost argue the cost of the NE is not a real factor in the decision to delay.

    I would like to note as well, that normally I am going to be trying to produce scientists in the GP Farm, and whatever other GP types I consider most valuable. Even though the odds are less certain, and the GP pool is a little bit muddied from the NE and any other wonders, I still feel that having the NE overtake other cities multiple times in GP production is worth it because they are GP types I want anyway. If it were the case that the GP farm had to produce GP types that I really didn't want, then there'd be a real problem running a NE city. But this does not happen, in my limited experience.

    I have to agree the whole argument is a bit fuzzy, because you can't pin anything down into even a remotely simple formula. I would just be very interested to see a real game example where the decision to delay the building of the NE was a good move. I just had never thought of the idea of a parallel SE after NE before your post.
     

Share This Page