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. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    It's not so much whether it can be subjectively qualified as "expensive" as what is given up or delayed in order to build it. Everything has a cost. Even if you simply have the NE build units or Research, that's something you have to give up.

    That's an issue we haven't really discussed. For example, a parallel SE can spread out wonder builds to multiple cities, and make effective use of them. A serial SE, to maximize GPP production, has to delay wonder builds (i.e., has to queue them up) and this both risks losing the wonder to an AI and also delays getting the benefit of the wonder.

    This is where you say yes but this pollutes GPP type in the GP farm. Yes, and that is a reason why you may desire to build a wonder in a city you never want a GP from, period. Nevertheless, that is lost GPP and perhaps a tiny % risk of pollution is worth it to increase GPP production. Or, perhaps you're doing a golden age strat and thus any GP is put to good use. Or several other scenarios. Anyway.

    A parallel SE can focus different cities on different types of GP, and build appropriate wonders in each. GPP pollution can be controlled as desired, either to eliminate it or to choose to intentionally place it in certain cities. i.e., one city with 30% pollution but will only produce 1-2 GP over the course of the game may be more desirable than having a GP farm with 7% pollution and producing 15 GP over the course of the game.

    Yeah, that's certainly a big affecting factor. Choice of parallel or serial SE as part of overall game plan is part and parcel with what civics are desired to be run.

    No I think you misunderstood... I must have not said it clearly enough.

    Say in a parallel SE you have a city specifically producing prophets (for shrines) at the same time you have a GP farm. Delaying production of multiple prophets can mean a HUGE amount of lost gold.

    My usual problem is that I often HAVE TO HAVE a specific guy, say a merchant to found sushi. I have to "turn off" my GP farm (which is producing a huge chunk of my total empire reserach) in order to spit out that merchant.

    Well good I sparked a discussion then. I myself like game variety and experimenting different strats... I think it keeps the game fresh and interesting. Which is why I keep playing after what 3 years now?
     
  2. Roland Johansen

    Roland Johansen Deity

    Joined:
    Apr 29, 2003
    Messages:
    4,292
    Location:
    the Netherlands
    I wasn't really part of the discussion, but this is an element out of your post that I'd like to focus on. I often see players mentioning this, but I never use it in my games. I never have the luxury in a game to choose where I want to build a wonder. I almost always need to choose the optimal production city to have a chance to win the wonder race. And of course a city that is optimal for production isn't using specialists, it's using farms and mines and workshops. So this city won't be a good great person city. It likely will never produce a great person from the GPP produced by the later wonders.

    My skill is suited for playing a challenging game at immortal-deity level on huge maps (with random maps, not preselected), so I can see me adopting such a strategy at monarch-emperor level games where I can get seriously ahead of the AI in the technology race. But I can't see me using this at the really challenging games. Am I missing an element or is this strategy more suited for games where you get ahead of the AI in the technology race?
     
  3. obsolete

    obsolete Deity

    Joined:
    Dec 17, 2005
    Messages:
    6,201
    Location:
    Planet Earth
    You are making multiple mistakes by assuming specialists beat wonders for GPPs. Part of the reason is your looking only at the points of 3 vs 2. But wonders do much more than specialists.

    You pick your best 3 specialists... = 9 GPPs
    I pick best 3 wonders... (TGL, ToA, SoL) = 8 + 5 + 5 = 18 GPPs...

    I got you crushed 2 to 1 here...

    You want to go another round?
    Ok, lets add 2 more things into the mix..

    You now pick a total of 5 specialists... = 15 GPPs
    I add to my list Parthenon [2 + 50% (global total)] and Taj (2 + 100% total GA-Time)

    You still don't have a chance of catching up...

    Try again?
    Ok, you pick one more specialists (+3) = 18 GPP total only..
    I'll be nice and throw in a harmless Mausoleum... [2 + (All GA-factors * 50%)]

    You still wont beat me...

    Alright, lets open another slot... for each.
    You put one more specialist into the mix, (+3) = 21 GPP
    I put my first National Wonder, the National Epic... [1 + (100% * Total GPPs)]


    You still don't stand a chance, and never will...

    Getting near 15 is attainable at deity...just not always.
     
  4. Roland Johansen

    Roland Johansen Deity

    Joined:
    Apr 29, 2003
    Messages:
    4,292
    Location:
    the Netherlands
    You know exactly what I mean so there was no need for such an arrogant reply.
     
  5. PieceOfMind

    PieceOfMind Drill IV Defender Retired Moderator

    Joined:
    Jan 15, 2006
    Messages:
    9,319
    Location:
    Australia
    Yes but I also said that if you're going to build the NE at all it seems a non-issue. If anything, I'd say delaying the NE and building it later is the cost. If you are going to build the NE at all, generally it is going to be the sooner the better. If you're about to get DoW'd on or something then of course priorities are going to be a bit different. But I've never thought about possibly building Research in the likely GP farm when the NE can be built. Surely it's fairly reasonable to say the NE is a far far better return for the hammer investment, unless we are talking short-term enough that 5-10 turns is a big issue.

    It may just be the way I play but I don't let the GPP from wonders affect my GP strategies much at all, at least not after the first few GPs anyway. I don't build many wonders in the first place, but if I can I'd rather be placing them in only a couple of cities - the ones with either a lot of hammers or a lot of GPP already (eg. the GP farm).

    For me, the main way to maximise the benefit of every wonder is the the main effect of the wonder - not the GPP it generates. I have to agree with Roland's comment that if wonders can be built in a city of choice, and frequently, something about the game seems a bit easy. Having said that, I know I'm not much of a wonder builder anyway, and I don't play IND leaders very often I admit.

    Maybe. I find that example a bit odd because I'd take 15 of any GP over the 2 best GP any day.

    But this doesn't seem relevant. If delaying the production of Great Prophets means a huge amount of lost gold, then why not make the GP farm produce prophets too? If you did that you'd actually increase the number of great prophets born. Each great prophet that is lost due to having a GP farm outpacing it is going to be made up by the GPs the GP farm is producing. It would be questionable value to have your GP farm outstripping all your other cities with rubbish GPs like artists (or whatever you think is the worst in the particular game) but usually it's not too hard to ensure the GP farm is heavily biased towards the preferred GP.

    I really don't agree that delaying a few great prophets in one city is a huge loss, if another city is producing them sooner.

    I agree there are many points of the game where you want a specific GP. This is a situation where it would be better if you had produced fewer GPs previously, because each GP now can cost a huge amount of GPP (e.g. if we are talking the time of corporations). At this point, I'd be planning well in advance what GPs I want, and would probably time my GP births so that I was producing the desired GP one or two GPs before I actually need it. If I'm after a merchant, for example, and I can only get roughly 70% odds of one in the GP farm, then I'd reasonably prepare to get one or two non-merchants before I get the merchant. It depends how badly I want the corporation.

    :goodjob:
    Forgive me for this is probably a dumb question, but I have no idea what rationale you're using here. How do you get 18GPP from 3 wonders? Isn't it only 6GPP?
     
  6. obsolete

    obsolete Deity

    Joined:
    Dec 17, 2005
    Messages:
    6,201
    Location:
    Planet Earth
    Each World Wonder comes with a base of 2 GPPs, and then gets BONUS/MODIFIER points added with that base.

    TGL = Base of 2 + (2 x 3) = 2 + 6 = 8
    The reason for the extra 6 GPPs is because the running specialists it gives you also contributes.

    ToA = Base of 2 + 3 = 5
    Remember the free priest adds 3 GPP.

    SoL = Base of 2 + 3 = 5
    Remember, the free Specialist it gives you adds 3 GPP.

    8 + 5 + 5 = 18
     
  7. PieceOfMind

    PieceOfMind Drill IV Defender Retired Moderator

    Joined:
    Jan 15, 2006
    Messages:
    9,319
    Location:
    Australia
    Oh, for some reason I'm used to TGL meaning great lighthouse, and the great library is just GL.

    Anyway, those GPPs are coming from the specialists. Yes I know building the wonders is what grants the free specialists but RJ never said those 10 specialists were not already coming from wonders.

    In any case, your comparison is biased because you allow the parthenon to be included in the wonder GPP total but not for the specialists. At 5 specialists or wonders they would both even out at 30GPP/turn. And even then, you are assuming you'd get all 5 of those wonders lol.

    And golden ages are irrelevant, or at least you haven't included them fairly, because they would also multiply the effect of the specialists. And then you go an include the NE for your wonder count but not for the specialists. It seems to me you just wanted to drive home a minor point in the most pedantic, biased way possible. The calculations are correct but they are, IMO, almost useless because they do not even get near to describing typical gameplay.
     
  8. fed1943

    fed1943 Emperor

    Joined:
    Nov 28, 2005
    Messages:
    1,185
    Location:
    Lisbon
    My goal here is to learn, so please tell me where I'm wrong and why.

    The serial way never can lose to the parallel one from a quantity/speed point of view.

    Then, the question arises if a type of specialist is wanted. But, which one and when?

    By when I mean the beginning, middle or end game. Well, in the beginning I'm in parallel,
    that I want it or not. In the middle probably I can have Caste, also in the end until
    Emancipation is a must.

    And what type of GP? Scientist,merchant or artist? I would say no, as my one or two
    GPfarm can handle it.
    A prophet not in the 1st part of the game? For a late Shrine? Or?
    An engineer again not in the beginning? To bulb or rush? Corporation, yes, just keep one.
    A spy again not in the beginning? Hum... I give the word to warmasters.

    It looks the serial way has a price, but it's the way.

    Best regards,
     
  9. obsolete

    obsolete Deity

    Joined:
    Dec 17, 2005
    Messages:
    6,201
    Location:
    Planet Earth
    Well that would be a really silly cop-out wouldn't it?

    BTW, I noticed some people use TGLH for better distinguishment between the two wonders.

    Anyhow, specs require an extra pop-point (must be happy), whereas wonders require no food.
     
  10. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    Several thoughts. A GP Farm, you're correct, often is utilizing max food and specialists to get a lot of GPP. It can be difficult, especially on high level, to make any wonder in that city, period. Thus, on high level, you might have to make a choice to build a wonder in other than your GP farm, which means you are not maximizing your GPP production.

    Note that not all GP farms use high food and specialists. A wonder/production GP farm is very possible. It uses just a few specialists along with mines/workshops etc. You typically end up with more of a mixed bag of specialists, but this can actually be a good thing especially if you are using a golden age strategy.

    Anyway, you point out the exact negative of having a GP farm that I was responding to. The serial SE has to either give up on maximizing GPP production (and build the wonder elsewhere or don't build it at all), or it has to risk not getting the wonder and also delay wonder builds by queueing them up. (Which is my quote you were responding to.)

    Because we're talking about SEs, after all. Scientists are almost certainly your main source of beakers. And your GP farm is probably running a lot of scientists, often with Oxford. So, if you switch it to priests for a bunch of turns, that's a LOT of lost beakers.

    Yep. But don't you agree it would be easier to plan this in a parallel SE? You don't want that merchant for Sushi until pretty late in the game, so isn't it actually a benefit to be able to plan it out so your merchant-producing city generates him at about the right time?

    Instead, as you say, a serial SE has a GP farm and by the time it gets to the point of wanting him it can only make it come out to 70% odds, and it's just an iffy proposition.
     
  11. Roland Johansen

    Roland Johansen Deity

    Joined:
    Apr 29, 2003
    Messages:
    4,292
    Location:
    the Netherlands
    I can see that a great person factory based on production and wonders can of course produce the wonders and thus use their great person points in one combined pool, sometimes called a wonder economy. But you were talking about adding the wonders in the cities that are going for specific types of great persons in parallel. In that case, all of those cities would need to have a good production to be able to do that (or the difficulty level would need to be less challenging). And maybe even a good production is not good enough in a challenging game, maybe you need a great production. It's unlikely to have that many cities with a production as great as your best production city.

    So a game where you're able to build the wonders in the cities where their GP-points are contributing to the exact great person that you're trying to acquire in that city is likely not pushing your skills to the maximum.

    That was just the simple point, I was trying to make.

    Of course, it is nice when you're able to place the wonders exactly where they are performing the best. By the way, if these cities will at some point in the game stop producing great persons, then from that moment on, the great person points from the wonders are wasted. So pooling the GP-points from wonders in one big production city is often the choice that I make.
     
  12. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    A wonder economy is not quite the same thing as having a wonder GP Farm.

    Two things:
    1) Quite a lot of cities are capable of good production, they simply don't use it in most games. If you're doing a parallel SE then you can certainly work hills (where as otherwise you might work farms or coast) until the wonder is made. e.g., make the Oracle, and THEN turn on priests in that city.
    2) The "difficulty level" argument is staid and unfair. Everybody knows that the AI does not always prioritize wonders, either by beeling techs or by building the wonder as soon as it gets the tech. It depends on the wonder, and it depends on what AI leaders are in the game. It also depends on other factors, such as if the AI is at war.

    And the point is?

    We're talking about maximizing GPP production and about generating specific GP. Speed of building wonders is a factor but it's not the ONLY factor. Simply being able to point out a concern does not invalidate anything that has been under discussion. Heck, I've pointed out several concerns with serials SEs, but that doesn't mean a serial SE (aka "GP Farm") is a bad strategy.

    This conclusion is based on invalid assumptions / premises, so it does not logically follow.

    Bottom line, by no means do I advocate a parallel SE is better than a serial SE. I honestly believe its as good as. Which means it's an effective alternate strategy. A valid candidate in order to have good game variety, keeping the game fresh & fun.

    Another way to state what I just said is that there are pros and cons with *either* strategy. A GP farm is not perfect and has its costs and negatives. Likewise, a parallel SE is not perfect and has its costs and negatives.

    True enough. But, cons and negatives about your preferred strategy have already been noted in previous posts.
     
  13. TabascoBob

    TabascoBob Chieftain

    Joined:
    Sep 23, 2005
    Messages:
    90
    Location:
    Morris County New Jersey
    I've been following this thread with interest while finishing one game and starting another, all with the specific intention of improving the quality and number of GP generated. Without pretending to have followed all of the specific point/counterpoint above, it still "seems" that I can harvest GPs earlier, and get specific ones at that, by producing them in parallel. The fact that I've played PHI civs for the past couple games has helped generate these GPs that much sooner, so much sooner that it has made a big difference. Further, having a GSpy, a GP, a GA and 2 GScientists pop from 3 different cities within a few dozen turns of each other has been a game changer twice in a row.

    Now I'm "stuck" in a game with a turbo-charged Washington (capital city, Oxford, NE) running 5-6 GS, 3 GE and 3 GSpy; a New York City (with Wall Street) running 5-6 GM; and a captured capital (Berlin) that is food rich and well-populated, so it's running a motley collection of GP (I modifiy the "auto-selected" choices when I go through the City Screen). I'm managing the generation of GP in all three to make sure the 2 non-NE cities get theirs out.

    Obviously, this means that the "primary" GP farm is going to produce fewer total by the end of the full number of turns. I accept that. This also means that, in total, I will generate fewer GP than I otherwise would over the full number of turns. I also accept that.

    What I am unwilling to do is abandon my "pure" GM farm (which is my Wall Street City, which under Representation is a jewel) which will continue producing GM GPpts; however, it likely will never produce a GM unless I shut the other two off.

    In the captured capital, there's really nothing for all those specialists to do but produce maximum "GP" benefits, which include the production of GPpts (and, ipso facto great people. It's got about 10 specialists and (because of tile overlap) is otherwise a fine example of "full employment."

    In order to pop great people from the secondary GP farm, all I have to do is occasionally "turn off" the specialists in the primary. Right now, Washington has just produced its 7th or 8th; I've produced 5-7 others in 3 other cities. It may be that Berlin is getting to the point where it will never catch the NE city again, but it's really got nothing else to do but run specialists without completely reorganizing the nearby cities (which might be "better" but that kind of MM is not as much fun as marching off to war, YKWIS?)

    I guess the GPpts in the "second" city that get generated after a certain point, maybe as soon as the next GPerson, might be considered "lost". I'll have to look at that city to see if there is some minimal MM that will at least optimize what I'm getting, if I'm not getting another GPerson. For example, Priests are generating hammers and beakers because of a wonder and my civics; if I'll never produce another GPerson, then it doesn't really matter if I am running engineers or priests.

    However, running the GMs in my WS city are worth it regardless of whether or not they ever produce another GPerson.
     
  14. Roland Johansen

    Roland Johansen Deity

    Joined:
    Apr 29, 2003
    Messages:
    4,292
    Location:
    the Netherlands
    The discussion of serial versus parallel is a discussion you were having with PieceofMind, not with me. Maybe you have us mixed up or something? We were discussing something different. Or at least I was trying to discuss something different. I don't want to start discussing serial versus parallel as I don't have a really preferred method. It just depends on the game for me. I'll typically use more than one city to create great persons.

    I made the quite simple observation that if you're able to build a wonder with a sub-optimal production city, then you're apparently not really pressured in that wonder race. If you're regularly in that position, then it seems to me that you're not really challenged in wonder races at all. If you're not really challenged in wonder races at all, then it sounds like you are in control during that game. It typically happens when you're ahead of the AI or at least even with the most advanced AI's or have diplomatic means to control the flow of the game. It doesn't sound like a game where you're struggling to keep up in the technology race and are a second rate civilisation in the technology race for a large part of the game. It also doesn't sound like a game where large hostile civilisations are toying with you.

    So I concluded that picking and choosing your building sites according to the type of GPP produced by the wonder is only feasible when you're more or less in control of the game, not when you're struggling to keep up.

    As I said, it's not a remarkable statement.
     
  15. popejubal

    popejubal Emperor

    Joined:
    Jan 13, 2006
    Messages:
    1,034
    Location:
    Allentown, PA
    The Globe Theatre can turn that city into a :whipped: and drafting production powerhouse. Also, if it has lots of food (and if it's a GP farm, then I'm sure it does), then you can farm/pasture just enough tiles to irrigate all the farm resources and get all your bonuses and cottage/workshop/watermill the rest. Great GP farms also make outstanding production cities with Workshops/Watermills. When you hit the point where that second GP farm will not keep up with your primary, it's time to just run the specialists that are appropriate to the empire's needs and ignore the GP points.

    Just work useful tiles ahead of specialists and build improvements that decrease/don't increase available food unless it's a resource. Also, even without GP points, specialists are still good. They're just not as good as appropriately improved tiles.
     
  16. DrakenKin

    DrakenKin Prince

    Joined:
    Jan 8, 2010
    Messages:
    461
    Location:
    Montreal
    Good job, great topic. I only read the first two pages (for now) but I learned a lot.

    However I already started a game with 3 GP farm in mind so I will finish it, while tweaking my cities with what I learned here. Let me know if it makes sens :

    I will at first tweak my cities so they generate roughly the same amount of GP points. Example first city makes 20 GP, second 20 GP, and third 10 GP (with national wonder that's 20). So they never go above 1:1:1, thus the d ratio. The d ratio never goes below 1 in the chart, so I will be safe.

    Now if I look at the chart in this post, I see excellent news. It seems that as I generate great people, I get wider and wider margin of error. While at first I had to have a 1:1:1 ratio between my 3 cities, I can now have as much as 2:2:1... To accomplish this I will just have to alternate building wonders at each city, and use the specialists to equalize these ratios.

    The other good news is that I can get what I want. Right now my first city is getting engineers (pyramids etc), my second city is getting artists (Sistine, etc) and the third one located in a nice sea of grasslands is still in production and might go for artists too since I might go for a culture victory.

    That being said, I believe that the one city approach is NOT the ultimate answer, even if the data present here are very well thought out. There are a couple of simple, fundamental reasons for this :

    1) You WILL generate more GP points on multiple cities than a single one. You can dream up the perfect city all you want : there is only so much a single city can do. Again, there is a maximum GP output out there and you will hit it eventually. Since national epic basically gives you 2 cities in one, you'd need at least 3 GP cities to make more great persons than your primary, but I believe it is fairly possible to do, even consistently so. You will end up with more great persons in a game this way.
    2) Now that we know there is a ratio out there beyond which your other cities won't produce great persons, the only thing we have to do is stay under that threshold and we can successfully extract Great persons from all 3+ cities at the same time. We can even choose what to make!

    You don't even have to slow down a city to accomplish this, the only thing you need is to even out GP production among your cities.
     
  17. Wodan

    Wodan Deity

    Joined:
    Jun 3, 2005
    Messages:
    4,867
    Location:
    In transit
    Some misc. thoughts.

    GP earlier are better than GP later. Spreading out means you get your GP later in the game. Which means the benefit you get from them is later (e.g., you get an Academy X turns later and thus lose X turns of benefit).

    Devoting multiple cities, even partially, deducts from city specialization. Devoting one city out of my empire to be a GP farm is one thing. Devoting 3 or more is a big deal.

    You can have NE in one city only. So, spreading out means losing a good bit of "free" GPP.

    Whatever victory being pursued, all GPP in cities that don't produce a GP by game end, are lost. A multi-city approach will increase these lost GPP by a large amount.

    As a general rule, a multi-city approach works better early game, then loses most of its efficacy, for the above reasons.
     
  18. TheDS

    TheDS Regular Riot

    Joined:
    Jan 18, 2002
    Messages:
    1,454
    Location:
    Daytona Beach
    In my experience, even though I've built a ton of wonders in a single city, and have a ton of specialists in that same city (usually my science+capital city), and even have NE in that same city, there are other cities where I have lots of specialists (Wall Street + Shrine(s), Nat Park) and these generate enough GPPs, ESPECIALLY when I'm in a Golden Age (during which I switch to Pacifism for +200%) that I usually can get a GP or two out of each city. This is without even doing it intentionally. It makes a lot of sense to have 7 Merchants in Wall Street, and maybe a few Priests too. Am I going to sabotage my economy just to avoid GPPs? Heck no.

    That +200% boost (+300% if you're Phil) makes a huge difference in where your points get generated. These bonuses magnify the effect of Specialists much more intensely than they do those of Wonders. And it's because of that that I usually have 3 or 4 cities that wind up producing GPs, even though I'm focusing on building a Wonder city.

    In the early game, I could see a point to trying to have more Specialists in your Wonder city to boost your GPP, but never rob yourself by refusing to assign a Specialist.

    IMO.
     

Share This Page