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

[PTW] New City Build Order, Worker Mgmt and Railroads

Discussion in 'Civ3 - Strategy & Tips' started by HEEL_caT666, Mar 7, 2020.

  1. Aabraxan

    Aabraxan Mid-level Micromanager

    Joined:
    Jun 29, 2006
    Messages:
    5,297
    Location:
    Arkansas
    That certainly should be a good one. (It's early morning and I'm pre-coffee, so please don't ask me to crunch numbers.) The bottom line on settler pumps is that the town needs to grow by two in the same time it produces a settler. Ideally, that's a 4-turn pump, but it can also be a 5 or 6-turner. With a little fudging of the numbers, you can build worker pumps, too.
     
  2. SuedecivIII

    SuedecivIII Warlord

    Joined:
    Apr 20, 2019
    Messages:
    119
    I've got to disagree hard here. Growth isn't exponential. Your surplus food at size 1 is generally going to be equal or higher than your surplus food at size 5, at least until you switch out of despotism. So you won't cripple your growth by building settlers early.

    As for production: yes, floating low population levels will cripple your production. Who cares? Unless you're going for an early wonder or a military rush, production is irrelevant at the margin. You have all the production you need to produce scouting units and military police and the shields towards settlers and workers while you're waiting for your cities to hit size 3. What else do you need?

    (Granaries are obviously an exception, they increase growth, it's worth delaying a settler for a couple of them)

    What else do you need to build during the expansion phase that is so urgent that you will give up city spots?

    The advice just simply doesn't apply on standard size, 70% water maps. Maybe on very big or land heavy maps where the expansion phase bleeds into the next phase of the game because there's so much land.
     
  3. Lanzelot

    Lanzelot Moderator Moderator

    Joined:
    Jun 25, 2007
    Messages:
    5,758
    Location:
    Heidelberg
    Of course you are free to disagree, but I think the success of this strategy speaks for itself: I joined Civfanatics in 2007 and at that time I was a mid-level Emperor player (winning about 50% of my Emperor games and losing the other half). I participated in the GOTM competition and reached placements in the middle of the pack. At that time I was playing the same way you recommend here. Then I started analyzing the games of the top GOTM players, understood what they were doing differently and adopted this style of playing. By 2008 I was winning my first gold medals and consistently ranked in the top 5 of the GOTM ladder. (And at that time there was still a very stiff competition, with over a hundred active players and many world class players still competing, who unfortunately have meanwhile retired.)
    So it worked for me.

    First of all, I'm not giving up city spots. Quite the contrary: as I already said, with the above method I am able to settle a big stretch of land much faster than when I try to build settlers "early" (which in fact should be read as "late"...)
    The "build-settler-at-size-3" method may be able to settle the first 5 towns faster than the other approach, but by town 6 I will have caught up, and I'm pretty confident that by the time I reach 20 towns (plus have a strong core with Barracks, Libraries and Markets in the important non-corrupt cities), the other approach is somewhere between 10-15 towns and still has a weak core.

    And secondly: in some situations, depending on difficulty level and desired victory condition, it may even be wise to forego settler-building and let the AI do that for you...! How else do you think BC domination victories are achieved? The human player will never be able to build enough settlers to cover 66% of the world before 10 AD. But on the higher difficulty levels (let's say Emperor and above) you can use the production- and food-bonus the AI gets, and let the AI do it for you... The AI is programmed to cover every free spot, even if it doesn't get any benefit from it. So while the AI is busy settling the world for me, I just build a small powerful core asap (let's say first and second ring), then build nothing but Barracks & horsemen and then go and collect all the towns the AI has meanwhile settled for me... (If our tribe has a good early UU, like the Mounted Warrior or Gallic Swordsman, this approach is even more effective. For an illustration see my "Asterix" game, where I reached domination in 10 AD despite the fact that it was not a Pangaea, but a difficult Continents map, requiring lots of ship transport.) On lower levels up to Monarch this may not work, as the AI is too weak, and then the human player has to build more settlers himself in order to achieve a fast domination date.

    Here I have to disagree: especially under Despotism it is very seldom to have +5 food just from one single tile. The standard case is cow/wheat on grassland with irrigation, which is +4. So you need at least a second tile to get to +5. This means that (assuming the Granary is already in place) you need 3 turns to grow from 1 to 2, compared to 2 turns when the town is bigger!
    If your best food tile is only +3, the situation becomes even worse: you need 4 turns to grow from 1 to 2 and 3 turns from 2 to 3, resulting in 7 turns to gain two citizens, compared to the 4 turns you need to grow from 3 to 5, if you can work three +3 tiles.
     
  4. SuedecivIII

    SuedecivIII Warlord

    Joined:
    Apr 20, 2019
    Messages:
    119
    Obviously there are some times where that is producing settlers as fast as possible is not the best goal. Like if you're going for a 1 city cultural victory, going for a key wonder like the great library that will carry you, or as you mentioned, if your conquering. In those cases production is important.

    You are quite good at the game so there's probably something I'm missing here. What I'm proposing is aside from a couple granaries, you build settlers when you hit 3 population. This is the fastest way to produce settlers, if producing settlers is your main goal. What is the mechanism through which you catch up if you don't do this? If you let your cities grow to size 4 or 5 before getting out their settler, then you're giving up all the food that the new city would have produced if you planted it sooner (in addition to the other yields). You might gain more shields and commerce keeping the citizens at home for a little while, but you don't gain more food.

    What you mentioned about settler pump cities is an edge case, I don't think I'd have the production to pump out enough settlers to keep those cities at 1 pop if I tried. Realistically, aside from flood plains starts, 90-95% of the city spots available to you will not have access to more than one tile with 3 food in despotism (without build a harbor or chopping jungle). In those cases, sure, float 2-3 pop instead of 1, but otherwise floating a lower pop maximizes the speed at which you're claiming city spots.
     
  5. Lanzelot

    Lanzelot Moderator Moderator

    Joined:
    Jun 25, 2007
    Messages:
    5,758
    Location:
    Heidelberg
    Right. The nice thing about Civilization is, is that there is no "silver bullet" that applies to every situation. For example, everything I said in the previous couple of posts would be very wrong in a Sid level game: there you have to crank out the first 4-5 settlers as fast as possible, even if it harms all other aspects of your game. Because if you don't, the AI will have settled all spots around you and you will be playing OCC for the rest of the game (which fortunately will not be that long... ;))

    But the original poster stated that he was playing on the low difficulty levels (even as low as Chieftain), and I think for this level (and for pretty much all "standard" cases on the levels up to Emperor) the advice I gave will yield very good results and allow the average player to play a comfortable and successful game. Of course it needs to be adjusted to the given circumstances. (Both of us have already stated some examples where adjustment to the "general" strategy is needed: an important resource you need to get under your control, before the neighbor does, or a certain Wonder that is important for your plan for victory.) But the player will learn by experience after a couple of games, where such adjustments are necessary.

    I have already mentioned one case where "cranking out settlers as crazy regardless of the consequences" is not the most efficient strategy: when you want to build up a powerful army quickly and then simply capture the towns that you did not found. (To put it in a nutshell: if you build the settlers, you will have the towns. If you build horsemen, you will eventually have the towns and the horsemen... :D -- At least in some situations.)

    But let me also comment on another important factor of early gameplay, that we have not yet discussed. Perhaps this is, what you have been missing so far: the Despotism penalty.
    The Despotism penalty hurts your empire really hard: less of everything (though mainly food and commerce, since early in the game you probably have not yet built any mines on hills/mountains whose production would then be affected by the Despotism penalty), and on top of that, this "less" that you have is then reduced by the higher corruption rate of Despotism... :D
    Therefore it should be a top priority to get rid of that penalty as fast as possible and reach a proper government. Now for doing this, you need to research fast, and for this you need a high commerce output. And here we have two goals that more or less contradict each other: the goal to expand quickly and the goal to create high commerce early. Why is that? Because of three factors:
    1. If the central core towns constantly operate at size 5-7, they generate much more commerce than if they would constantly operate at size 1-3. (This is true on the lower difficulty levels, were 4 citizens are born content, but even on Emperor, where you only have one content citizen, it is true, if you create the necessary military police and hook up one luxury. But even if you have to increase the luxury slider, this statement is true, if you have enough river tiles or other 2-commerce tiles that the extra citizens can work, because it takes only 1 commerce to keep that citizen content via the luxury slider!)
    2. When building a settler in the capital, you basically trade two citizen (that may harvest you 4 gpt) for one insignificant town that might only contribute 1-2 gpt. (At least until it has grown and developed a bit, which may take a long time.)
    3. And finally: corruption again! The strategy of "growing tall" instead of "growing wide" has the advantage that all of your population is concentrated closely around your capital, keeping corruption low. Consequently, if we compare two empires that have exactly the same amount of population, but one has a few size 5-6 towns around the capital while the other one has lots of size 1-2 towns spread out further away, then the "tall" empire will create much more commerce and reach Republic much earlier.
    The importance of establishing an early Republic is underestimated by many. But in my experience, under Despotism your empire grows only linearly, while once you are a Republic, it takes off the ground exponentially...

    Also take into consideration: if you are too slow and miss the Republic slingshot, you will be facing an extra couple of centuries under Despotism... (Ok, this applies only to C3C, not to Vanilla/PtW.) So in my experience it is worth it to refrain a bit from expanding too wide in the early phase and in return for that get Republic much faster, at which time you will "explode" and easily catch up & overtake the AI, which has meanwhile deployed the "settler-at-size-3" strategy.

    Coincidentally, I just found a short discussion between two excellent players, templar_x and PaperBeetle (who both were #1 in the GOTM ladder at some point of their career) and myself, about when it is better to build the first settler early, and when it is better to let the capital first grow a bit and build a granary (and maybe 1-2 workers) first. See posts #110 and #117-120 in the above mentioned Asterix game. Interestingly, all three of us agreed that the "granary first" approach was better in this situation, even though there were two reasonably good town spots available not too far away (one with a cow and another with two floodplains). The reason simply being that either of those spots would have required lots of worker turns before it would become productive, and therefore it was better to first improve the capital to full potential, before settling the second town. (Unfortunately all my attachments in that thread have been eaten by the forum upgrade a few years back, so you can't look at the situation yourself. I really need to check sometime, whether I can still find the screenshots and sample .sav files on an old disc. That thread has lost most of its value because of the missing attachments.)
    So anyway, it's interesting that I am not the only one, who does not consider "cranking out settlers at all costs" to be the top priority. "Becoming powerful" is the top priority.
     
    Last edited: Jun 25, 2020
    Aabraxan and need my speed like this.
  6. Aabraxan

    Aabraxan Mid-level Micromanager

    Joined:
    Jun 29, 2006
    Messages:
    5,297
    Location:
    Arkansas
    Well said, @Lanzelot

    I have never gotten around to playing PTW, but I know that in C3C, getting out of Despo is huge. Emperor has always really been my fun/comfort zone, but even at that, getting into Republic quickly is usually a big part of my early game. Not only does it get the cash flowing, it gives me valuable techs with which to trade.
     
  7. justanick

    justanick Emperor

    Joined:
    Oct 6, 2010
    Messages:
    1,049
    Location:
    Germany
    You are exaggerating. Unless your empire is very small those 2 or 3 granaries will not suffice to fivefold growth. And you need to discount the growth that those towns with granary could have had if they had not transfered it to other cities.

    In practice a good part of the value of a granary is gained after reaching city size.

    I would mildly disagree. While same net food is gained by granary or settler, settler is preferable. This is even more true if this helps to connect luxuries. It becomes less true once corruption hits in. So there is some tradeoff there.

    During despotism a good deal of mines and irrigations can be ignored at first. Roads however are important for research and for moving settlers and for reducing distance corruption. The later is high in despotism.

    So my point would be that settler first is the (very slightly) better strategy (for an an early republic) unless the granaries do increase net food to a greater degree than settlers would. In practice that is rather big unless since most players will rather choose a starting position with some nice food bonus.


    If however expansion is expected to be limited and an early swift to war production will be needed, then granary first will be preferable even at same net food as this will help to max out the production of not settlers but of workers. It is worker production where granaries come in handy. For settlers the tradeoff is almost on par.
     
  8. SuedecivIII

    SuedecivIII Warlord

    Joined:
    Apr 20, 2019
    Messages:
    119
    Interesting. I can probably learn a bit from this. I often delay switching into republic, until I have my civilization more "online". That is to say, I think I can function without the military police, the extra units on defense, and the risk of war weariness if someone attacks, and survive the anarchy in the meantime. If it was possible to switch earlier there'd be benefits to that. I'll definitely start considering this playstyle in situations (like island starts) where I'm not competing for city spots.

    In your 3 factor list there's something enormous you're missing though, and that's food. Food is ultimately the biggest barrier to development, it lets you grow your cities, and it lets you build workers, without which the extra population isn't worth much. If you build a settler, your surplus food is usually unchanged. But as soon as you plant the new city, you get 2 surplus food for free, more if you plant next to bonus food. If you delay building that settler by 10 turns, you're basically sacrificing a worker (and 10 commerce minimum, and 4 unit support). Just for what? To have 1-2 extra shields and commerce per turn in your capital.

    Interestingly, one of the reasons I've always seen commerce as low priority early is that's there's no power spike tech early on. There's nothing like steam power or replaceable parts or military tradition at the end of the ancient era that rewards you for teching hard. But if republic is that tech for you, it makes sense. For republic to be a big boost though, you'd need a higher population in your cities, in which case... you'd get to republic faster. So I think both of strategies here are internally consistent. It's ok for me to delay getting to republic because I play a playstyle that doesn't benefit from the immediate switch, and vice versa.

    I think a lot of our differences come down to assumptions about maps and difficulty level. I think with bad land food maxing shines, because the benefits of high population are relatively lower and the benefits of the city center yields are higher (I didn't even mention slaving). And I was mostly thinking about things from a demigod+ perspective. On emperor or lower, of course, production is king, just spam 10 archers and you can take enemy capitals right off the bat. But it's dangerous to tell the players who play on emperor or below to focus on growing their cities. 90% of the time, the issue is not expanding enough (militarily or with settlers), and a lot of them aspire to play peacefully. If they ignore the conquering part, but pay attention when you tell them to grow their cities, that's how you end up with players like the OP of this thread, getting beaten in expansion by the AI on chieftain difficulty.
     
  9. justanick

    justanick Emperor

    Joined:
    Oct 6, 2010
    Messages:
    1,049
    Location:
    Germany
    Getting out of despotism should be your priority number one in the ancient age. Donnot delay republic. Not unless there is an insanely good reason for it like having to survive an ongoing war.

    Not quite. If you still have low population per town during anarchy you can still grow during anarchy. You need the high population per town after you enter republic, not necessarily before that. Once you are a republic you may easily use the luxury slider. But while still in despotism it can be sensible to watch out how much population you can sustain with no or near no luxury slider.
     
    WeirdoJoker likes this.
  10. Lanzelot

    Lanzelot Moderator Moderator

    Joined:
    Jun 25, 2007
    Messages:
    5,758
    Location:
    Heidelberg
    Agree with justanick. Republic benefits even the smallest of empires, simply because it gives you lots and lots of food that you don't have under Despotism. And because it increases your commerce by somewhere between 33% and 50%. The extra food clearly outweighs the bit of food I lose due to not building that many settlers in the early despotism phase, and the extra commerce balances the extra unit support you have to pay as well as the luxury slider you have to crank up because of missing military police. But after a very short time, you will have grown enough, that the unit support will go away, and you should have acquired enough luxury resources, that the lux slider can also be lowered to an acceptable level (sometimes even back to 0%).

    And then you have all that extra commerce available for research or for connect/disconnect. (Depending on your desired victory condition.)
     
    WeirdoJoker likes this.
  11. Dragon1965

    Dragon1965 Chieftain

    Joined:
    Jun 9, 2020
    Messages:
    29
    Thanks for the reply. I have to ask are you talking about something other than vanilla Civ 3? That's what I'm playing, the game from off the shelf in 2003 with no DLC.

    My questions:
    1. Temples - if I don't build temples how can my culture grow? I build temples not just for morale, but also so my culture will expand and fill in gaps in my border, to deter settlers from other civs (somewhat). I'm not looking for a culture victory per se (although I always leave it enabled), but mostly to fill in those gaps.
    2. Dialing up the luxuries instead of building temples - if I do that then my research level will necessarily go down (I usually have it as high as possible while still making a profit), yes?

    I'll check into the "settler factory" concept elsewhere on this site, thanks for the tip. It's interesting seeing the newfangled perspectives some people have, maybe it'll improve my game!
     
  12. Lanzelot

    Lanzelot Moderator Moderator

    Joined:
    Jun 25, 2007
    Messages:
    5,758
    Location:
    Heidelberg
    Both questions are already answered in the statement you quoted... ;)

    Let me add a few more details:
    1. For culture expansions: a Temple gives you 12 additional tiles and costs 60s. A settler gives you between 9 (worst case) and 25 (best case) new tiles and only costs 30s! (If you settle the new town intelligently, meaning at distance 3 from existing towns, then the "gaps" are filled automatically! This is how you can gain 25 tiles by settling inside a 5x5 hole in your territory. All adjacent tiles around the 3x3 square of the new town are automatically included into your territory.)
      So the benefits are obvious: the settler needs only half the shields, provides about the same number of new tiles and instead of costing you money (=research), it gives you money, food and production!
      I would say, Temples slow down your research (and your expansion), they are not allowing you to "research faster"!

      There may be situations, where a settler cannot be used, then a temple might be ok. But even in those cases: if you don't need the extra tiles urgently, wait until you can build Libraries and then use a Library to get those tiles into your borders. (The Library will at least help with research in contrast to the Temple, which only hurts you...)

    2. For happiness: a Temple costs 1gpt and gives you one content face. And gold that you constantly pay for Temples, can also not be invested into research! Keeping one citizen content via the luxury slider, also costs 1gpt, so the cost is exactly the same. But for the temples you pay forever, while the lux slider can be dialed back, when you hook up/trade another lux resource! Or if happiness is a problem only for 1-2 cities, you could temporarily use a clown/scientist there to keep it happy.
    There is no DLC in Civilization 3. There are three versions, all of which came out via CD: Vanilla (2001), PtW (2002) and C3C (2003). So if you are playing the version from 2003, it is not Vanilla.
    But anyway: all these basics we are talking about here, apply to all three versions. The difference with regards to game-play strategy are only minimal (and will probably not even be noticed by the beginner). If something I say applies only to a certain version, I clearly indicate that.
     
  13. Aabraxan

    Aabraxan Mid-level Micromanager

    Joined:
    Jun 29, 2006
    Messages:
    5,297
    Location:
    Arkansas
    1. If you're not looking for a culture victory, why do you need culture? I understand the 'filling in the gaps' rationale, but as was explained, you can do that with settlers instead. Temples costs maintenance. Settlers become towns, which generate income.
    2.Yes, luxuries and research come out of the same budget. So does maintenance on temples. The sliders give you a lot more flexibility, though.
    Might I suggest the War Academy: Spotting Settler Factories

    ETA: Remember that your towns need nothing to survive. The only question is what the empire needs for them to have.
     
    Last edited: Jun 27, 2020
    Dragon1965 and tjs282 like this.
  14. SuedecivIII

    SuedecivIII Warlord

    Joined:
    Apr 20, 2019
    Messages:
    119
    I have to hard disagree here. Sometimes you need to wait before switching. Let's not talk theory though, let's dig into an example :)

    The file attached is a demigod game as Sumeria. This is a game I won without too much trouble, without any conquest. It's turn 93, we could easily have republic by now if we wanted it, but I prioritized currency instead, because it's a more reliable trading tech.

    We are under the unit support cap in despotism. If we switched to republic we'd be 18 units over. We only have 8 workers here, we'll probably need 15 more. We can probably disband a few of our scouting enkidus, but we need everything else. The mayans and romans are strong to us, and running troops through our land. If anything we're cutting it kind of close here. We can't afford to disband any enks at home, and if we do it will be to replace them with swords or horses. TLDR, we could easily be paying 45 gpt in unit support.

    We'd also be giving up a ton of military police. We'd have to replace them with the happiness slider.

    The one huge benefit to switching here is the additional food in city center tiles, but that's unique to agricultural civs (and the lack of fresh water here). That's not normally a benefit you'd have.

    If someone attacked us, it'd be manageable in despotism. We could quickly slave out some swordsmen. In republic, an attack would be much more likely to be fatal. Sure, we could buy units with cash, but a military alliance would probably be the better purchase in either government.

    I think you could make a case here that it's worth switching into republic here. But I'm happy to stay in despotism until I have more worker moves, all my cities planted, my border secure, and more population in all my cities.
     

    Attached Files:

  15. Aabraxan

    Aabraxan Mid-level Micromanager

    Joined:
    Jun 29, 2006
    Messages:
    5,297
    Location:
    Arkansas
    Clearly, there are times when staying in Despo is a better option than switching to Rep, but I'd suggest that those are the exception rather than the rule.
     
  16. Lanzelot

    Lanzelot Moderator Moderator

    Joined:
    Jun 25, 2007
    Messages:
    5,758
    Location:
    Heidelberg
    This is an extreme example: the start position was quite mediocre, no food bonus anywhere and only a very small stretch of uncorrupted land to work with. To win this start on Demigod is definitely an achievement!
    But still I have to say: if your capital is still size 1 in 690 BC :eek:, then something must have gone badly wrong.

    Can't really say, how "my" strategy would work out in this situation, but I would probably not have founded all those size 1 cities. I would probably try to implement a plan like this:
    1. Build 4-5 towns around Ur and make them productive asap.
    2. Build up an early military and try to take Chichen Itza. (Or even better a town by the Wheat that the Mayas have.)
    3. Disband Ur and jump the Palace to more fertile and productive lands.
    4. Then take it from there...
    Do you still have the 4000 BC save? I would really like to try how I would fare in this situation. Also, what is the VC you are shooting for?
     
  17. Lanzelot

    Lanzelot Moderator Moderator

    Joined:
    Jun 25, 2007
    Messages:
    5,758
    Location:
    Heidelberg
    Never mind, with the seed and the world properties from CivAssistII, it is possible to recreate the map:

    properties.png

    (I only need the desired VC. And did you have SGLs turned on or off?)
     
    Last edited: Jun 27, 2020
  18. SuedecivIII

    SuedecivIII Warlord

    Joined:
    Apr 20, 2019
    Messages:
    119
    SGLs on, Domination/Cultural/Diplomatic/Conquest/Space Race on. Wonder victory off.

    I'm interested in how your strategy would do in terms of expansion here, and what kind of benefits it would yield going into the mid-game.
     
  19. Lanzelot

    Lanzelot Moderator Moderator

    Joined:
    Jun 25, 2007
    Messages:
    5,758
    Location:
    Heidelberg
  20. SuedecivIII

    SuedecivIII Warlord

    Joined:
    Apr 20, 2019
    Messages:
    119
    Here's the game end save. The self imposed challenge was winning with mediocre land without conquering my neighbors or building the great library.

    I think I might have benefited from a second granary early in my capital, but the opening was such a scramble for city spots. So there's definitely some optimization possible there, having a granary would make it easier to transition into wonder building or stacking units.
     

    Attached Files:

Share This Page