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

Granary or settler?

Discussion in 'Civ3 - Strategy & Tips' started by ThinkTank, Sep 26, 2007.

  1. ThinkTank

    ThinkTank RL Addict

    Joined:
    Feb 2, 2007
    Messages:
    1,085
    Location:
    An inaccessible possible world
    I agree, and

    Indeed, for the numbers that come out of the calculation it is only relevant that it is a 3 food 2 shield tile.
     
  2. haphazard

    haphazard Chieftain

    Joined:
    Jan 30, 2004
    Messages:
    70
    This is a great discussion, but lets see how tough everyone really is.... How about a friendly little competition?

    The following scenario is designed to compare REX skills and the other topics discussed in this thread on a tiny pangea map. The computer players have been removed to eliminate that random element of play, but I have added in a few extra barb camps to force players to produce at least some military units.

    When choosing the map, I wanted a normal random generated map with sufficient room for people to compare and contrast their growth strategies on. Since many people, either manually or by using one of the tools available, specifically hunt for a starting position that already has fresh water, and other
    starting conditions nearby, I did the same type of search. The first random map that met those conditions is the basis of this scenario. Note, the starting position has been tweaked to give some options that allow a strategy to be demonstrated. (No I am not starting anyone in the middle of a desert.)

    Since some players make their strategy based upon if more shields or if more food is available I have given you a choice. The starting position is on a mountain. All players will have to move to found their first city. If you move north, you will have two cows. If you move east, you will have 2 flood plain
    wheat. For those of you that can not decide, if you move north-east, you will get one of each.

    The objective of the scenario is to see how many turns it takes to get 10, 20, and 30 cities. Post your dates/turns for each of these 3 levels, along with a short summary of how you managed your cities as far as granaries/settlers/workers/etc. Since this is not a real contest, the only award you will get is the unending respect of your peers. :lol:

    Here are the world parameters for the map that was produced:
    Tiny, 60% Ocean, Pangea, Random seed=1020939890. The rest were defaults.

    I then modified the map as follows:
    The western most starting position was assigned to player 1.
    the northwest wheat was deleted, and 2 cows were added.
    The players starting position was changed to a mountain to force a first turn move.
    The other three starting positions were deleted and replaced with barb camps.
    A few additional barb camps have been placed as well.
    I then created a small 3x3 island where I placed a new (second)
    player starting position.
    Scenario properties were changed setting number of players to 2.
    I modified the rules to change world size settings for "Tiny" to
    match some of those for a standard world, namely optimum city number
    was changed to 20, and tech level was set to 240.

    As a recommendation, I would suggest that the Maya would be a great choice for everyone to play. They are agricultural, industrious, start with pottery(Granary), have a UU that may even supplement their worker pools with a few free slaves. If you think you need a different civ to demonstrate your strategy, feel free since this is just a friendly competition. Oh yes, I am playing at Emperor level.

    And finally, good luck and have fun. :goodjob:


    http://forums.civfanatics.com/uploads/41993/REX-Challenge.biq
     
  3. Chamnix

    Chamnix Chasing Time

    Joined:
    Aug 10, 2004
    Messages:
    8,941
    Location:
    NJ, USA
    I would suggest a different objective to make it more realistic - if the objective is just to get cities, then there is an incentive to settle very close by (instead of leaving appropriate room to grow), and there is a strong disincentive to build workers.

    I'm sure you don't want to make this as complicated as the GOTM Quick Start formula, but I think the best simple indicator of how good a start you have is total # of citizens + total # of workers (+ # of settlers if you have any running around still).
     
  4. haphazard

    haphazard Chieftain

    Joined:
    Jan 30, 2004
    Messages:
    70
    The reason I chose number of cities is that is what most of the spreadsheets and analysis are driving at. (At which point does a certain strategy pay off more than the other.) Choosing 10, 20, and 30 cities was arbitrary and hopefully will show a reasonable slope. Other than that this is just to have fun and see how other people sprawl out rather than play accountant and debate the models used in a spreadsheet.
     
  5. ThinkTank

    ThinkTank RL Addict

    Joined:
    Feb 2, 2007
    Messages:
    1,085
    Location:
    An inaccessible possible world
    I think this is a nice idea. I have two proposals. First, let's remove any RNG factors that can influence the result. Bad luck does not disprove anything. So no barbs. Instead, let's tighten the rules and require this: no tighter build than CxxC, and the milestone numbers are:

    10 cities + 10 workers + 10 military units
    20 cities + 20 workers + 20 military units
    30 cities + 30 workers + 30 military units

    Second, I like to modify Chamnix' proposal, since 8 size 1 cities are not equal to 1 size 8 city. I was thinking number of worked map tiles, as rough measure of overall production; this is equal to # of cities + # of citizens. Maybe also account for improvements on tiles. And yes, number of workers and settlers.
     
  6. haphazard

    haphazard Chieftain

    Joined:
    Jan 30, 2004
    Messages:
    70
  7. haphazard

    haphazard Chieftain

    Joined:
    Jan 30, 2004
    Messages:
    70
    Well, I made a run through the sample map I provided and this is how it went(using Maya and Emperor):
    Went North with the settler to get the cow, west with the worker to chop the woods. My intent is to build a granary in the capital and then settlers.

    2350BC I had 10 cities, 11 workers, 10 warriors, 17 squares worked, turn=35.
    I am conducting no research, building no barracks, and have the lux slider around 50% to keep everyone happy. Found one barb camp to the south so I guess I missed deleting that off of the map. :blush: I will choose to let the barb pillage my city and only loose a few gold rather than loose the warrior.

    1750BC I have 20 cities, 22 workers, 22 warriors, 2 settlers, 41 squares worked, turn=50.
    Ended up building 2 extra granaries to speed up settler production. These were also built with forest chops. Performing some infill, but finding it takes longer to get to city sites. Workers are now spending time on harder terrain such as clearing marsh and roading mountains. Lost a warrior to a barb.

    1550BC I Have 30 cities, 32 workers, 31 warriors, 66 squares worked, turn=58.
    Lost a second warrior along the way. Barb activity is roaming so I guess that is why they keep showing up. It has taken as long as 6 turns for a settler to reach his city site, but overall I would guess the average was 3-4 turns. I left 2 easy to reach sites for cities 29 and 30. Otherwise I would have taken a couple of extra turns.

    Here is the final sav file. Let me know if you want the other 2.

    http://forums.civfanatics.com/uploads/41993/Smoke-Jaguar_of_the_Maya_1550_BC.SAV

    Now I do not consider myself to be a super player, but the number of cities produced exceeded the expectations from the spreadsheets. The differences were probably that I had 3 cities sharing the 2 cows and they produced the bulk of the settlers. Also travel time to city sites was shorter than expected, being about 3-4 turns instead of 5 as was used by ThinkTank. Having the granary up quickly with a forest chop was probably key.

    I would be interested in how other people approached this problem. It seems like the spreadsheets are a bit on the concervative side.
     
  8. ThinkTank

    ThinkTank RL Addict

    Joined:
    Feb 2, 2007
    Messages:
    1,085
    Location:
    An inaccessible possible world
    To be honest I almost could not believe your results. :eek:

    I had a look at the save, which explains it: you used accelerated production. This implies growth and build time all twice as fast. In my spreadsheets I assumed normal production. Also you used a fair amount of CxC builds, where I proposed not tighter than CxxC.
     
  9. haphazard

    haphazard Chieftain

    Joined:
    Jan 30, 2004
    Messages:
    70
    :blush:
    :stupid:
    Wheres the Do'h icon?
    OK, I knew I was going to embarrass myself by going first. I need to find some time to replay this because GUESS WHAT!, its going slower this time.....

    I knocked out the first 10 cities in 61 turns (actually 1 extra settler is also going for a distant site) so this is still going rather well considering I have to build workers and warriors as well. Here is the 10 city report:

    1475BC, 10 cities, 10 workers, 10 warriors, 1 settler, 23 squares worked, turn=61.
    Same rough strategy as before but this time I immediately built granaries (assisted by wood chops) in the first 3 cities that are sharing the 2 cows. This may have given me a slower start, but I expect to make it up as time progresses.
    I also started a little research this time since excess gold will not do much for me. It is interesting to note that you can almost fly through the lower end techs with this size of an empire. You will not be the tech leader, but by sacrificing everything to that first REX phase, you are in a good position to catch up.
     
  10. ThinkTank

    ThinkTank RL Addict

    Joined:
    Feb 2, 2007
    Messages:
    1,085
    Location:
    An inaccessible possible world
    :)

    The new numbers do seem to more in line with what I'd expect. I'll have a go at this myself over the weekend. Maybe we should just continue until the whole continent is settled and make note of all subsequent milestones of the form (10xn cities + 10xn workers + 10xn military units).
     
  11. haphazard

    haphazard Chieftain

    Joined:
    Jan 30, 2004
    Messages:
    70
    Here are the results from the latest try. For this run I remembered to turn off accellerated production.

    1475BC 10 cities, 10 workers, 10 warriors, 1 settler, 23 squares, turn=61
    1000BC 20 cities, 22 workers, 21 warriors, 1 settler, 38 squares, turn=80
    670BC 30 cities, 30 workers, 30 warriors, 0 settlers, 67 squares, turn=94

    What I was trying to get a feel for in this experiment is how the charts ThinkTank produced relate to a game where you have to dedicate some of your production to military and workers. Rather than all cities being pure settler pumps. The core cities I had, had about 5 surplus food, but the empire performed closer to the 4 surplus food column. (I know ThinkTank calculated on all follow on cities only having an even 2sfpt.) Also since none of my cities really grew that much, I had a mixture of CxxC and CxC density, but I do not think this affected the outcome.

    If I had to throw in more infrastructure, and a better quality military than all 3hp warriors, that would probably knock me down another level in the chart, eg 3sfpt.

    If someone wants to give this a go without granaries, go for it. I agree that a few key granaries make a difference in a long term game. Also note that I like to get them up quickly with a forest chop or 2. Practically every other city was busy with other types of builds so it really was limited to just these 3 cities creating settlers. This made it much more worthwhile.

    Here is the 30 city post:
    http://forums.civfanatics.com/uploads/41993/Smoke-Jaguar_of_the_Maya_670_BC.SAV
     

Share This Page