SGOTM 10 - Smurkz

[Backwards Logic]

My green site isn't wonderful? Uh, settling on a plains hill working three seafood and having at least three other hills to work isn't 'wonderful'? The city can set up as a GP farm mid game once all land is settled then transformed into a production powerhouse, something all other city options currently lack. Personally I think yellow is overrated - any city with 2 deserts in the BFC and a desert hill is hardly wonderful. It's like you're making an accuse just to settle the clams. It has less production early on for settlers which settling 2 E has.

Also settling the other red city 1 S of the iron is no longer an option in my opinion. That ship sailed after turn 0 passed. I believe I mentioned any city that was going to be settled up there needed to be moved that turn to get into position. Delaying a turn to settle a marginal city compared to other options doesn't sound very optimal to me. Plus doing so basically cans any decent spot up where green normally is, barring any additional resources show up.

Back to my other option, the point of this city early on is not for massive hammer potential - I'll concede it doesn't have the food to do so. All it really needs is enough hammer to spit out fogbusters at a reasonable rate and with 12 base at pop 6 (2 ivory @ 3 each and 2 mined grassland hills) it can certainly live up to this task. The capital gets settler spamming duties while this city spits out units. Once the green site is settled and up and running, this city stops working all of the ivory and hills and goes straight to working cottages. I should also point out we're one tech away from replaceable parts which gives a hammer boost to windmills, so it's not like it won't have any production.

 

[Xcalibrator]

My green site isn't wonderful? Uh, settling on a plains hill working three seafood and having at least three other hills to work isn't 'wonderful'?

Well, hmm, no? It may be my residual inclination toward ICS but it seems like the cities don't fit the available terrain as well that way. The green city may end up with too many sea tiles and there probably wouldn't be a good location south of the gold. With option 4's yellow city, yeah, the 2 deserts are completely useless and the 4 non-clam sea tiles not so great, but that still leaves room for 14 citizens (and a mine on the desert hill is useful) and any more pop can be used as specialists.

Your point about settling 2E being stronger initially is quite true and probably very important, though. You and zyxy have put a lot more effort and thought into this than I have so I leave it up to you to decide (with welcome input from anyone else who would like to contribute, of course).

 

[Niklas]

ICS rocks!

Not that I have paid enough attention to what you're discussing to have any truly relevant input, I just wanted to say that.

 

[Backwards Logic]

Just messing around, I've exceeded my first posted test run. In fact, I've smoked it by quite a lot.

On turn 40:

• 4 cities (all settled by turn 32) - Moscow (pop 7, working irrigated rice, corn, pasteurized cattle, both mined gold, iron, and running two scientists (yea, I found a slot to squeeze a library in). Great Person would pop next turn), St. Pete (pop 5, working 2 farmed floodplains, 1 camped ivory and two cottages while running an engineer), Rostov (settled my green site, pop 3, working crabs and clams both with fishing boats as well as grassland mine and engineer) and Novgorod (settled 2 W of clams, pop 2. City being set up to run two scientists and could be ready in four turns).
  
• 5 workers
• 3 workboats, with one more in 5 turns. (2 hooked up on clams, one on crabs)
• 4 fogbusters, not including the explorer which remained exploring. Units are predominately archers since they are cheap and will be able to withstand a random barb attack if necessary, though I did sprinkle in a longbow or two in there. Again depending on how much land we have to fogbust this number can be increased.
• Garrison in each city, again mostly archers
• Another settler due to be complete in five turns

Now depending on how much land is around, I can probably get 6 settlers out in the same time frame, though our GP production will suffer as a result. I only settled land that we know about, hence the city 2 W of the clams. However if we run a handful of farms down there it could end up being pretty decent in terms of a GP farm.

Regarding barb cities, it seems like turn 20 is the magic number when they start appearing. When they do, they usually have 3 longbows garrisoned in each city. I also noticed their borders pop shortly after their establishment.

As for our unit production, we lose the ability to make both warriors and regular archers once the iron is hooked up. Something to keep in mind.

I'm willing to wait until Monday/Tuesday for a few reasons. One, it'll give everyone some time to test out their scenarios and report back what they accomplished. Two, Rolo theoretically will have BUFFY up and running, so he'll be able to have some concrete thoughts on the subject. And three, if anyone is feeling adventurous maybe some spreadsheeting is in order. Finally, can one of you explain what this ICS thing is?

 

[Xcalibrator]

ICS = Infinite City Sprawl, the city placement method of true warriors. Such as Niklas, apparently.

4 cities by turn 32 is nice. Which city scheme did you use? Post 71 option 6 or post 73?

Did you use Worldbuilder to check for barb cities or was that just when you saw them while playing? Any feel for where and how often they appear?

I'll try to do a better job of collecting info and screenshot links on the first page. Soonish.

 

[Backwards Logic]

Post 71 Option 6. On turn 20 I saw the edges on barb cities. If it wasn't fogbusted by anyone there's a pretty good chance it'll pop up. For giggles I played through a second time just to see of the barbs would pop up in the same place, and so long as no one can see them they seem to settle in the same places if you let them. Moral of the story is if it looks like it could be a useful spot for a city later go stick a unit by it.

EDIT: Ran another test. Four cities by turn 31, 5 by turn 38. I'm improving

 

[Xcalibrator]

Maybe it's possible to do better, but I'd say BL's scheme is looking very good. I tried settling W of the coastal elephants and S of the western gold and here's what I had after 47 turns:
4 cities (pops 9, 8, 3, 2) + 1 settler
4 workers
4 garrison units (warrior and archers) + 4 fogbusters (longbows and warriors) + explorer
4 seafood resources, 2 gold, 1 ivory, and iron hooked up
1 freshly minted Great Engineer (100% probability--hadn't run any sci's yet).
Library and barracks in Moscow

I built more roads than optimal if just going for settlers and probably let the cities get bigger than necessary. I had 4 farms, 3 mines in addition to the gold... Still, I really doubt I could better BL's settler and other production using the 2 starting sites I did. Anybody else do some trials? zyxy?

What were your tactics, BL? Just focus on chopping plus the bare minimum of other improvements, or something more complicated?

 

[Backwards Logic]

A lot of it was the timing of building workers/settlers vs fogbusting units. The chops were saved exclusively for the workers and settlers since the only time the cities can grow is when units or buildings are being built. I had unit builds timed so that when they finished the city had just grown and/or another improvement was ready. Since cities have forges from the start, improving tiles that give extra hammers added a bigger bonus to cities producing settlers than say the farms did. I don't have the numbers in front of me, but I think I had all of the 5 workers I had out by turn 10 or so and the majority of fogbusters out before turn 20. The trick is getting the fogbusters in place before barb cities pop up.

I played the game out rather lackadaisically just to see the AI's trading tendencies when it came to certain techs. I researched Printing Press first and the only tech I could get in return for it at the time was Gunpowder. Replaceable Parts on the other hand was far more lucrative. It seems like the AI's hold Astronomy in pretty high regard and it may be possible use it to get more techs in return. Later on by the industrial age I held a commanding tech lead over everyone using some timely trades, so our tech deficiencies early on can definitely be overcome.

I'll run back through and do my best to replicate my numbers while also putting together a list of everything I did for your perusal when I get time. In the meantime I'd love to see some more test runs from others for the various proposals, keeping in mind that the number of settlers produced say by turn 40 be the measuring stick.

 

[Xcalibrator]

What, nobody else has any input?! I ran another test settling N of the corn and E of the coastal ivory--just in case things look ugly west of the coastal forest where we'll likely settle. I got settlers on turns 31 and 32 and a 6th worker on 34. By 35 I had 1.5 libraries, 1 barracks, 6 warriors, and 1 archer. It would take another 8 turns for a 3rd settler in Moscow.

[Edit: As of turn 35: 3 workboats, all resources improved, population 7 in both cities.]

 

[zyxy]

That's very impressive BL!

I did some calculations to check the alternatives. Our eastern settler can go 1S to the Yellow site, or 2N to the Red 2E site. I assumed this settler will found the capital. The western settler has a lot of options. I checked the Blue coastal site 3W1S of the starting spot, the Blue coastal ivory site 3W1S 2W1S, and the Blue inland site (BL's site at 1W3S).

Some general findings:

1. For a very fast REX, forget about improving tiles (almost). Chopping workers and settlers is the key. It is quite possible to get 2 settlers and 2 workers done by turn 21.
2. Growth is rather pointless until tiles are improved. Together with the previous observation, this leads to the following strategy: quickly chop a few workers. Then improve some tiles and chop some more while training a settler. If correctly timed, the town has a set of tiles ready for quick growth once the settlers is done. For example, once Red 2E starts growing, it takes 6 turns to get to size 3, and after that only 7(!) turns to get to size 7. Yellow is similar, the blue sites are a bit slower.
3. Having a lot of forests in the BFC is a key ingredient for this strategy. This is where Red 2E, Blue inland and to some extent Blue coastal ivory really beat the other options.
4. Some performance indicators by (roughly) turn 40:
   • Red 2E: Size 7. Produced 2 workers, 2 settlers, 1 warrior MP, library. All resource tiles in BFC improved, plus a few more. Produced about 820 gold (at 0% sci).
   • Yellow: Size 9. Produced 2 workers, 1 workboat, 1.3 settlers, 1 warrior MP, nearly a library. Almost all visible tiles in BFC improved. Produced about 1000 gold (at 0% sci).
   • Coastal blue: Size 7. Produced 2 workers, 2 workboats, 2 archers, 1 warrior MP; 147 hammers into a settler. All visible tiles in BFC improved. Lots of surplus worker turns. Produced about 250 gold (at 0% sci).
   • Blue coastal ivory: Size 5. Produced 2 workers, 1.5 workboats, 1 settler, 1 warrior MP. Six tiles in BFC improved. Lots of surplus worker turns. Produced about 175 gold (at 0% sci).
   • Inland blue: Size 6, growth in 2. Produced 2 workers, 1 settler, nearly 2 archers, 1 warrior MP. Nine tiles in BFC improved. Produced about 180 gold (at 0% sci).
5. Of course, comparing the two options for the western settler is not really possible because we don't know half the tiles for the coastal option. I am still worried about the potential of the inland one because of the limited food, and the calculations confirm this. At size 7, it has more or less run out of good tiles. (Note that two of the grass tiles are shared with the Red NE capital, so blue will not get them.)
6. The Coastal blue site also runs out of tiles very quickly. We can hope there will be something in the fog, but of course this is a gamble.
7. The capital gets 75% hammer bonus. So Iron is a really strong tile (much stronger than cow when training settlers or workers).
8. These sheets are not quite optimal yet. For example, for Red 2E the worker first chops a forest outside the BFC, losing 16 base hammers (28 after multipliers). In fact, the sheet for Red 2E is the only one that I serously tried to optimize.

The spreadsheets for the capital:

Spoiler :

Red 2E option:


Yellow option:

And for blue:

Spoiler :

Blue coastal option:


Blue coastal ivory option:


Blue inland option:

The sheet (Open Office format) is attached.

 

[Backwards Logic]

I checked the Blue coastal site 3W1S of the starting spot, the Blue coastal ivory site 3W1S...

Help me here, but isn't this the same site? What's the difference?

[*] Of course, comparing the two options for the western settler is not really possible because we don't know half the tiles for the coastal option. I am still worried about the potential of the inland one because of the limited food, and the calculations confirm this. At size 7, it has more or less run out of good tiles. (Note that two of the grass tiles are shared with the Red NE capital, so blue will not get them.)

Why?

One of the two overlap tiles has to be a farm for the irrigation chain, so why not give this tile to my inland blue city? Fixes the lack of food plus the capital doesn't really need it with the rice, corn and cows. The short term plan for this site is producing fogbusters and chopped workers/settlers which it handles fine. The long term plan for this city is to be cottaged up, and with all of the grassland available it definitely can fill this role better than the other sites. 2 Floodplain farms, 1 grassland farm, 4 riverside cottages and still growing at pop 7 after using the two ivory to get some hammers out early on... What's the problem exactly?

I'm also certain of the tiles immediately to the west that are obscured by the fog for the coastal 1 W of the ivory city. There's definitely no forests there, and barring any resources no additional food on the plains and desert hill tiles. The only things that can go there would be wheat or cattle, and I don't think Gyathaar would give us all three granary health bonus resources right from the start. The site simply isn't that good for anything specific, and the lack of forests for chopping in my opinion knock it out of the running for settling consideration.

 

[Xcalibrator]

Help me here, but isn't this the same site?

I think this is just a typo, BL--the coastal blue ivory is the tile next to your blue.

Very nice comparions, zyxy. Forgive me for doubting you.

You guys have done a great job analyzing the various options and basically it looks like all of the 5 sites under consideration, although having different strengths and weaknesses, would do the job for REX using mad chopping, at least for the first 3 or 4 or so settlers, and by then we will probably (?) have time to fill in the land we've blocked off or be able to reach unclaimed land by sea. True?

I'm happy to let you two hash out the decision, but here are some other things that it might be good to consider. What are your feelings for the long-term potential of these sites, keeping in mind the 50% Bureau bonus for commerce and hammers? How important is a river, and can we expect to find few or many other river sites on this kind of map? Would having a bit more overlap between inland Blue and the eastern city be a significant problem, and if so by when? Mid-game or too late to matter?

How do the various choices compare in terms of their effect on future city placement, particularly in cases where a future city would have a substantial number of tiles in unexplored areas? This is more of a concern in island maps such as ours, although we may turn out to be on a fairly large and non-snaky landmass where we don't have to worry about fitting cities in just the right location.

What about health? Will having delayed access to seafood with some city placements be a problem (particularly after we've chopped so many forests early on)? Or will we be able to build coastal cities and net the fishies before that becomes a problem?

I eagerly await your pronouncements and the next turn.

 

[zyxy]

Help me here, but isn't this the same site? What's the difference?

Thanks, it is was a typo.

One of the two overlap tiles has to be a farm for the irrigation chain, so why not give this tile to my inland blue city?

Ah, good comment! This got me thinking. The rice and corn can (eventually) be watered from the tile SW of the corn. The resulting +7 fpt for the inland blue site can be used to work the ivories, 4 mined grass hills and a cottaged plains tile. Meaning it can get to size 14. It will keep growing pretty fast if we temporarily windmill some of those hills. At size 14, it will have 2 ivory camps, 2 fp farms, 1 grass farm, 4 grass cottages, 1 plains cottage, 4 mined grass hills. Nine of these are riverside locations. This brings a total of 0 fpt, 20 hpt and 17+ copt (=commerce pt). This is rather good, even long term.

Do you have any idea what the 2 fogged tiles are?

I would definitely give the tiles in the overlap to the capital, because it has better modifiers.

So all in all, I think B:'s site is fine. If this were my private game, I would probably take the gamble and go west, reasoning that it could be a lot better than what we see (there can be wheat, or seafood, for example), and that I lose at most one turn on settling BL's site if the fogged tiles turn out to be the Gobi desert . But going straight for BL's site is fine too.

Once we agree on a settling site, I would like to try and optimize the initial micromanagement.

 

[zyxy]

What are your feelings for the long-term potential of these sites, keeping in mind the 50% Bureau bonus for commerce and hammers?

Good questions!
Red 2E is the best site IMO, but it is fairly close with yellow, because yellow has a slight advantage in gold output initially.

At size 14, for example, using all land tiles, it can have 6 resources, 4 grassland cottages, 3 plains cottages, 1 mined plains hill, for +3 fpt, 19 hpt, 25+ copt. Eventually all tiles in the BFC are usable. Each tile beyond size 14 adds 2 copt.

At size 14 Yellow gets 5 grass cottages, 1 plains cottage, 2 golds, corn, cow, clam, grass hill, 1 desert hill, 1 coast for +4 fpt, 15 hpt, 29+ copt. Two tiles in the BFC are unusuable. This site has a bit more gold, but less hammers, and lower growth potential (because of the unusuable tiles) than Red 2E. Also, it takes longer to develop.

The blue sites have substantially lower gold output, so don't qualify for the capital IMO.

How important is a river, and can we expect to find few or many other river sites on this kind of map? Would having a bit more overlap between inland Blue and the eastern city be a significant problem, and if so by when? Mid-game or too late to matter?

I would rather minimize overlap so that our cities can max out in size in the midgame. No idea how abundant rivers will be. They bring money and some health. For health it is better to be coastal than riverside, given the choice (seafood).

How do the various choices compare in terms of their effect on future city placement, particularly in cases where a future city would have a substantial number of tiles in unexplored areas? This is more of a concern in island maps such as ours, although we may turn out to be on a fairly large and non-snaky landmass where we don't have to worry about fitting cities in just the right location.

The two decent sites for future cities that we can see are Green (N of crabs) and somewhere down south instead of Yellow. Any coastal blue will reduce the potential of Green, the inland blue will reduce the potential of Yellow-ish. Given the choice, I think Green is the better site.
It would be nice to know the lay of the land, so that we know where to block.

What about health? Will having delayed access to seafood with some city placements be a problem (particularly after we've chopped so many forests early on)? Or will we be able to build coastal cities and net the fishies before that becomes a problem?

Good point! It will be an issue initially, so I really need to check my spreadsheets for this once we have agreed where to settle. Perhaps the corn and rice need to be roaded sooner.
Eventually, with 6 health resources visible already it should not be a serious issue I think.

 

[Xcalibrator]

On the progress page you'll see that Murky's score dipped around turn 35 even though their power kept rising. What would cause that? And does it perhaps indicate some barb trouble?

Meanwhile, the decision on where to move the settlers this turn is....?

 

[Backwards Logic]

Well, I would have had time to post sooner, but having my wisdom teeth removed today knocked out most of my time and I wasn't exactly in any condition to think.

On health: During my tests it was around when the city (settled 2E) hit pop 5 that I had issues. The mass deforestation and delay in roading the health resources is entirely responsible for this. I've found using the rice as the tile roaded helps in a variety of ways. One, it gets the +1 granary bonus bringing the bonus to +2. It also brings us one step closer to hooking the inland blue city up to the capital for resource sharing (later needed for ivory) and trade routes.

On capital: Whichever city has the gold gets capital. Plus 2E has much more base hammers than the yellow site does, so this the 50% hammer bonus will be larger here as well.

On which settling combo is better long term: Once again sounding like a broken record, but I still like mine the best. Settling blue coastal weakens green too much for my liking, and I've found the green site in my settling scheme is extremely flexible. There's enough food for at least four specialists, and that was when the city was working three grassland hill mine tiles and 2 workshop grassland tiles. GP farm or hammer center, take your pick. I think the remaining surrounding land will dictate what this city ultimately becomes. I'm not a fan of settling yellow at all unless it's a secondary city. There's just not enough hammers to spit out enough settlers early via chopping or building them long-term. As zyxy pointed out the Iron is a massively important tile for our strategy.

On the 2 fogged tiles that remain in my blue inland site: South of the grassland hill is a plains tile and west of that is a grassland tile. Neither have forests, but the tile south of the plains tile does.

I'd also like some confirmation from those involved that they agree with the settler moves 2 N for capital and 1 SW, 1 SE for inland blue site before I move anyone.

I'm feeling up to it now, so I'm hoping within the next few hours I'll have re-enacted my test save where I had 5 settlers by turn 38. I'm also going to mess round with different chop orders (ie, worker -> settler vs settler worker in a few instances).

EDIT: On second thought, it won't be done tonight. I'll try to have it up tomorrow sometime.

 

[Xcalibrator]

I'm OK with your plan, BL, so play another turn when you're ready. I'd just add that doing simultaneous chops on more than 1 tile allows more time for growth if you switch in and out of settlers--but you knew that. I hope you feel better soon after the wisdom teeth. That was quite awhile ago for me but I remember it wasn't fun.

 

[zyxy]

settlers 2N and SW-SE is OK for me.

But that doesn't help to end the turn, because we also need to decide on the worker. Chopping where it stands loses 28 hammers, I would rather avoid that if possible.

 

[Xcalibrator]

But that doesn't help to end the turn, because we also need to decide on the worker. Chopping where it stands loses 28 hammers, I would rather avoid that if possible.

Why would the hammers be lost? The chop won't complete until the cities are settled, although we may need to be careful about the settling order so the hammers go where we want, and perhaps canceling the chop order so it doesn't complete before settling. It doesn't matter that the forest is outside a BFC.

 

[Backwards Logic]

No, worker definitely moves 1 NE and chops in the capital. There's real no other chop location alternative that maximizes hammers. Plus the worker remains setup to run over to St. Pete and chop out the worker being built there without any extra turns lost. The worker actually gets done the same turn either chopping in place or moving NE then chopping, but there's a much higher overflow with moving that gets the second built worker out quicker. I think your SS has it in there and I know mine does as well. @XC the hammers are lost because the tile isn't in any borders when the chop is finished. Moscow's borders pop in 8 turns (assuming no second religion founded here), the chop is done in three. It's +29 hammers outside the border and +44 inside.

Once I get an okay from our three man team, I'll go ahead with these moves, settle the cities, start the chop and move our longbows and explorers around. First builds in each are workers. The explorer moves SW to start then waits to see if anything else opens up before moving a second time. The northern longbow moves 1 W and the southern longbow likely moves 1 SW.

And then we have to pick a starting tech. I'm leaning towards Astronomy for the tech trading benefits - it would seem the AI value this one the most out of what we have to start.