[SPOILERS] swance bitten, twice shy

[aetryn]

When you say flexibility and safety, under what circumstances do you see yourself Impi rushing, and at what % would you estimate the chance that someone rushes you?

hmm, i think i will consider a dedicated impi rush if we find a neighbor who is definitely not close to bronze working (maybe subjective but i'm thinking mysticism open + went for a religion as one of their first two researches) and whose lands are easy to reach (so, close to the min 14 tiles away and not on the other size of some hideous hill range that negates the impi movement advantage, bonus if they settle toward us). so probably not highly likely that all those things combine, maybe 5-10%? but what i think is significantly more likely is needing access to fast impis to secure an exposed, possibly semi-forward settle like at the newly-discovered floodplains region, given the pretty lackluster quality of the spots near the capital. but the exact odds of that are harder to precisely estimate without more scouting info

odds of me getting rushed, well, we have 3 other leaders who are probably at least considering it with superdeath + AGG rome, gavagai egypt, and GT boudica. odds of us neighboring at least one of them, given 4 neighbors per player, are 1 - (6/9 * 5/8 * 4/7 * 3/6) = 88% if that math is correct (which it may not be i'm rusty af). in the case that we do neighbor one, would we be their presumptive target, and tempting enough a target if so for them to commit? we are AGG zulu yes, but it sounds like these kinds of rushes can usually land around t40 (praets probably excluded) with AH first, we get bronze on t32 which is fine, but the issue is that unless we open scout - worker - worker (and the second worker would have nothing to do because it can't chop) or have our settler sit around for a few turns, we will already have committed the second city location before we reveal copper. so we can't have it actually settled and hooked up until the third city comes out which is in the t40s and therefore in the danger zone. if an opponent, through a combination of scouting and demo-scrying, discovers that we settled our second city before discovering bronze and around t30 (and we are sufficiently easy for them to reach), i think that coupled with what might generously be termed our "new player discount" would make us a good rush target. hard to put precise odds on that, but if we open AH first and neighbor someone who wants to rush i think we are a better target for them than the average of their neighbors, so maybe 88% * >25% * odds that they commit to a rush at all = something in the 10-15% range? but maybe "odds that they commit to a rush at all" is smaller than i give it credit for here...and, yeah, i am noticing that my OWN odds of committing to a rush are significantly lower than that, but my odds will go up if WE discover info like that about another player via scouting and demo-scrying

but if we get BW first, especially if we settle for copper but maybe even if not, those odds should be very low. not sure how low, but if it's WCs i think we'd only need a handful of impi to defend against them, and if i'm egypt and my zulu neighbor already has bronze when we make contact (which it sounds like they will know from the soldier count, plus we'll be in slavery lol) i'd consider them pretty much off the menu even if there is a possibility they haven't yet actually settled for copper.

re the foodhammers we lose from BW instead of AH, we start working the Capital Cow on t21 in the AH first plan, and t35 in BW-first (but after slavery flip which makes up 1t of the difference), and the marginal gain of working it over the deer is 2fh + 1c/turn. so long-term i guess we have spent something in the range of 1.5 capital chops (and therefore 4.5 turns of worker labor, but BW plan spams workers early so i think we make that back) to achieve whatever the odds deltas above are. was that, like, the /right/ decision? lol idk, but the fact that the "spent" hammers appear to only really materialize when we would run out of capital chops (so like t70) instead of right away makes me feel significantly better about it

^^ this is basically where my internal logic was when i committed to this yesterday but, uh, as always feel free to point out any cognitive distortions you may observe within LOL

@williams, if you see this, could i please have edit access to the spreadsheet i requested it as "swanfangcollective@gmail.com" as is my idiom lol

Rome is too slow for a Ancient Era rush if they are waiting for legions - nobody's going to get Iron Working and build them before we have BW in any case. GT is Ethiopia and probably not rushing us at axes either (I'd expect a classical or standard Knight era push instead, or even waiting all the way to Oromos). I think only Egypt is a reasonable threat to rush us in the time frame you are thinking of, both by the combination of an actually useful unique unit to do the rush with and the personality of the player running it, but even then, why attack an opponent when you can peacefully settle land? All you're doing is letting other people peacefully settle the land you aren't settling. Given how painful it is to take territory in Civ 4 and how unprofitable land war usually is, getting it without having to fight is pretty much always better.

I think the odds are heavily against us needing to defend a rush at T40. Maybe 5%? If Egypt is far from us I think it's 1% or less unless we provoke it with some kind of HIGHLY offensive settle. This isn't the map for that kind of rush. All it does is take you and your victim out of the game.  I'd heavily advocate we only consider attempting one ourselves if the player is totally unprepared (I could see India if MJMD is running a really heavy farmer's gambit, or Ricketyclick's Inca given he's a green and may have no idea what MP warfare takes) and even then I'd be surprised if it's a better play than simply taking the open land and coming back in a favorable era.

If you ask me what the odds are sight unseen a neighbor attacks us at Knights if we appear weak? I'd rate that at 75%-80%. That's the era where offensive warfare works well. That's the era we should be targeting for offensive war - get to Knights first and use them to roll over an opponent. We have some of the tools to do that - a fast start from matching techs, lower maintenance costs from barracks we'd want to build to boost our Knights anyway, a unit that makes people nervous of pink-dotting us to protect our marginal land on the border- and I'd argue we should be aiming to convert those to a tech and land lead as much as possible without fighting. Why take Zulu over, say, Byzantium if you want to attack at Knights? Because getting to Knights first is more important than having boosted Knights that your opponent has time to prepare for, and Byzantium is sloooow out of the gate.

[ljubljana]

wait do..... this is so embarrassing...... do barracks and stables stack for knights? iiiii kinda thought only stables gave XP to mounted units LMAO.... although i'm now remembering having this same discussion before 15 or so pages ago, oops

re rushing someone, i think this analysis is basically fair. one thing to add is, in what i think is the only instance of "not playing according to my sense of optimal play given the time and resources i have" that i'm generally willing to really entertain in these games, i am a little hesitant to rush specifically rickety out of their first game even if it seems plausible, as i kind of recall having been narrowly spared from something similar in MY first game

re settler at size 3: i did try it but i think a bit suboptimally; i'll rerun that and let you guys know

re scouting yeah, once all the first-ring tiles (and the now mandatory-looking floodplains region) are scouted i think that's a good next step

[aetryn]

wait do..... this is so embarrassing...... do barracks and stables stack for knights? iiiii kinda thought only stables gave XP to mounted units LMAO.... although i'm now remembering having this same discussion before 15 or so pages ago, oops

re rushing someone, i think this analysis is basically fair. one thing to add is, in what i think is the only instance of "not playing according to my sense of optimal play given the time and resources i have" that i'm generally willing to really entertain in these games, i am a little hesitant to rush specifically rickety out of their first game even if it seems plausible, as i kind of recall having been narrowly spared from something similar in MY first game

re settler at size 3: i did try it but i think a bit suboptimally; i'll rerun that and let you guys know

re scouting yeah, once all the first-ring tiles (and the now mandatory-looking floodplains region) are scouted i think that's a good next step

Yes, Barracks and Stables stack for mounted units. I'm also fine with not rushing Rickety - i probably wouldn't either - but it's one place where a rush might well catch someone totally pants down.

[ljubljana]

AH first, settler at size 3 notes

so, i guess one reason i initially didn't like this option is we grow to size 3 significantly before there's a third tile to work while not actually producing another unit in the meantime - here's the situation when we hit size 3 at turn 22:



2h short on the scout, only 1f overflow, so i don't think there way a way to hit both benchmarks, and we have 6 turns before the deer is up. at this point i will swap to settler for growth curve reasons, maybe in the real game i'd finish the unit idk [EDIT: in the real game we definitely would, it turns out that in this game production you spend on a unit stored in the queue apparently decays over time??]. a second worker doesn't seem correct as they will run out of things to do before bronze comes in



turn 32 we found the city and the capital still doesn't have 3 improved tiles to work as the second city steals the cow. will be fixed in 2t when the ivory camp completes. BW is in in time for us to settle for copper and revolt to slavery while the settler is in transit which, yeah, is in noted contrast to any other plan



second city will have a few turns working an unimproved cow... but the alternative was doing a second worker before the settler which would have left us in this situation



= 4 wasted worker turns which seems incorrect. but the BW plan also has the second city working an unimproved tile for 2t so in this respect we're just at "not better" rather than "worse". back to the mainline sim:



turn 42, if we go for the same number of workers before the third settler (4) we are still a bit behind BW, but should catch up as the deforestation mounts  third city being slow to found means it will be slower on its granary too and not getting effective double food from the wheat, but in this case the capital is working it so maybe that's fine



third city steals the wheat and pushes the capital onto an unimproved tile, but that might not happen in production if there's more food in the fog. as it stands i think i'd probably whip the end of this settler and overflow into granary



turn 50, one city has the granary but the others are a bit behind, fourth city is missing a TR, etc. i think this picture is basically consistent with having saved about 1.5 chops over BW first, but being behind a bit on worker labor, so that we actually have 3 extra forests left instead of 1.5. 

hmmm, idk, i think size 3 vs 2 may depend on what's in the fog at the third city, especially if it's good enough to make that the SECOND city and let the capital keep the cow. i am hesitant to grow the capital before the FIRST worker/settler build onto an unimproved tile, but maybe we should squeeze it in there somewhere depending on what we find...



as i am trying to convert our extra chops into extra settlers and workers....uh, the capital's borders expand and now we have more chops. this is why i say that the lost chop hammers don't manifest until ~t70 rather than t50, because in BW-first we get more chops available just as we're starting to run out of them from the fourth city and the capital border pop. but maybe that means i should have more workers instead of one or more of these granaries/settlers



turn 60, by which point the lines look pretty converged. at this point, both here and in the BW sims, i'm mass-building cottages instead of using the new chops - no idea if that's correct (aetryn, any thoughts?) maybe that means both of these plans need a fifth worker sooner than i've been making one...

edit: whatever the answer would be in isolation, surely it makes sense to not be chopping if we're actually 9 turns from mathematics at this point lol

[aetryn]

One thing we probably do not want to do is steal all the capital's food at any time - and the deer tile probably isn't good enough that I'd consider that enough. I think it's okay for the cow OR the wheat to be borrowed by other cities but we shouldn't have the capital without one of the two tiles. Even when it's at the happy cap we want to grow OVER the happy cap so we're not whipping off productive citizens when we hit things like Markets and Courthouses (and Knights) that we want to produce immediately. I suppose there might be a case for a few critical turns somewhere while another city gets started, but really we should be looking to have our capital be our strongest city and that means growing and that means keeping its food.

The bare ivory tile is good enough that I don't mind working that unimproved for a few turns. Inevitably when starting up a city it ends up on unimproved tiles for a few turns - even if you had workers in position, usually you can't improve the tiles before they are in culture anyway.

Do we expect another city to be a worker pump? A worker pump city can probably delay a granary for a while (since it will be continually producing workers it won't have any food for growth and won't be whipped much - just whip the last worker and then grow into a granary). I'm not sure we have any city sites with enough productivity to really manage that at this point, but it's something we should keep in mind also.

[williams482]

You should have write access to the spreadsheet now. Let me know if that's not the case still.

Is it generally considered wise to chop in the 3rd ring of the capital to speed up early expansion? There's a pretty severe cost for distance from the destination city, and that's on top of the penalty for chopping before math. is 16h(?) per chop on t70 really a better investment than 30h into slightly later satellite city infrastructure, or knights for our military push?

[aetryn]

Interesting, thanks for trying. Each approach has inefficiencies.

A slight improvement could be that instead of the ivory the worker goes directly for the new cow. Should be able to start it just as the second city is founded. Meanwhile Ulundi can work all improved tiles until T33 and then whip 3->2 worker. After that I'd grow it to 4 so it's ready to whip a granary when pottery is in.

Yeah, I like this one a lot. Also gets the second city on its feet fast.

Also, units in the queue decay but only after 10 turns of delay. The interface will warn you when you're getting close and when you're actually losing hammers.

[ljubljana]

screenshot provided to illustrate: we flip to slavery, start the cow pasture, and found the second city all in the same turn! which, i guess means we're actually 1t late on slavery to avoid wasting a turn at city 2 (BW just came in this turn)... but we're planning on doing that anyways under BW-first

You should have write access to the spreadsheet now. Let me know if that's not the case still.

yes i do, thank youuu 

btw: https://drive.google.com/file/d/1T_EDL6s...sp=sharing in case anyone wants to mess w the sim

[ljubljana]

turn 42 of this line, whipping 3-2 worker and then growing the cap to size 4 by the turn pottery comes in



hmm, seems a little more unit-y than i would like (not visible: an ikhanda completed in the capital in addition to the 4 units you can see). the whipped worker is nice (converts 26 food to 30 hammers even without granary, a fact that in my civ4ing so far i have already become very good at forgetting) but it seems like the turns we spend regrowing and not making workers might make it net to being counterproductive

we are off by quite a bit.... it looks like even if we worked the ivory the whole time we'd still be 2 short:

[ljubljana]

also we have a Turn 11 - Zululand



LOL you have GOT to be kidding me

thoughts on where to scout next? i'm assuming next turn i'll end on the gold hill but i think whether i go SW-SE or SE-SW depends on if we're turning east or west after this (respectively). given that the water coming into view seems like it may continue to the east i'm guessing west is most likely... motivation for east would be if we want to take the long way around and end up scouting further up Cottage River in search of accessible neighbors to murder. but given how far away we are that may be a task for scout 2 or even 3

note that in some important ways this floodplain area might be better allocated to two cities - a city 1NE of the scout's position reclaims 6 windmills we would otherwise likely never have access to. but that misses the gold, which is absurd, and with the gold exactly 8 tiles from our capital i'm guessing one city down here is likely to be all we can reasonably expect to get. unfortunately this area seems to pretty much guarantee SOME kind of conflict with a neighbor in the fog 8 tiles away in another direction.... fortunately as AGG zulu we're in a good position to win that conflict, but this land might be good enough that even taking it with a forward settle could push our opponent into gamelong vendetta mode...

here is another omen of fish