[SPOILERS] swance bitten, twice shy

[ljubljana]

uh oh, thinking 

im still asleep but delirious first impressions in-between stress-dreams:

tokugawa and similar wetboys look out  with otherwise-balanced trait options there's no way we should take fishing over something that is actually on the way to where we want to go in the opening

this looks favorable to india + early the wheel, those forest positions are beyond almost-annoying (and therefore into "actually annoying")... one wasted worker turn between wheat and cow, two wasted turns between cow and deer, and then even more to bounce around to all those third-priority tiles until we get the forests roaded? seems gro55

with quite a bit of free commerce from city center + both foods on the river, plus the wasted worker turns to move between food sources, being tech-gated in the extreme early game seems unlikely and we may end up with extra beakers to play with. perhaps (agriculture + mysticism start +) an early religion is favored relative to a random start? and the double rivers with one leading to the coast look okay from a maximize-early-spreads perspective (but then again highlands is otherwise bad for this) (and maybe i should be skeptical that even with 3 free gpt we can do all that and still have hunting and AH online in time)

"marble means wonders", but at first glance none of these here early marble wonders read as especially worth prioritizing except maybe for mr-hard-to-quantify himself, the oracle? in an MP context it seems a little unlikely that he is sandbaggable for anything reachier than, like, monarchy/CoL/metal casting.... maybe something likeeeee sumeria into oracle MC into (have been ind/org this whole time and) whip ziggurats and forges everywhere could be worth a look?

ivory is our friend, i think? as boosting the happy cap to size 5 for not-the-capital enables at least some settler triple-whipping. which in turn means cha is relatively /not/ our friend since we've already avoided its most compelling use case of a total early-game happy desert wherein only the capital can ever whip settlers?

baseline sim coming someday (= to-day, some-time)

myuhhh anyway back to bed before i say anything even ludicrouser lol

[ljubljana]

omg i know i knew this would happen but yall civ4 players are SO smart and knowledgeable about detailed mechanics questions omg <3

so wait, libraries are not worth making (early) in most cities, but the similar-impact but almost twice as expensive markets are? that...that doesn't mean everyone's running around with an expansion-crashed economy and 80% gold for most of the game, right? i'm surprised if so as i think in SP it feels way past the point of diminishing snowball returns, and here we'll be even more cash-strapped due to aforementioned gigantic piles of military...

re "what we want our traits to do", idk really, but internally i've been vaguely following the Thrawn School thus far, of such thoughts along the lines of "if all trait combinations are compatible with a winnable midgame due to general balance and the self-balancing impact of diplo working against early leaders, then the best traits are ones that most quickly translate a strong push in said midgame into an unassailable position in the endgame". in my head that counsels for econ and military traits over early-game snowbally stuff like exp or cre, but what do i know really lol

and i think fin and other low-skill cap traits are indeed good for me if all else is balanced, not because i don't want to play something challenging but because i think ruthless maximization of win probability at every decision tree branch requires accounting for the fact that in expectation i am less likely to play hard traits well enough to compete w easy traits in a first game... and re "fun factor", point taken... for me, the funnest thing is ruthless maximization of win probability (or as ruthless as i can from a knowledge- and competence-limited position, anyways) so i should rate FIN more highly for my purposes than picking statistics suggest

[aetryn]

omg i know i knew this would happen but yall civ4 players are SO smart and knowledgeable about detailed mechanics questions omg <3

so wait, libraries are not worth making (early) in most cities, but the similar-impact but almost twice as expensive markets are? that...that doesn't mean everyone's running around with an expansion-crashed economy and 80% gold for most of the game, right? i'm surprised if so as i think in SP it feels way past the point of diminishing snowball returns, and here we'll be even more cash-strapped due to aforementioned gigantic piles of military...

re "what we want our traits to do", idk really, but internally i've been vaguely following the Thrawn School thus far, of such thoughts along the lines of "if all trait combinations are compatible with a winnable midgame due to general balance and the self-balancing impact of diplo working against early leaders, then the best traits are ones that most quickly translate a strong push in said midgame into an unassailable position in the endgame". in my head that counsels for econ and military traits over early-game snowbally stuff like exp or cre, but what do i know really lol

and i think fin and other low-skill cap traits are indeed good for me if all else is balanced, not because i don't want to play something challenging but because i think ruthless maximization of win probability at every decision tree branch requires accounting for the fact that in expectation i am less likely to play hard traits well enough to compete w easy traits in a first game... and re "fun factor", point taken... for me, the funnest thing is ruthless maximization of win probability (or as ruthless as i can from a knowledge- and competence-limited position, anyways) so i should rate FIN more highly for my purposes than picking statistics suggest

Generally I expect to spend the early game at around 30% science and the midgame outside of GAs at around 50% science, and I'm probably not aggressive enough in expanding. Ruthless maximization of win probability requires riding the line of economic collapse as closely as possible. If you wait to take the land when it's economically convenient it won't be there to take, someone else will have settled it already. This is FAR less true in single player, as the AI is a derp at many difficulty levels and won't settle aggressively even if you cheat in free settlers for them via World Builder. Currently playing a hot seat multiplayer game on Monarch with 5 AIs, and the players are all at 12+ cities, while no AI has more than 8 and France is still on 3(!). 

Markets give you more productive citizens because they increase happiness, which almost every city wants. A city with almost no commerce outside Trade Routes that has a lot of mines or lumber mills to work isn't going to need a Library, not because it's totally pointless but because it's not that effective and production cities need to be helping your empire, not building endless infrastructure. But it still wants to WORK those mines or lumber mills and it can't do that if the citizens aren't happy. You already are guaranteed ivory since it's in your screenshot, and it's pretty likely you'll end up with either Furs or Silks, so +2 usable population cap from a market on top of the gold bonus, which never hurts.  There's also a post around somewhere that talks about how gold multiplier buildings are often functionally more impactful to science than science multiplier buildings, but I'm not that deep in the weeds on mechanics to really speak to it.

While it's generally true that diplomacy effects will help corral an early snowballer, it's pretty rough to count on it in an individual game, where geography or circumstance may make it hard for some players to respond. With diplomacy through the in-game interface only, it's also REALLY hard to coordinate effectively. Or heck, even with full diplomacy on - look at PB70 where MJMD had the dominant early game, built Hanging Gardens with an absurd number of cities, got attacked by 3 of the other 4 players, and still came out ahead, because even WITH diplo, different people had different ideas about when to go in, how much to send, how critical the war was, etc. I'm not saying you should build for an early snowball, but don't discount it either - there are plenty of players who prefer that over a long-term advantage

[ljubljana]

Would that be banks produce beakers? The logic there only applies with science slider under 50% IIRC but that sounds like the normative situation anyways. Which is interesting...I mean, expanding as quickly as possible while still on the right side of the line of "impede ability to reach next econ breakpoint" is good SP strategy too, but on Deity 30% science has been well south of that in testing. On Deity difficulty, which this isn't, so it makes total sense that we can skate by with lower % science here.

baseline sim coming someday (= to-day, some-time)

promises made: 1, promises broken: 1

[Tarkeel]

Libraries vs Markets vs Grocers is an interesting discussion, as they all have a similar main effect: +25% to one type of commerce and two slots of the appropriate specialist. The differences lies in cost and secondary effect; Libraries cost 90h, while Markets and Grocers cost 150. The secondary effect of Libraries is culture, while Markets boost happiness and Grocers boost health.

I agree with Aetryn that you're likely to build Markets in more cities than Libraries, but that doesn't mean you're going to build them earlier. In fact, the first Library is usually built pretty early in a high food city to benefit from the specialist slots, to generate an early GP for a GA or Academy. You're also likely to have Libraries as the border popping building in cities if feasible, but that should still be a city with some commerce and/or food. Markets on the other hand come a bit later when you both need the happy more, and have enough of the resources it doubles. Lastly you're likely to build Grocers during the industrialization, when health becomes a premium as well.

Banks on the other hand are 200h for a 50% increase, with no secondary bonus, and is one of the most efficient commerce multipliers that you actually want to build in most cities. Universities are a lot worse, since they cost the same but are only a 25% increase.

[ljubljana]

I agree with Aetryn that you're likely to build Markets in more cities than Libraries, but that doesn't mean you're going to build them earlier. In fact, the first Library is usually built pretty early in a high food city to benefit from the specialist slots, to generate an early GP for a GA or Academy. You're also likely to have Libraries as the border popping building in cities if feasible, but that should still be a city with some commerce and/or food. Markets on the other hand come a bit later when you both need the happy more, and have enough of the resources it doubles. Lastly you're likely to build Grocers during the industrialization, when health becomes a premium as well.

Banks on the other hand are 200h for a 50% increase, with no secondary bonus, and is one of the most efficient commerce multipliers that you actually want to build in most cities. Universities are a lot worse, since they cost the same but are only a 25% increase.

i see i see, this basically makes sense to me yeah, and is about what i'd expect if one took an SP game and removed the least essential 50% of infra builds there with military lol. universities in particular seem iffy from a balance perspective even in SP (where it seems like only Oxford makes them worthwhile), hard to see how the building that boosts the lesser-used currency by a lesser amount can get away with being the same cost

of course that's still a tiny nitpick compared to civ6 where the list of worthwhile buildings is more or less

- monument
- walls
- lighthouse
- market just for the trade route
- granary in rare corner cases
- allegedly library but i was never in a position to make them in 3 games
- buildings specifically buffed above baseline by religious beliefs/civ abilities/etc (and we lost PBEM22 in part by making even these instead of units)
- maybe factory and shipyard if the game is still undecided after 150 turns, which has happened like twice

Hi guys Magic 8 Ball
While I appreciate the enthusiasm for more Sumo content, taking on the role of a daily turn player for an unfamiliar game and producing extensive Sumo essays simultaneously might lead to burnout.

i'm worried about this too tbh let's see if i get genuinely excited about it between now and game start. another obstacle is that i think i peaked in sumo-engagement in 2022 and might not be as competent at blogging about it now.... but the *story* just *needs* to be told

[ljubljana]

Let's hold on the picks and sims

yayy procrastination rationalization

(may sim today anyways depending on How I'm Feeling)

[ljubljana]

baseline sim, and i even changed my civ language back from frapanese just because randomly failing to instantly parse info didn't seem win-probability-maximizing lol

attempt #1!

leader is Tokugawa (no econ bonuses from traits at the start) of Zulu (agriculture/hunting starting techs, plus i think this civ is indeed on my shortlist for serious consideration come actual picking). you'll notice that it turns out i am a giant idiot and starting techs are tied to civ and not leader lmao

tech order AH -> the wheel (an incorrect choice it looks like, making roads this early with one worker didn't feel worth it) -> mining -> bronze



i got through 2 warriors and the second settler comes out on t31. this seems quite bad! but that's why it's called the baseline

something seems wrong with my setup though as the tech costs still seem too high? according to the changelog mysticism should cost 60 beakers and yet:



tech sizes are map-size-dependent yes? should i be assuming large map size costs? or are we working with a custom size with customly-scaled costs? slash, maybe this is just normal for 60 beakers * monarch * large? i confess i haven't looked into the formula yet and am a little intimidated by the prospect of doing so...

edit: and here is what would happen if we did the second worker at size 2 working wheat+cow and correctly open ah-mining-bronze:



too many improved tiles to work, workers with nothing compelling to do for 2t until bronze comes in, and no settler even started at t30. ok, even though a million chops will come in after this once we hit bronze i imagine this is what's known as "below baseline"... however i think i'd better go out to like t40 before bindingly making decisions like that

trying to figure out how to get this thing to run in borderless window fullscreen on my second monitor so i can take screenies, write, and play at the same time. i do not know how to do this yet hence the notably non-borderless screenshots above...

according to CFC tech cost should be: floor(60 (base cost) * 1.15 (monarch modifier) * 1.4 (map modifier)) = 96 as observed. that's expensive, wow! ok, very little chance a religion opening is going to make sense if these are the true costs

[Tarkeel]

tech order AH -> the wheel (an incorrect choice it looks like, making roads this early with one worker didn't feel worth it) -> mining -> bronze

Early wheel is mainly to give the worker something to do while waiting for other techs to come in, and build half-roads so as to not "waste" turns while moving between resources. It's rarely worth researching that early unless you're going for chariots.

something seems wrong with my setup though as the tech costs still seem too high? according to the changelog mysticism should cost 60 beakers and yet:

tech sizes are map-size-dependent yes? should i be assuming large map size costs? or are we working with a custom size with customly-scaled costs? slash, maybe this is just normal for 60 beakers * monarch * large? i confess i haven't looked into the formula yet and am a little intimidated by the prospect of doing so...

The actual tech costs scale by map size, and there's no feasible way to make custom sizes for a map. Map size has yet to be confirmed, but it usually is normal but large is a possibility since you're 10 players.

trying to figure out how to get this thing to run in borderless window fullscreen on my second monitor so i can take screenies, write, and play at the same time. i do not know how to do this yet hence the notably non-borderless screenshots above...

I'm not sure that's possible. I run bordered window to do the same thing.

[ljubljana]

Tarkeel - Seems about right re the wheel (plus, I guess, turning on trade routes at the second city, but I imagine that's not worth it before BW even for PRO...?) Re bordered window, nooo that's what I was afraid the answer was. So do all your IV screenies have the windows menu bar at the top of them, or is there a ton of labor-intensive cropping work involved

thrawn - that was a short month 

ok, point taken about the thread title - certainly it was /consciously/ intended as more of a tongue-in-cheek tie-in to williams lurking here again, but that doesn't mean it couldn't subconsciously become a source of defeatism, which would be not ruthlessly maximizing etc.... i'll do a brainstorm. and also at least run a practice game with spi - taking it would probably be better for my growth as a player long-term, but in the short term i've literally never played a full BTS game with the trait and perhaaaaps i should be interpreting PBEM norms of "best attempt at optimal" play as weakly suggesting that i "should" optimize for success in /this/ game where possible

HRE i've been taking a look at as well... as the rathaus seems to have much more staying power than the UU i'd think the obvious synergy pick would be something like org over pro, though? who's org/spi, asoka? this could be insufficient economically though if the only hard econ boost is whipping half-cost rathauses, in which case we could split the difference with org/pro, org/agg, ofg/fin, etc. although i do really like spi for a mysticism + religion start because of nearness to orgrel and monarchy as you've said. will we have enough beakers to rush both monarchy and CoL quickly? maybe not buuuut the combo we're considering puts us really close to the oracle and there's marble right there at the capital............. (and is also a backup way to pick up religion, as has never been relevant to me in SP before but i guess i should know lol)

i think i'll have to sim out religion openings once starts and map size are confirmed to see if early meditation is viable.... unfortunately i don't think we'll have time to do it before agriculture if we take HRE, tests suggest this will leave the worker sitting around for a few turns with nothing to do which i'm told is a no-no lol. i'm not really suuure there will be space before AH if the overriding goal of the opening is indeed to improve all the food resources ASAP. but maybe it could work if we take agriculture + mysticism and open like meditation - AH - hunting? i'm seeing that inca are the unique civ that start with that combo, and they seem reasonably playable at a glance but we'd really need to take EXP to make them work i think. as much as i like the rathaus, i'm not sure it makes sense to take HRE if an early religion is unviable due to tech-gating in the opening, surely we need to get more value from our starting techs than the ability to make monuments (which we most likely won't strictly /require/ until the first non-capital-food-stealing city i'd imagine). or maybe i have the wrong idea about just how "early" the early religions go in IV, and it actually is possible to open agriculture - AH - (religion tech) and have it still be on the table?

re keeping armies as pop instead of standing troops - that'd be preferable, but i'm getting an ominous sense (that IV players should ofc correct me on) that the numbers involved wouldn't work out... if the average player in these games ends up with like 10-12 cities post-landgrab, and the talk i'm seeing upthread about attacks with like 32 knights and 20 catapults is representative (potentially with just one turn lag between when war is declared and when 2-movers can hit a border city), it sounds like whipping units will be an essential source of units but we'll need to have significant standing forces ready ahead of time as well to not die to such a stack even if we see it coming. also, this could be wrong but i thiiiink spies are banned/off in this game, in which case we might not have much warning outside of the power graph