4 ms·

marvelous, sad game I remember being frustrated/saddened enough to dive into source code to find optimal strategy, and I found that even with this knowledge ga

marvelous, sad game I remember being frustrated/saddened enough to dive into source code to find optimal strategy, and I found that even with this knowledge game kept my interest: - each "year"(second) each death reason rolls its probability. If civ survives, there's also per year chance of developing tech on its own. It's optimal to advance tech asap - technologies and a couple of lategame events change these probabilities - usually by removing small chance of early reason and adding new higher chance for later reason. It's mostly optimal to develop techs in clusters that cancel most of each other out and advance self-tech chance. - tech tree roughly separates into 3 ages: early (up to and including sailing), middle (up to and excluding "modern times" with biology, transistors and newspapers) and late early tech has (1)writing+math+astronomy techs that don't give additional death chance in any combo, but increase tech chance - research first (2)agri+fish+tools and fire+construction, metal+sailing that have higher death chance when you have all but 1 or 2 of them than if you have full cluster - focus if self-tech triggers (also agri and fish before tools) (3)architecture->plumbing decreases death chance - focus middle tech doesn't change death chance, so you have a breather late tech is most deadly - (1)taxonomy->germ->genetics tech branch has great death decrease on 2nd step, but last one gives high chance of bioweapon death (so don't. you'll fear self-tech doing it too); (2)exploding nukes event increases world war death chance, world peace event decreases it; (3)Ai is great danger but necessary - do it as late as possible (iirc there's like 1/3 chance of death on these last 1-2 steps) When you learn all the numbers it becomes a great "tension - breather - tension" game, but still not with that high of a winrate. Nearby supernova death having a chance all the way through is kinda funny --- fun fact - I learned math behind Markov Chains just solve this game. Civs have like 1e-12 or 1e-15 chance of survival on their own, iirc (it was something very small, I don't remember now)

Supermancho 3y ago
> marvelous, sad game Only sad when you can seriously believe the various scenarios that play out...like teaching fishing could cause a civilization to cease to exist. It's a silly game that has no bearing on reality.
- NooneAtAll3 3y ago
  a) empathy has no reason to only work in reality - or whole genres of fiction wouldn't exist b) it's hard to judge what's possible or not without experimental data (on otherworldly civilizations), nor we're even explained the mechanisms of "us teaching others". But it's a schematic way of showing Great Filters in a simplified, gamified manner - and that's still a possible, valid and argued for answer to a Fermi Question
  - me_me_me 3y ago
    > b) it's hard to judge what's possible or not without experimental data (on otherworldly civilizations), nor we're even explained the mechanisms of "us teaching others". But it's a schematic way of showing Great Filters in a simplified, gamified manner - and that's still a possible, valid and argued for answer to a Fermi Question Exactly, the whole discussion is already extrapolation for sample of 1. Who knows how civilization would develop if we were more like bonobos less like chimpanzees.

Supermancho 3y ago

> marvelous, sad game Only sad when you can seriously believe the various scenarios that play out...like teaching fishing could cause a civilization to cease to exist. It's a silly game that has no bearing on reality.

NooneAtAll3 3y ago
a) empathy has no reason to only work in reality - or whole genres of fiction wouldn't exist b) it's hard to judge what's possible or not without experimental data (on otherworldly civilizations), nor we're even explained the mechanisms of "us teaching others". But it's a schematic way of showing Great Filters in a simplified, gamified manner - and that's still a possible, valid and argued for answer to a Fermi Question
- me_me_me 3y ago
  > b) it's hard to judge what's possible or not without experimental data (on otherworldly civilizations), nor we're even explained the mechanisms of "us teaching others". But it's a schematic way of showing Great Filters in a simplified, gamified manner - and that's still a possible, valid and argued for answer to a Fermi Question Exactly, the whole discussion is already extrapolation for sample of 1. Who knows how civilization would develop if we were more like bonobos less like chimpanzees.