Yeah maybe it’s more accurate to say that Industrious helps Huayna, because it helps him finish stupid wonders faster.
|
Civ4 AI Survivor: Season Four
|
|
I have to take this opportunity again to complain about how Deity difficulty skews the AI behavior and competitiveness. In Civ 4, the AI cost discount does not apply to wonders, to give the human player more of a fair chance at building some. Wonders are therefore considerably (1 / 60% = 1.66x!) more expensive relative to every other building item on Deity difficulty. Any leader whose personality or trait prioritizes wonders puts himself at a significant disadvantage compared to leaders who stick to discounted items. Industrious is poor not in itself but as a symptom of that.
(An example of the numbers: Oracle costs 150 on all difficulties. A monument costs 30 unmodified, or 18 on Deity. Oracle should cost 5 monuments, but instead on Deity it costs 8.33 monuments, which is relatively 1.66x more. Anybody building any wonder is wasting 40% of his production compared to building a discounted item.)
Yeah, of course when we talk about good or bad leaders in the context of these games, we're talking about the specific conditions under which they are being run. Running the games on Deity difficulty has a number of different effects that advantage certain traits and certain leaders, and disadvantage others. (Running the games on Deity is pretty essential though - they go *MUCH* slower on lower difficulties, and many of the games would end in an uninteresting Time finish after 500 turns.) However, we could say the same thing about any number of other settings aside from difficulty level. If we used a different map script with multiple continents, the more peaceful AI leaders would win significantly more often. If we turned on vassal states, we'd see far fewer eliminations and some of the AIs would do much better. If we didn't turn on Aggressive AI, we'd see far fewer wars and much more peaceful games. And if we didn't turn off tech trading, we'd be playing in a completely different world entirely.
All of the settings that we pick are arbitrary to some extent. Through trial and error, we've found settings that work well for this AI competition and consistently deliver fun and exciting games to watch. On that note, we had 181 entries into the picking contest today, once again setting a new record. I'm pretty happy with how things are working out.  Fluffball: as yuris said, difficulty level can be changed in a Worldbuilder save file. I plan to make them available after we get done with the current season. Here's a batch of links for the next game: Main Season Four Webpage Sullla Twitch Page Sullla Discord Channel AI Reference Guide from CivFanatics Game Eight Livestream Next Game: Wildcard Game Schedule: Scheduled for Friday, 14 June 2019 at Noon EST Wildcard Game Map Setup Video Preview Wildcard Game Written Preview Wildcard Game Picking Contest Entry Form Congratulations to BlueOrchardBee, Steez, and artdeco as the winners of Game Eight's picking contest with 18 points.  Scoring for the group as a whole was fairly average in this game, right in the middle of the spectrum. We are continuing with our normal Friday date for next week's Wildcard game. Who's going to take advantage of their last chance to punch a ticket into the playoffs?
(June 7th, 2019, 06:56)Sullla Wrote: I disagree about the Industrious trait; I think it's one of the weaker options in the hands of the AI. Who gets what wonder is almost entirely determined by which AI happens to be leading in tech, not by who has the Industrious trait. And the overall cost discounts are large enough on Deity that the +50% production bonus (only on wonders) is fairly insignificant. The double speed production on forges is probably the better half of the trait for the AI games. This made me think about the other AI modifiers, since the Industrious trait is not made equal (for example, Stalin only has a build wonder weight of 20 (out of 50) compared to Ramsses who has a wonder construct weight of 50). Alongside the traits, each AI personality has 'strategies' that it uses to help push it in a certain direction. These are Gold, Religion, Military, Culture, Growth, Production, and Science. Typically these are given as a primary and secondary strategies (so Stalin has a primary Military strategy and a secondary Production strategy, with a weight of 5 and 2 respectively) though some AIs are all in on one strategy (like Gandhi or Temujin who value Culture and Military respectively with a weight of 10). Here's a couple of tables relating the respective strategies when compared to Sulla's power rankings:   It definitely seems like production is the hands down best strategy, but more as a secondary strategy. It's probably great as something to help certain AIs stay balanced. You can tell that when Civs don't have a secondary strategy, they tend to go off the rails (the X strategy implies no secondary strategy, and civs without these are more famine than feast. Growth also seemed like an okay strategy to have, but it didn't do nearly as well as I thought it would. With that said, I don't know the specifics how an AI prioritizes 'Growth' as a strategy (could mean that, for osme of them, they try to build vertically as opposed to expanding). Science was a big fat failure though. Seems like it might be too easy for those civs to get bullied and they never get to take the tech lead that they want (conversely, Gold as a strategy seems to do the same thing but better). We'll have to cross reference these with Build Unit/Peace weights to see if it's just that the Science strategy is bad, or if it's more a correlation with never building units (or both). Lastly, the power rankings referenced here are based at the end of last season, so we'll have to see how everything fairs at the end of season 4. I'll put these tables on google sheets when I get home if people are curious to look at them. (I have a second table that lists ranking/civ/trats/leader/wonder weight/strategies/build unit weight/war decc weight).
I am excited to announce that through the first round of the competition, Fatkid, random.org, and I are all in contention for the coveted Germany award: a participation award for those of use who fail completely. The three of us are neck-in-neck at a total score of 45 points. That's an average of 5.63 points per game!
Why a Germany award? The other Civs with a score of 0 in the power ranking have given us wonders like Sitting Out Bull, Hatsheput of North Korea, Victoria's Nuclear Exchange, and Washington's Apostolic Cheese Survival. Nothing can compare to the ineptitude of the German civilization, who also dazzled us this season by ... dying second in both their games. They're honestly not even worth picking for first to die any more! I'd like to thank my wife and children for their encouragement as I attempt to fail my way into the annals of history. |
![[Image: survivor3-14-4.jpg]](http://www.sullla.com/Civ4/Survivor/S3/survivor3-14-4.jpg)
