New AI Survivor from Sulla

[lymond]

Have to watch replay tomorrow..hard to watch live for me.

 

[Nogbadthebad]

I was wondering- Hatty was certainly building a lot of units before Shaka unsurprisingly attacked. Given that Asoka at that point had fairly few units Hatty could have been planning to attack Asoka while he was weak, which at the time would have been a good move for her. Obviously whilst the war with Shaka was on, Asoka went oracle-mon-HR and threw out units for happiness, so the window had closed.

Is it possible that Hatty's AI was still planning to attack Asoka, and held on to this plan through the Shaka war. Can the AI do this?

 

[Johnstrain]

Unsure if that was planned by HATTI or not: its fun to watch and you learn how AIs behave but...the stupidity of the AI! Sullia sets the game up to learn about AIs in general. in Five played out SEASONS of this series a few changes were put into play to solve problems of viewing the results. And the entire 4 seasons played out are available either on utube or twitch or discord.

 

[Fippy]

I think all plotting AIs do is cancelled if they get declared on.
An exception are refused demands, then they might still come back later.
(but only if they asked threatening)

 

[Johnstrain]

Some AI clashes do cancel out; however some don't - watch a few from the series and see - most are 4 hours long for the first 3 seasons of AI Suvior play

 

[Nogbadthebad]

Some AI clashes do cancel out; however some don't - watch a few from the series and see - most are 4 hours long for the first 3 seasons of AI Suvior play

Iv seen most of them i think- started watching it towards the end of series 1 (am a bit worried about how long ago that was). It just left me wondering how long an AI will keep a war plan against you even if you bribe someone else to attack them first.

 

[lymond]

I picked Boudi in this one with Mao as runner up. It's an odd bunch here. I think Boudi is decently balanced while still being aggressive. I'd like to see her win won of these and this one may be it. I'm reaching here and having HC die first.

ha..first game I've picked this season as I keep forgetting.

 

[Nogbadthebad]

I couldn't bring myself to bet against HC- he always seems to find a way. Put Boad down for second for roughly the reasons you mention, though Mao has the ideal runner-up traits of protective/well out of the way. Finish date is tricky though, the ice seems to be slowing this season down a lot.

 

[Fippy]

Boudi played one of the worst games i ever saw, even forming stacks with warriors
If they skip wheel they are always in bad shape, and in Boudi's case..terrible shape

HC really benefited from having Monty close, honestly he's worse than Lincoln.
All warmongers are pretty terrible under those settings and Monty gets further crippled by having no real Swords.

Augustus also bad..he made 2nd place but geez did he play terrible after grabbing the most land, any good leader might have stolen this from HC just by absorbing Mao and not ignoring military techs.
And Lincoln always does well early with okay starting techs and while units are similar, then he goes into Democracy and Physics lalaland and gets eaten.
Feels a bit like watching Groundhog Day, they are really predictable at this point. Sometimes one gets lucky but most never have a plan and HC often does so..

Next game could be a good chance for those who want to make up scorepoints, Justi will be picked by 66% or so and his start looks terrible.
HC had a good cap this game, but Justi wow he's coastal and tundra close already with nothing good visible at all.
Peter has great land (like 3x of what Justi will work with) and is at least average as AI leader..a good pick here.

 

[dankok8]

Yea saying Boudica was bad is an understatement!

Augustus is a peaceful AI with Praets. It just doesn't fit well together. If this was Julius in his place, he would win Domination. Especially since he would build more units and completely steamroll Boudica the first time. Gilga and Mao are pretty strong leaders but Mao struggled with the seafood start and then being dragged into the abyss by Boudica and Gilga just froze. I can't explain him not doing anything after 60 or so turns.

 

[lymond]

Boudi let me down again Man, Biz was pretty terrible in that last game early.

Ok..I'll not bet against HC again..ha..just always root against him.

It would be interesting to figure out why some Leaders do better than others regardless of traits. Some combinations of their coding that just works better. Like Stalin seems to do very well but on paper does not seem like he would be so good. I think not being able to declare at Pleased hurts a leader in this type of combination unless somehow they just luck out and stay safe.

I think the maps could be a little better balanced. It would be interesting to see how things fared in this competition is some leaders did not start in such terrible situations. No question thought that AIs make some terrible and baffling decisions early like Biz not getting TW until waaay late.

Actually watching the game now. Wow...Sullla actually picked Boudi which I found a bit surprising.

 

[Johnstrain]

It's a matter of coding and whom is neighbor to whom that decides war declares. The map is purposefully set to start wars and fun on Mr Sullias machine.
That said the thing with some leaders is they have no clue as a human would how to proceed on some cases.

Its possible to play out some games yourself.
I agree a stack of warriors used that way was....dumb...
Furthermore the AI has but limited comp power to make do.

 

[dankok8]

To be honest, the sample sizes in these AI Survivor contests aren't nearly large enough to provide reliable data but they are in good fun.

 

[Nogbadthebad]

To be honest, the sample sizes in these AI Survivor contests aren't nearly large enough to provide reliable data but they are in good fun.

On the subject of sample size, I do hope someone gets to replay yesterdays game a few times to see how often that happens (no spoiler).

RIP my average score

 

[dankok8]

Alex deserved that win. The way he smashed hopeless Zara was almost what a human would do and then the pariah Churchill who everyone hated. If he didn't keep his 100-unit SOD waiting for so long during the China war he would have won even much earlier. It's one of the most dominant performances in AI Survivor history. Realistically speaking, Justinian played a great game improving his seafood and his tech choices were mostly spot on. Oracling Monarchy was huge. China with good starting techs also opened really well and even settled nice land but didn't have any happy resources or religions for a while. Peter and Churchill also started well but the latter was pretty much everyone's worst enemy. This was actually a quality game from an overlooked group of leaders.

I picked Peter to win IIRC or at least to finish second. He had a lot of good land around him and he's kind of your middle of the road AI. I've seen him strong in some games. Justinian got the short end of the stick after playing well and is really a powerhouse AI leader. In almost all of my offline games too he's a top tier leader.

 

[Fippy]

I think Peter could have won this game without Zara folding.
Justi played good but Peter also not bad + his cities were higher quality.
(this was visible in the preview and 75% picking Justi hmm..questionable)
He also got unlucky that Church had so much culture, rare for him.

Overall i would say the Zara insane stupidy factor was > Alex skill.

 

[ArchGhost]

This latest game showed just how random the AIs really can be.

Zara utterly imploded by chasing then failing to commit to religious path 3 times, all the while ignoring critical development tech. Sullla pointed out he still had warriors defending just before Alex inevitably attacked him, he had Iron but it was online for such a short time it wasn't a factor, the poor settling locations and expansion speed, etc.

Justin on the other hand was playing quite well from a decision making standpoint except for his 3rd city, and got screwed by the map (also not making campaigns for the nearby barb cities). When I saw where he put city 3 I knew he wouldn't be winning, as it was just an open door for Qin to squeeze him.

Peter was looking incredibly good, with his land and lack of contestation due to weakness in Justin and Churchill's general incompetence in deciding to go for religion. I can't believe he didn't win this one, but then again when a civ becomes a runaway by devouring two other guys he's a runaway indeed.

I think I was most impressed with Churchill of all people in this game. He's usually SUCH a lame AI, but here he managed to grab SH for culture and border control, was utilizing his CHA happiness, took away the religion Zara that was chasing and caused him to self-destruct, etc. Even if he had 1v1 Alex only he still would have lost after the roll-over of Zara, but he was still impressive compared to his normal self in this game. I think the last AI Survivor game I saw him in he just turtled in a corner losing wonder races and underexpanding, nice change to see him pull it off a little better this time.

 

[dankok8]

Wildcard game will be interesting. Lots of militaristic AI's in that one.

Toku is one AI who is absolutely hopeless with tech trading on due to his isolationist ways but with tech trading turned off, his attitude doesn't hurt him. And I think due to his science flavor he seeks out economic techs more than the likes of Khan and Monty. I picked Mao to win because I think he will just pull ahead in tech but Toku has a chance here.

 

[Fippy]

I like Genghis start in this one..building lots of units and researching military techs could also benefit him against Raging Barbs.

So i dare saying he could be considered the favorite here, just so much nice land very close (pigs + copper spot, 2x corn, Ivory for happy in his cap, gold north).
If he follows his usual path of early BW he should also have copper fast.

 

[Nogbadthebad]

Raging barbs with this much tundra & ice is going to be messy- Toku and Bis in particular are likely to get wrecked. Im still trying to persuade myself this isn't a runaway win for a GW owning Hammurabi.