Wreathing and Writhing and Arithmetic

[Drachen]

It seems possible that C3C 1.22 has changed the AI research priorities and nerfed the GL Gambit on Sid level standard maps with default settings and random opponents. In the last four or five games I’ve started I’ve found that Alphabet and Writing are discovered much earlier than was usual in 1.15b and that the AI Great Library build starts much earlier as well. The sample size is small but the results are consistent and in these few games the AI starts the GL so early that it becomes much less likely that we can get to literature in time to build it ourselves. I also noted in one game on a large map that the GL was obtainable but that I almost lost the race for it. That’s unusual for large maps and while it could be an aberration it could also be the result of changed AI research priorities. If any of you are observing this trend perhaps you could post your findings so that we can increase the sample size and get a clearer idea if AI research changes have actually occurred.

[DrSpike]

Hmm, I hope you are correct.

[Harovan]

I played around with the English a lot (starting with Alphabet) and set my beeline on Writing and Philosophy, both on a 50 turn pace. Out of about ten ancient ages I tried, I only once missed on getting Writing and Philosophy first of all.

[DrSpike]

On what level? 50 turn philo is sloooowwww.

[Harovan]

Emperor and Demigod. I could've easily raised the pace, but I tried to save some cash and set it to 1 beaker by purpose. What I forgot to mention is, that most games were archipelagos. This may explain it a bit. Still I can't confirm, that the AI values Literature any more. It was mostly one of the last techs researched in the Ancient Age.

[Drachen]

Originally posted by Sir Ralph
Emperor and Demigod. I could've easily raised the pace, but I tried to save some cash and set it to 1 beaker by purpose. What I forgot to mention is, that most games were archipelagos. This may explain it a bit. Still I can't confirm, that the AI values Literature any more. It was mostly one of the last techs researched in the Ancient Age.

It may be that the change is only apparent at Sid level since the AI is able to research so much faster on that difficulty level.

[Drachen]

Originally posted by DrSpike
Hmm, I hope you are correct.

Yeah...I know what you mean. The Great Library is just so unbalancing on Sid the the poor AI doesn't stand a chance.

[DrSpike]

Hehe I just meant the AI undervalues literacy on all levels.

[Harovan]

Originally posted by Drachen
It may be that the change is only apparent at Sid level since the AI is able to research so much faster on that difficulty level.

True, but does it have different research priorities?

[Drachen]

Originally posted by DrSpike
Hehe I just meant the AI undervalues literacy on all levels.

I agree Dr Spike. It’s been the cornerstone of the Glib Gambit.

[Drachen]

Originally posted by Sir Ralph


True, but does it have different research priorities?

That’s exactly the question I’m trying to answer. My sample size is too small to determine if research priorities have changed or I just had a bad run of luck with the RNG. I will note that I went back to 1.15b and promptly built the GL in my first attempt. You might want to try a couple of Sid level runs at the GL on a standard map with default settings. It only takes an hour or so and it doesn’t really matter if you're running 1.15b or 1.22 since either can add to the database. As for me, I’m going to continue my present game and see what my brand new Conquistadors can do.

P.S. If you’ve got the link handy I’d really like to take a look at some of the work you did on city placement for large maps. It’s an area I really need to work on in my games.

[Harovan]

Originally posted by Drachen
P.S. If you’ve got the link handy I’d really like to take a look at some of the work you did on city placement for large maps. It’s an area I really need to work on in my games.

I don't remember to have written about this.

[alva]

Maybe he's refering to Ralphing.

[DrSpike]

What's your problem Drachen?

[Harovan]

Even though this is completely off topic here...

On large maps, I tend to build areas with variable density. Some areas, preferably with a good mix of hills, mountains and grassland and not too corruption ridden, I build loose, so that the cities have the potential to work 18-20 tiles in the end. The same I do in plains areas, which aren't irrigatable before Electricity, because they can't get a good production early on anyway (because they're unable to feed many citizens), but can kick butt later. Those are my production and wonder bases after hospitals. Other areas I build more dense, so that each city has only 10-12 tiles to work. They are my production bases pre hospitals. On some occasions I use Ralphing, which I don't deserve to give it my name (it should rather be called Aesoning), i.e. building camps around an OCP build and dissolving them later.