Paradox, can you have a dev diary on the AI?

[King Travis]

It seems like with this genre, AI is always an afterthought. Will the AI be different, or is the focus on multi player mean that its probably just going to be the standard AI stuff. Not saying it will be bad, but it would be nice to have a focus on it. Galciv used to pride itself on its AI strength, and it was pretty amazing. (not sure if its that way in the newest one).

 

[Krajzen]

Yes Paradox we need AI races!

...

No but seriously, I agree wholeheartedly. AI in strategy games is VERY important.

 

[valergain]

While I agree that AI is important not entirely sure what an entire DD about would serve. Just letting them know is good enough in my opinion.

Yes Paradox we need AI races!

This will be like the first or second DLC I swear...

 

[DKinator]

It seems like with this genre, AI is always an afterthought. Will the AI be different, or is the focus on multi player mean that its probably just going to be the standard AI stuff. Not saying it will be bad, but it would be nice to have a focus on it. Galciv used to pride itself on its AI strength, and it was pretty amazing. (not sure if its that way in the newest one).

I dig what you're saying, but paradox has proven themselves time and time again as having some of - if not the best - AI in strategy games, so I wouldn't be too concerned on that front. It's really quite a marvel, honestly. If it's anywhere remotely as good as EUIV's AI as of now, it'll be more than sufficient. I'd wager, with all they've learned from CK2 and EUIV, the AI's baseline will be incredible. I'm just hoping Wiz gets very little sleep in the coming months and works on all their projects. Poor guy.

I haven't played Galciv3, but I remember one of the goals being "learning AI" or some such where the game monitors how people play the game and apply it to the AI's logic. No idea how that'd work, if it worked, or even how well Galciv3 is doing. But you've reminded me that enough time has passed since its initial release and a lot of the early release bugs from a kickstarter game have probably been ironed out by now (at least, I think it was a kickstarter...).

 

[EntropyAvatar]

If they were to do a dev diary on AI, it could focus on areas where the AI is different from previous Paradox games. Like how it makes decisions on anomaly event chains, how it decides to design ships, they it plans fleet movement with multiple FTL movement types, how it reasons about possible enemy FTL movements. There are real-time battles where fleet formation and ship positioning, target priority and so on could be important. How does it decide that stuff?

 

[King Travis]

One thing i would like to see is that the AI steals ship designs from the player base. Humans come up with some awesome, crazy designs.

 

[Aries666]

One thing i would like to see is that the AI steals ship designs from the player base. Humans come up with some awesome, crazy designs.

But would it be able to tell the good designs from the bad ones? I can imagine sabotaging the AI by purposefully creating crappy ship designs and letting it steal from me

 

[King Travis]

But would it be able to tell the good designs from the bad ones? I can imagine sabotaging the AI by purposefully creating crappy ship designs and letting it steal from me

This is true. Maybe it tracks the users success with said design. Not sure.... maybe itd be fun to have some crap ships too.

 

[FelixG]

While I agree that AI is important not entirely sure what an entire DD about would serve. Just letting them know is good enough in my opinion.

I think a whole DD would be good, they could lay out what they are doing different, how the AI can adapt for long term plans, how a few traits will contribute to AI behavior, etc.