Clever AI version named in last QA stream

[Fediuld76]

Dear Paradox,

Would it be possible to have as optional that more opportunistic AI mentioned in the last stream?
Or if you not going to use it and is deprecated, could you release it for us to create a mod and improve it?

Players really need better challenging AI than a cheating AI in higher difficulties.
Also since the games (all based on the engine) aren't that graphic demanding, have you considered to those of us having powerful compute graphic cards like Vega 56/64 or RTX2080 to provide some better AI, using out GPUs deep learning capabilities? This is possible also today with Hip via ROCm, which unifies NVIDIA and AMD gpus under a common programming language, and the compiler just generates the correct GPU version code.

I know is a long shot, but I believe a big part of the community want more clever and challenging AI.

 

[Chaingun]

I suspect you misunderstood Johan. Until AI is super good, for sure the best version possible will be enabled at Normal difficulty with exception of special "niceness" behavior against player.

Aside from that, do a forum search on deep learning as those threads pop up fairly often. My post recent post regarding this and other modern AI approaches: https://forum.paradoxplaza.com/foru...rove-paradox-a-i.1142483/page-2#post-25030111

FWIW the earliest version of Imperator AI used deep neural network for unit AI (reinforcement w/ custom temporal credit assignment algorithm) but I replaced it because of performance issues and doubts whether functional concerns could be handled before release. Making a new game on time isn't nearly the same thing as being able to do research on a static game.

 

[Fediuld76]

I suspect you misunderstood Johan. Until AI is super good, for sure the best version possible will be enabled at Normal difficulty with exception of special "niceness" behavior against player.

Aside from that, do a forum search on deep learning as those threads pop up fairly often. My post recent post regarding this and other modern AI approaches: https://forum.paradoxplaza.com/foru...rove-paradox-a-i.1142483/page-2#post-25030111

FWIW the earliest version of Imperator AI used deep neural network for unit AI (reinforcement w/ custom temporal credit assignment algorithm) but I replaced it because of performance issues and doubts whether functional concerns could be handled before release. Making a new game on time isn't nearly the same thing as being able to do research on a static game.

Thank you