sunfish

Indeed Sunfish often takes a more lengthy approach than say Micro-Max, because Sunfish is also meant to teach chess programming.

Any program which has greater than this level of precision, and equally precise searching, should be able to beat sunfish.

I really like this work but the performance of the network against Sunfish is not particularly informative.

Saying it can beat Sunfish (a python program with grade of?maybe ELO 1100?

You are right of course:) Sunfish is nowhere the level of 'real' engines.

That deep learning could not break even against Sunfish seems rather unimpressive.

When sunfish (with its just 111 python lines of chess logic) 'evaluates' a position, it uses the perhaps simplest known effective method: A piece-square table.

It only lacks the whoosh-rush-boom I recall from turning my Sunfish sharply across the wind.

I’m currently using Negamax with alpha-beta pruning, whereas Sunfish uses MTD-f MTD-F is not better, just a different way to accomplish more-or-less the same thing.

As the author of sunfish, I'd like to point out something about learning in chess, which is a topic that interests me a great deal.

But, I had fun on a little Sunfish as a teenager.

But apparently sunfish-kernel only contains binaries.

Honestly getting a Sunfish is probably cheaper than the a VR headset if you want to "train sailing"

I tried to make to make a sunfish and it was like "not a fish".

Note that Sunfish isn't a particular "good" AI, it would be more interesting comparing it to a "proper" chess AI.

Like Sunfish it also doesn't do underpromotion: "which I considered a dull input problem".

Sunfish move generation is tested pretty thoroughly, generating tens of millions of positions and comparing with other engines.

Sunfish is impressive because it's written in Python and in a tiny number of lines while still being readable.

From my (brief) understanding of Sunfish, the evaluation function is essentially equivalent to a hardcoded 1 layer network in Deep Pink.