TalkChess.com

Fairy-Stockfish

That is NNUE, not the LeelaZero NN + MCTS I was inquiring about. Still, good to know that classical search techniques + NNUE work well on these games. (First time I'd heard about the Makpong variant.)

Are there any Leela-based engines for any of the southeast-asian chess variants? Since these are all more strategic, less tactical, I would think NN would be a good fit. I'd be curious if such engines would discover the same opening strategies used in Makruk, or play the endgames better than in ches...

I suggest asking Larry Kaufmann for a review of your work. He is knowledgable in both computer chess and shogi.

Thank you so much for those references! I guess there were quite a few variations on that theme. I'm delighted that we can still enjoy the creative brilliance of the finest minds of a thousand years ago, even though the game has changed. (See also Al-Suli's Diamond, a corresponding-square Shatranj e...

Hello Álvaro: https://www.youtube.com/watch?v=iCG5lVDZWmc There are gaps in the video. I found a longer video: https://www.youtube.com/watch?v=HhrvwHrceRg They also featured three three-movers: 2) Very old problem: [Result "1-0"] [Setup "1"] [FEN "5K2/7r/4k1rn/4p3/4P3/5N2/8/3R1R2 w - - 0 1"] 1.Ng5+...

I am also extremely grateful for all of Larry Kaufman's contributions to computer chess. His treatise on Handicap Shogi is also one of the only sources of deep shogi knowledge available to western students of shogi. In fact, wasn't Larry the one who urged the NNUE cross-pollination between computer ...

Another possibility is to use stricter endgame counts which subsume the 50-move rule when training. As an example, consider the Thai chess variant, Makruk. It has a host of different counts, which depend on the amount of material left.* Does anyone know if there is a project to apply the LC0 framewo...

Also, there's this historical gem: https://www.chessgames.com/perl/chessgame?gid=1044024 Fischer-Pupols is a rough gem, indeed. A teenage future master beats up the 12-year-old future world champ! There are many unpublished picturesque mates and faster attacks on the sidelines, well worth analyzing.

I'm even more impressed by this design for self-moving pieces: https://regiumchess.com/ Honestly, once I saw how it was implemented (a 3x3 grid of small, strong electromagnets under each square and captured piece area), I was amazed that nobody had tried it before. The mechanism is practically solid...

Spark 1.0 gets it in a few seconds at depth 16. It is a better than average mate finder.