TalkChess.com

jonkr wrote: ↑Mon Jan 11, 2021 9:59 am
Udpate On FRC play :

I ran some more testing and in FRC 2.5 was weaker than 2.4 even with AVX. (Although standard it was still 40+ elo stronger, really shows how specific neural nets can be.)

jonkr wrote: ↑Mon Jan 11, 2021 5:05 am SlowChess Blitz 2.5 is now available at
https://www.3dkingdoms.com/chess/slow.htm

Gabor Szots wrote: ↑Wed Jan 13, 2021 2:05 pm Where are the updated endgame nets? The last one I have was from v2.4 and dates 30th October.

jonkr wrote: ↑Mon Jan 11, 2021 5:05 am SlowChess Blitz 2.5 is now available at
https://www.3dkingdoms.com/chess/slow.htm

You should use slow64-avx if your computer supports AVX2.
The alternative is slow64-noPop. It is plain SSE2. Hardware popcount doesn't really matter since not used in neural net eval, so I didn't compile any more versions.

Slow 2.5 now uses neural nets for every stage of the game. I decided it was time to drop Classic from the name. The endgame nets are the same as 2.4, except with more training and the addition of a Q v R(s) network. The main addition is a network to replace the general evaluations. It's the same structure as the endgame nets, but with a more neurons on first later ( 320x32x32x1 ) Inputs are the same too, just piece/sq, with horizontal symmetry based on white king.

Playing style is similar to 2.4 but more different than any of the hard-coded evals were to each other. Replacing a hard-coded eval with a neural net is a big change, even one based partially on the same training positions and partially on search with same eval. Older hce versions with just certain terms added and factors tweaked showed an extra high draw % playing against previous version, and this one doesn't show that. I think eval definitely main factor in how an engine plays.

On the plus side

• I estimate it's about 40 elo stronger in standard chess than 2.4
• Can beat Stockfish 8 in 2-moves book matches by a few elo on my computer.
• The play based on win percentages does feel more human-like to me.

On the minus side

• requires AVX2 for strongest play
• struggles in Chess960 variant (can't recommend it over 2.4)
• can take longer to close out games. Probably from valuing sure wins so high. (Some day I may see if I can train it try to close out as aggressively as possible.)

I think it's of some interest to have a neural net program that is not standard NNUE and not based on it in any way. (Although it's clearly much weaker than any stockfish NNUEs, and I'm not sure how much is based on my lack of computer resources for training, and how much is from not being the best possible architecture or training process.)

At this point the idea of working more to make it stronger sounds too tedious so I'm releasing it, but if the past is a guide I expect eventually I will start experimenting again then make enough progress to make another version.

Carlos777 wrote: ↑Sat Jan 16, 2021 1:30 pm Thanks for this new version.

The nopopcnt version does not work on my old PC (AMD Phenom II x4), 2.4 version works fine.
When 2.5 starts thinking (out of the book), it crashes.

Carlos777 wrote: ↑Sat Jan 16, 2021 1:30 pm Thanks for this new version.

The nopopcnt version does not work on my old PC (AMD Phenom II x4), 2.4 version works fine.
When 2.5 starts thinking (out of the book), it crashes.

Tibono wrote: ↑Sat Jan 16, 2021 5:01 pm +1 (Intel Pentium Penryn, Instructions sets: MMX, SSE, SSE2, SSE3, SSSE3, EM64T)
Cheers,
Tibono