Chess Forum

janus wrote: ↑Sat Oct 14, 2023 8:21 pm From vondele >>

The right thing here is to setoption name Move Overhead value 1500, i.e. the option is exactly there to inform about expected (worst case?) GUI delays.

Receiving negative (even including zero) btime could lead to all kind of weird stuff. This is really a GUI bug.

Aloha! From where this comment comes? I lost the trace.

Skynet wrote: ↑Sat Oct 14, 2023 9:57 pm

janus wrote: ↑Sat Oct 14, 2023 8:21 pm From vondele >>

The right thing here is to setoption name Move Overhead value 1500, i.e. the option is exactly there to inform about expected (worst case?) GUI delays.

Receiving negative (even including zero) btime could lead to all kind of weird stuff. This is really a GUI bug.

Aloha! From where this comment comes? I lost the trace.

Ha !
You got me - I dont think it was TalkChess but maybe referenced by/to another Forum where vondele seems to be active.
Maybe Discord but I dont think thats where the comment was made.

I just made a copy of his comment and had it on my desktop.
Sorry but I'll keep looking !

EDIT : any wonder I'm a genie-ous ..

Here you go mateys >> https://github.com/official-stockfish/S ... ssues/4000

janus wrote: ↑Sun Oct 15, 2023 6:01 amHa !
You got me - I dont think it was TalkChess but maybe referenced by/to another Forum where vondele seems to be active.
Maybe Discord but I dont think thats where the comment was made.

I just made a copy of his comment and had it on my desktop.
Sorry but I'll keep looking !

janus wrote: ↑Sun Oct 15, 2023 6:01 am Ha !
You got me - I dont think it was TalkChess but maybe referenced by/to another Forum where vondele seems to be active.
Maybe Discord but I dont think thats where the comment was made.

I just made a copy of his comment and had it on my desktop.
Sorry but I'll keep looking !

EDIT : any wonder I'm a genie-ous ..

Here you go mateys >> https://github.com/official-stockfish/S ... ssues/4000

Hah, thanks, i thought that it was from recent issues but this one is from 30/04/2022, supposed to be fixed.

Skynet wrote: ↑Sun Oct 15, 2023 8:51 pm

janus wrote: ↑Sun Oct 15, 2023 6:01 am Ha !
You got me - I dont think it was TalkChess but maybe referenced by/to another Forum where vondele seems to be active.
Maybe Discord but I dont think thats where the comment was made.

I just made a copy of his comment and had it on my desktop.
Sorry but I'll keep looking !

EDIT : any wonder I'm a genie-ous ..

Here you go mateys >> https://github.com/official-stockfish/S ... ssues/4000

Hah, thanks, i thought that it was from recent issues but this one is from 30/04/2022, supposed to be fixed.

Oops I misread the date - need new eyes too ..

In Stockfish, simple eval (evaluate.cpp) has now been removed.
https://www.abrok.eu/stockfish/

Author: cj5716
Date: Fri Feb 9 19:06:25 2024 +0100
Timestamp: 1707501985

Remove simple eval

Now Stockfish solves the two critical positions immediately, for which I had previously criticized Stockfish.

r1n2N1k/2n2K1p/3pp3/5Pp1/b5R1/8/1PPP4/8 w - - 0 1

Analysis by Stockfish 090224-avx2:

1.Ng6+ hxg6 2.Rg2 Be8+ 3.Kf8 Ra4 4.b4 Rxb4 5.c4 Rxc4 6.d4 Rxd4 7.Rh2+ Rh4 8.Rxh4+ gxh4 9.fxg6 Bf7 10.g7+ Kh7 11.g8Q+ Bxg8
= (0.00) Tiefe: 50/23 00:00:07 4454kN, tb=33763
Die Stellung ist ausgeglichen

2r2r1k/qb3p1p/p2p1Np1/np1Pp1P1/2P5/3RP2P/5R2/4Q1K1 w - - 0 1

Analysis by Stockfish 090224-avx2:

1.Rf4 h5 2.Qh4 Rxc4 3.Rxc4 Rc8 4.Rxc8+ Bxc8 5.Kh2 Nc4 6.e4 b4 7.Rf3 Nd2 8.Rd3 Nf1+ 9.Kh1 a5 10.Nxh5 gxh5 11.Qxh5+ Kg7 12.Qh6+ Kg8 13.Qf6 Ba6 14.Qd8+ Kg7 15.Qf6+ Kg8
= (0.00) Tiefe: 50/31 00:00:07 84494kN, tb=1715
Die Stellung ist ausgeglichen

Unfortunately, Stockfish 090224 now has problems with this position:

Qxe5
4q1kr/p6p/1prQPppB/4n3/4P3/2P5/PP2B2P/R5K1 w - - 0 1

Analysis by Stockfish 090224-avx2:

1.Qa3 Rxe6 2.Qxa7 Qe7 3.Qb8+ Qe8 4.Qb7 Qe7 5.Qc8+ Qe8 6.Qxe8+ Rxe8 7.a4 Kf7 8.Be3 Rb8 9.Bd4 Kg7 10.b4 Nc6 11.Bf2 Rhc8 12.h4 Ne5 13. Bd4 Nc6 14.Bf2
= (0.00) Depth: 54/28 00:00:21 204MN, tb=13734

That's why I'm going to change the parameters for the small network again. It will look like that:

Analysis by Stockfish X 090224-avx2:

1.Qxe5 fxe5 2.Rf1 Qe7 3.Bd1 Rc4 4.Bb3 b5 5.a4 a6 6.a5 g5 7.Rf5 Qa7+ 8.Kg2 Qe7 9.Ba2 Qg7 10.Bxg7 Kxg7 11.Bxc4 bxc4 12.e7 Re8 13. Rxg5+ Kf6 14.Rf5+ Ke6 15.Rh5 Rb8 16.e8Q+ Rxe8 17.Rh6+ Kd7 18.Rxa6 Kc7 19.Rh6 Re7 20.Rb6 Rd7 21.Kg3 Rd1 22.Rb4 Kc6 23.Rxc4+ Kb5 24.Rc7 Kxa5 25.Kg4 Rf1 26.Rc5+ Ka4 27.Rxe5 Rb1
+- (4.27) Depth: 33/67 00:00:09 105MN, tb=29546

Yes, I like that better. I will continue to optimize and then implement the optimizations into my own engines. Stockfish X is just an initial test version on the way. My own new engines run with Random Op. MultiPV 2, and have no problems with this and other positions, but Stockfish dev unfortunately has some problems here.

Stockfish development versions, with Minimum Thinking Time and Slow Mover. My compilations:
Stockfish 170224 (avx2, bmi2, sse41, modern, source code) https://pixeldrain.com/u/WpkeK7ak

Stockfish 16.1 released! Stockfish now also supports avx512, vnni256, vnni512, m1-apple-silicon and armv8-dotprod. My compiles for Windows 64 bit as avx2, avx512, bmi2, sse41 and source code. I implemented Minimum Thinking Time and Slow Mover for online players. Download Stockfish 16.1:
https://pixeldrain.com/u/o4ukVzFc

Eduard
https://solistachess.jimdosite.com/

Stockfish dev now with network Dimension 3072:

Author: Viren6
Date: Tue Apr 2 08:49:48 2024 +0200
Timestamp: 1712040588

Update NNUE architecture to SFNNv9 and net nn-ae6a388e4a1a.nnue
https://www.abrok.eu/stockfish/