Discussion of anything and everything relating to chess playing software and machines.
Moderator: Ras
peter
Posts: 3410Joined: Sat Feb 16, 2008 7:38 amFull name: Peter Martan
Post by peter Thu Oct 26, 2023 10:19 am
Dann Corbit wrote: ↑ Thu Oct 26, 2023 9:49 am
Thanks to all for these efforts. Believe it or not , I did analyze all of these with multi-pv at long time control with several engines.
All the positions are fine, Dann, I just wanted to point out that some of them to me seem better and some less good for usage in automatically evaluated suites with short to moderate TC.
Aamof the 128 out of the 138 just fit better then the 138 to 128 other ones, I use in suite now and then too,
https://www.dropbox.com/scl/fi/rpjl89w0 ... nfwfl&dl=0
so they'd well add up to 256 all in all with same TC of about 15"/pos. SMP
Peter.
peter
Posts: 3410Joined: Sat Feb 16, 2008 7:38 amFull name: Peter Martan
Post by peter Sun Oct 29, 2023 10:58 am
E.g. with 8 threads of 16x3.5GHz- CPU (to let some instances run simultaneously), Lc0 with 2 and the 3070ti- GPU:
Code: Select all
Lc0 v0.31.0-dag+git.e429eeb0
Bisher gelöst: 139 von 256 ; 21:01m
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-------------------------------------------------------------------------------------
0 | - - - - 1 - 0 0 0 - 0 - - 5 2 - - 0 0 -
20 | 3 - - - 13 0 - - 1 0 0 - - 0 - 2 0 - - 0
40 | - 0 0 2 - 8 - - - 0 - - - - - 0 0 10 0 0
60 | 0 - - - - 0 - - 0 5 - - 2 - 0 - - 2 0 -
80 | - 5 3 0 - 1 3 - - - - 0 3 1 0 0 11 - - -
100 | - - - - 2 - - - 3 0 - - 0 0 - - - 0 - -
120 | 0 0 8 0 0 0 - 1 - 0 - - 0 1 - 2 - 0 0 0
140 | 9 3 - 0 0 0 - 0 - - 0 - - 0 - 0 5 0 0 -
160 | - - - 2 - - 0 0 - - 7 0 - 0 0 0 0 1 0 0
180 | - 2 0 0 3 - 1 5 0 - 0 - 0 0 - 0 - - - 10
200 | - - 0 6 0 0 - - - 1 0 0 1 0 - 5 - - 0 -
220 | - 0 - - - 7 - 0 0 - 0 - 0 - 0 - 0 0 5 -
240 | 2 0 - 0 6 0 3 0 10 0 3 4 - 5 1 0
K/s: 28.415
TotTime: 24:45m SolTime: 21:01m
Stockfish dev-20231008-7a4de9610
Bisher gelöst: 108 von 256 ; 47:03m
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-------------------------------------------------------------------------------------
0 | 0 - - 1 3 1 0 11 1 - 0 - 3 0 0 - - 1 - 0
20 | 3 - - - - 0 - - 0 0 9 - 12 14 - - - - - 2
40 | - - 7 - - - - - - - - - - - - - - 0 - 0
60 | 10 0 - - 0 - - - 5 0 - - - 0 12 12 - - - 1
80 | 15 - 0 - - 10 7 0 - 0 - 0 5 - - 0 - - - 1
100 | - - - 4 - - - 2 2 0 - - 0 0 - 15 - - - -
120 | - - - - 11 - - - - - - 11 11 - 7 6 - 1 0 12
140 | - - 7 4 - - 9 - - - - 3 - - - 5 4 - - -
160 | - - - - 5 - - 15 - - - 6 0 - - - - 15 4 -
180 | 0 4 - 3 13 15 - - - - 2 - 1 - - 9 - 12 8 -
200 | 15 - 0 - - - 0 15 8 - - 3 3 13 - - - - - -
220 | - - - 3 11 0 - 0 - - 7 2 9 - - 2 8 13 2 0
240 | - - 1 - - 1 2 - 1 - 12 3 - 12 - 7
K/s: 6.539.807
TotTime: 52:33m SolTime: 47:03m
Berserk 12
Bisher gelöst: 105 von 256 ; 45:09m
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-------------------------------------------------------------------------------------
0 | - - - - 6 7 0 - 9 - 0 - - - 0 - - - - -
20 | - - - - - 0 - - 13 - 1 - - 0 - - - - 1 0
40 | - - - - - - - - - - - - 8 - - - - - - 0
60 | - - - - - - - - - 1 7 - 4 12 - - 2 2 - 6
80 | 1 - 0 - 0 6 3 7 - 1 - - - - - 8 0 - - -
100 | - - - - 3 - - - 7 0 - - 0 0 - - - - - -
120 | - - - - 9 - 10 - - 4 1 - 8 - - 0 - 2 0 0
140 | - - 2 1 0 - - 0 - 6 1 - - 7 3 14 3 - - -
160 | - 13 - - - 14 - 2 4 - 3 1 14 6 - - - 2 2 15
180 | 0 - - - 1 6 - 15 1 - 1 1 7 - - 2 - 0 - -
200 | 15 - 0 - 15 - 0 - - - 0 6 - - - - - - - -
220 | 0 3 - - - 0 - - 1 - 4 1 7 - - 8 0 2 9 0
240 | 0 - 3 - 0 1 - 0 0 0 4 - 14 0 3 0
K/s: 11.660.696
TotTime: 49:51m SolTime: 45:09m
Dragon 3.3 by Komodo Chess 64-bit
Bisher gelöst: 92 von 256 ; 45:43m
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-------------------------------------------------------------------------------------
0 | - - - - - - 0 - - - - 0 - - - - - 0 - -
20 | - - 1 - - - - 7 6 0 1 - - - - - - - - 7
40 | - - - - - - 10 - - - - - - - - - - 0 - 0
60 | 8 2 - - - - - - 0 1 - 0 - 0 - - - - - 0
80 | - 0 0 - - - 9 - - - - 1 - - - 0 - - - -
100 | - - - - - - - - 7 0 4 - 0 0 - 8 - 4 - -
120 | 3 - - - - - - - - 1 - - - 0 5 3 - - 0 -
140 | - - 0 0 0 3 0 5 - - 6 - 9 9 - 1 13 - - -
160 | - - - - 2 - 0 10 3 - - 0 - 1 - - - 1 0 0
180 | 2 1 - 0 0 - - - - - 0 4 1 - - - - 1 0 -
200 | 13 - 0 - - - - - - 0 0 13 3 - 7 - - 1 0 -
220 | - 0 1 - - 0 - - - - 0 - - - - 9 0 - 7 -
240 | 12 - 7 - - 7 0 - 1 - 3 0 0 1 - 0
K/s: 14.179.260
TotTime: 50:57m SolTime: 45:43m
Again, I'm agreeably surprised how well Lc0 (net 2790M) scores. Of course 8 threads of this one CPU is somewhat too little "Leela- ratio" for A-B maybe, but those measurements stand each one for their own only anyhow. Seeing how much less time Lc0 needs all in all compared to the other engines (about the same as with MEA BTW), EloStatTS (Frank Schubert) would bring much further criteria into matching, but I couldn't let more than one instance run at same time with Fritz- GUI, so that's too hardware- time consuming to me for such results only, regards
Peter.
