https://github.com/amchess/ShashChess/releases/tag/29.1
In particular,
https://github.com/amchess/ShashChess/b ... 0-2023.pgn
ShashChess
Moderators: hgm, Rebel, chrisw
-
peter
- Posts: 3252
- Joined: Sat Feb 16, 2008 7:38 am
- Full name: Peter Martan
Re: ShashChess
Thanks for the many updates.amchess wrote: ↑Mon Mar 06, 2023 10:23 pm https://github.com/amchess/ShashChess/releases/tag/29.1
In particular,
https://github.com/amchess/ShashChess/b ... 0-2023.pgn
29.1 is the first one, that doesn't work correctly with MEA (Ferdinand Mosca). Given milliseconds for each position aren't interpreted correctly, everything <500 is ignored. No error message, run is completed and rated, just the TC is wrong.
Tried already to set Move Overhead and Slow Mover to minimum, doesn't help neither.
Best regards
Peter.
-
amchess
- Posts: 349
- Joined: Tue Dec 05, 2017 2:42 pm
Re: ShashChess
Hi.peter wrote: ↑Wed Mar 08, 2023 10:15 amThanks for the many updates.amchess wrote: ↑Mon Mar 06, 2023 10:23 pm https://github.com/amchess/ShashChess/releases/tag/29.1
In particular,
https://github.com/amchess/ShashChess/b ... 0-2023.pgn
29.1 is the first one, that doesn't work correctly with MEA (Ferdinand Mosca). Given milliseconds for each position aren't interpreted correctly, everything <500 is ignored. No error message, run is completed and rated, just the TC is wrong.
Tried already to set Move Overhead and Slow Mover to minimum, doesn't help neither.
Best regards
ShashChess 29 uses the win probability to classify positions based on Shashin theory and uses for this the formula in Stockfish wdl model. This formula is rather computationally onerous.
Therefore, in ShashChess 29.1, we introduced a hashmap (to be precise, unordered_map), initialized at engine startup in a similar way to syzygy, so that we have immediate access to the win probability, given a score and depth of analysis.
I am trying to figure out how this might impact this tool that I am not familiar with.
-
peter
- Posts: 3252
- Joined: Sat Feb 16, 2008 7:38 am
- Full name: Peter Martan
Re: ShashChess
Thanks for explaining and looking into it, Andrea, just to notice again, with version 29 the TC was interpreted correctly still, 29.1 is the first one aberrant as for the tool and its fixed time per move syntax, regards
Peter.
-
peter
- Posts: 3252
- Joined: Sat Feb 16, 2008 7:38 am
- Full name: Peter Martan
Re: ShashChess
BTW, as for Banksia 0.58 positional test tool it's about the same, 200 msec/pos. given takes about 8.5 min for 512, that's well 5 times too much:
Code: Select all
File name : 512.epd
Total test items : 512
Test for : best moves
Total engines : 1
Timer : movetime: 0.2
Expand ply : 0
Elapsed : 08:35
Laps : 1
Total tests : 512
Total corrects : 420 (82%)
Ave correct elapse : 17 ms
Status : completed
Correct/Total:
ShashChess 29.1: 420/512Peter.
-
peter
- Posts: 3252
- Joined: Sat Feb 16, 2008 7:38 am
- Full name: Peter Martan
Re: ShashChess
>750 msec TC- interpretation starts coming nearer to the one of other engines, but yet even with 1"/move, there's still about 150% time consumption by ShashChess 29.1 than by other engines in GUIs that allow so (and in console like with MEA).peter wrote: ↑Wed Mar 08, 2023 7:44 pm BTW, as for Banksia 0.58 positional test tool it's about the same, 200 msec/pos. given takes about 8.5 min for 512, that's well 5 times too much:
Code: Select all
File name : 512.epd Total test items : 512 Test for : best moves Total engines : 1 Timer : movetime: 0.2 Expand ply : 0 Elapsed : 08:35 Laps : 1 Total tests : 512 Total corrects : 420 (82%) Ave correct elapse : 17 ms Status : completed Correct/Total: ShashChess 29.1: 420/512
Shredder13- GUI with 1"/move ("exact time"- checked), like Dragon in this one example also other one SF- branches and e.g. Ethereal 14.00 take rather exactly 512", Shashchess 29.1 takes 850:
Code: Select all
Dragon 3.2:
Bisher gelöst: 405 von 512 ; 2:32m
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-------------------------------------------------------------------------------------
0 | - 0 - 0 0 - 0 - 0 0 0 0 0 0 0 0 0 - 0 0
20 | 0 - 0 0 0 - 0 0 0 - 0 0 - - 0 0 0 0 0 0
40 | - 0 0 - - 0 0 0 - 0 0 - 0 0 0 - 0 - - 0
60 | - 0 - 0 0 0 - 0 0 - 0 0 0 0 - 0 0 0 0 0
80 | 0 0 0 - 0 - - 0 0 - 0 0 0 0 0 0 0 0 0 0
100 | 0 0 0 0 0 - - - 0 - 0 - 0 0 - 0 0 0 0 0
120 | 0 - - - - 0 0 0 - 0 0 0 0 0 0 - - 0 0 -
140 | 0 - 0 - 0 0 0 0 - - - - 0 - 0 0 0 0 - 0
160 | 0 - 0 0 0 - 0 0 0 - 0 - 0 0 - 0 0 0 0 0
180 | 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 - 0 0 - -
200 | - 0 0 0 0 0 0 0 0 - 0 0 0 0 0 - 0 0 - -
220 | 0 0 0 0 0 0 0 0 - - - 0 0 0 - 0 - 0 0 0
240 | 0 - - 0 0 0 0 - - - - - - - - 0 0 0 0 -
260 | - 0 0 - - - 0 0 - - 0 0 0 0 0 0 0 0 0 0
280 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
300 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
320 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
340 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
360 | 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 - 0 0
380 | 0 0 0 0 0 0 0 0 0 0 0 - 0 0 - 0 0 0 0 -
400 | 0 0 0 0 0 - 0 - 0 - 0 0 0 0 0 - 0 0 0 0
420 | 0 0 0 0 0 0 0 0 0 0 - - 0 0 0 - 0 0 0 0
440 | 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 - 0 0 0 0
460 | 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0
480 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0
500 | 0 - - - - 0 0 0 0 - 0 -
1 sec -> 405/512
K/s: 1.973.838
TotTime: 8:32m SolTime: 2:32m
ShashChess 29.1:
Bisher gelöst: 430 von 512 ; 2:59m
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-------------------------------------------------------------------------------------
0 | 0 0 0 0 0 0 0 - 0 0 0 0 0 - 0 - - 0 0 0
20 | 0 - - 0 0 - 0 0 0 0 - 0 - - 0 0 0 0 0 0
40 | 0 0 0 - 0 0 0 0 - 0 0 0 1 0 0 0 0 - - -
60 | - 0 - 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 - 0
80 | 0 - 0 0 0 - 0 - 0 - 0 0 0 0 0 0 0 0 0 0
100 | - 0 0 0 0 0 0 - - - 0 - 0 0 0 0 0 0 - 0
120 | 0 - - - - 0 - 0 - 0 0 0 0 0 0 - 0 0 0 -
140 | 0 - - 0 0 0 0 0 0 - 0 - 0 0 0 0 - 0 0 0
160 | 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 0 0
180 | 0 0 0 0 0 0 0 - 0 0 - - 0 0 0 - 0 0 0 0
200 | 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 - 0 0 - -
220 | 0 0 0 0 0 - 0 0 - 0 - 0 0 0 0 0 - 0 0 0
240 | 0 0 - 0 0 0 0 0 - 0 - - 0 0 0 0 0 0 - 0
260 | - - 0 0 - - 0 0 - - - 0 - 0 0 0 0 0 0 0
280 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
300 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
320 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
340 | 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0
360 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
380 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
400 | 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 - 0 0 0 0
420 | 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 - 0 0 0 0
440 | 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 - 0 0 - 0
460 | 0 0 0 0 0 0 0 - - 0 0 0 0 - 0 0 0 0 0 0
480 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0
500 | 0 0 - 0 - 0 0 0 0 0 0 0
1 sec -> 429/512
K/s: 856.274
TotTime: 14:10m SolTime: 2:59mPeter.
-
Jouni
- Posts: 3425
- Joined: Wed Mar 08, 2006 8:15 pm
Re: ShashChess
In my test 29.1 + Tal is very good in testsuites! But Crystal 5 is even better in all suites.
Jouni
-
peter
- Posts: 3252
- Joined: Sat Feb 16, 2008 7:38 am
- Full name: Peter Martan
Re: ShashChess
5" is the shortest TC 29.1 at least in Fritz GUI interpretes correctly,
at 1" even in Fritz (18) 578" instead of 512 max given are used yet still:
EloStatTS relativizes the time to solution additionally anyhow.. For 5"/postion and 30 threads of the 16x3.5 GHz- CPU I like e.g. these 256:Ergebnis: 443 aus 512 = 86.5%. Durchschnittszeit = 0.72s / 9.38
https://www.dropbox.com/s/s3jn6pxp16diqvt/256.epd?dl=0
Code: Select all
Program Elo +/- Matches Score Av.Op. S.Pos. MST1 MST2 RIndex
1 Crystal5KWK : 3565 7 1603 60.8 % 3488 168/256 1.8s 2.9s 0.73
2 BlueMarlin15.6 : 3562 7 1534 60.4 % 3489 160/256 1.7s 3.0s 0.75
3 Eman8.7064-bitBMI2 : 3546 7 1452 57.7 % 3492 146/256 1.8s 3.1s 0.69
4 ShashChess29.1 : 3530 8 1454 55.2 % 3494 143/256 2.1s 3.4s 0.53
5 ShashChess28 : 3529 8 1442 55.0 % 3494 140/256 2.1s 3.4s 0.56
6 Dragon3.2byKomodoChess : 3491 9 1337 48.4 % 3502 102/256 1.8s 3.7s 0.47
7 Berserk11.1 : 3460 9 1211 42.8 % 3511 85/256 2.1s 4.0s 0.38
8 Ethereal14.00(NNUE) : 3424 9 1159 37.2 % 3515 67/256 2.3s 4.3s 0.26
9 Rebel-16.2 : 3416 9 1148 36.0 % 3516 60/256 2.1s 4.3s 0.27
10 Koivisto9.2 : 3406 9 1128 34.5 % 3517 52/256 1.7s 4.3s 0.28
MST1 : Mean solution time (solved positions only)
MST2 : Mean solution time (solved and unsolved positions)
RIndex: Score according to solution time ranking for each position
Peter.
-
peter
- Posts: 3252
- Joined: Sat Feb 16, 2008 7:38 am
- Full name: Peter Martan
Re: ShashChess
As for MEA, TC given 100msec/pos. for ShashChess 29.1 results in about same time consumption (with Move Overhead and Slow Mover set to min.) like with other engines default setting and 750msec/move (like e.g with ShCh 29 too), all engines execpt LC0 (Nividia 3070ti GPU) ran with single thread of 3.5GHz CPU over these 2345 positionspeter wrote: ↑Thu Mar 09, 2023 10:28 am >750 msec TC- interpretation starts coming nearer to the one of other engines, but yet even with 1"/move, there's still about 150% time consumption by ShashChess 29.1 than by other engines in GUIs that allow so (and in console like with MEA).
Shredder13- GUI with 1"/move ("exact time"- checked), like Dragon in this one example also other one SF- branches and e.g. Ethereal 14.00 take rather exactly 512", Shashchess 29.1 takes 850:
Code: Select all
Dragon 3.2: Bisher gelöst: 405 von 512 ; 2:32m 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ------------------------------------------------------------------------------------- 0 | - 0 - 0 0 - 0 - 0 0 0 0 0 0 0 0 0 - 0 0 20 | 0 - 0 0 0 - 0 0 0 - 0 0 - - 0 0 0 0 0 0 40 | - 0 0 - - 0 0 0 - 0 0 - 0 0 0 - 0 - - 0 60 | - 0 - 0 0 0 - 0 0 - 0 0 0 0 - 0 0 0 0 0 80 | 0 0 0 - 0 - - 0 0 - 0 0 0 0 0 0 0 0 0 0 100 | 0 0 0 0 0 - - - 0 - 0 - 0 0 - 0 0 0 0 0 120 | 0 - - - - 0 0 0 - 0 0 0 0 0 0 - - 0 0 - 140 | 0 - 0 - 0 0 0 0 - - - - 0 - 0 0 0 0 - 0 160 | 0 - 0 0 0 - 0 0 0 - 0 - 0 0 - 0 0 0 0 0 180 | 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 - 0 0 - - 200 | - 0 0 0 0 0 0 0 0 - 0 0 0 0 0 - 0 0 - - 220 | 0 0 0 0 0 0 0 0 - - - 0 0 0 - 0 - 0 0 0 240 | 0 - - 0 0 0 0 - - - - - - - - 0 0 0 0 - 260 | - 0 0 - - - 0 0 - - 0 0 0 0 0 0 0 0 0 0 280 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 300 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 320 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 340 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 360 | 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 - 0 0 380 | 0 0 0 0 0 0 0 0 0 0 0 - 0 0 - 0 0 0 0 - 400 | 0 0 0 0 0 - 0 - 0 - 0 0 0 0 0 - 0 0 0 0 420 | 0 0 0 0 0 0 0 0 0 0 - - 0 0 0 - 0 0 0 0 440 | 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 - 0 0 0 0 460 | 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 480 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 500 | 0 - - - - 0 0 0 0 - 0 - 1 sec -> 405/512 K/s: 1.973.838 TotTime: 8:32m SolTime: 2:32m ShashChess 29.1: Bisher gelöst: 430 von 512 ; 2:59m 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ------------------------------------------------------------------------------------- 0 | 0 0 0 0 0 0 0 - 0 0 0 0 0 - 0 - - 0 0 0 20 | 0 - - 0 0 - 0 0 0 0 - 0 - - 0 0 0 0 0 0 40 | 0 0 0 - 0 0 0 0 - 0 0 0 1 0 0 0 0 - - - 60 | - 0 - 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 - 0 80 | 0 - 0 0 0 - 0 - 0 - 0 0 0 0 0 0 0 0 0 0 100 | - 0 0 0 0 0 0 - - - 0 - 0 0 0 0 0 0 - 0 120 | 0 - - - - 0 - 0 - 0 0 0 0 0 0 - 0 0 0 - 140 | 0 - - 0 0 0 0 0 0 - 0 - 0 0 0 0 - 0 0 0 160 | 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 180 | 0 0 0 0 0 0 0 - 0 0 - - 0 0 0 - 0 0 0 0 200 | 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 - 0 0 - - 220 | 0 0 0 0 0 - 0 0 - 0 - 0 0 0 0 0 - 0 0 0 240 | 0 0 - 0 0 0 0 0 - 0 - - 0 0 0 0 0 0 - 0 260 | - - 0 0 - - 0 0 - - - 0 - 0 0 0 0 0 0 0 280 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 300 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 320 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 340 | 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 360 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 380 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 400 | 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 - 0 0 0 0 420 | 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 - 0 0 0 0 440 | 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 - 0 0 - 0 460 | 0 0 0 0 0 0 0 - - 0 0 0 0 - 0 0 0 0 0 0 480 | 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 500 | 0 0 - 0 - 0 0 0 0 0 0 0 1 sec -> 429/512 K/s: 856.274 TotTime: 14:10m SolTime: 2:59m
https://www.dropbox.com/s/hqeqqn31s5ynqr2/2345.epd?dl=0
Code: Select all
Kovisto 9.2:
Time allocation : GOOD!!
at <= et + mt and at >= et - mt
ExpectedTime : 1758.8s
ActualTime : 1766.7s
MarginTime/pos : 0.2s
MarginTime : 438.5s
ShashChess 29:
Time allocation : GOOD!!
at <= et + mt and at >= et - mt
ExpectedTime : 1758.8s
ActualTime : 1765.3s
MarginTime/pos : 0.2s
MarginTime : 438.5s
ShashChess 29.1:
Time allocation : BAD!! spending more time
at > et + mt
ExpectedTime : 234.5s
ActualTime : 1755.4s
MarginTime/pos : 0.1s
MarginTime : 117.2s
Code: Select all
EPD : epd\2345.epd
Time : 750ms
Max Total Time Hash
Engine Score Found Pos ELO Score Rate ms Mb Cpu
1 SF230305 29611 1908 2345 4108 32444 91.3% 750 8 1
2 BlueMarlin15.6 29467 1893 2345 4086 32444 90.8% 750 8 1
3 Eman8.70NN 29392 1897 2345 4077 32444 90.6% 750 8 1
4 ShashChess29 29383 1893 2345 4077 32444 90.6% 750 8 1
5 lc0230217-325M 29246 1896 2345 4054 32444 90.1% 800 8 2
6 Dragon3.2 28739 1842 2345 3987 32444 88.6% 750 8 1
7 Rebel16.2 28067 1792 2345 3892 32444 86.5% 750 8 1
8 Ethereal14 27926 1785 2345 3874 32444 86.1% 750 8 1
9 Koivisto9.2fix 27761 1777 2345 3852 32444 85.6% 750 8 1
10 ShashChess29.1 27666 1752 2345 3838 32444 85.3% 750 8 1
Created with MEA
by
Ferdinand
Mosca
Peter.
-
amchess
- Posts: 349
- Joined: Tue Dec 05, 2017 2:42 pm
Re: ShashChess
ShashChess 30
https://github.com/amchess/ShashChess/releases/tag/30
Tests result
Match:
https://github.com/amchess/ShashChess/wiki/Matches
Hard positions:
https://github.com/amchess/ShashChess/w ... te-results
https://github.com/amchess/ShashChess/releases/tag/30
Tests result
Match:
https://github.com/amchess/ShashChess/wiki/Matches
Hard positions:
https://github.com/amchess/ShashChess/w ... te-results