This is also a very good list. Most of the names ring true for their aggressive playing style. I hope you have tested engine called Disaster Area too? It does play exceedingly aggressive only when it is fed with those parameters sharted above. Otherwise, I am not very sure how it does with the standard values.adnoh wrote: ↑Thu Oct 06, 2022 7:57 am I am using Stefan's EAS 4.1 tool on my own hobby chess rating list and OpenTal 1.0 is by far the most aggressive engine I have tested based on that rating. Below is my top 20 using EAS and note another one to consider is Stockfish 5 Fury which is also based on modified settings via UCI settings.
CSTal (#11) is Chris's work.
Each engine played several thousand games.
Code: Select all
***************************************************************************** *** Engine Aggressiveness Tool V4.1 Score points Ratinglist ***************************************************************************** Score is: (percent*100) of the percent-values of the sacs (1-5+ pawnunits) calculated out of the won games by the engine, only. So, a weak engine (with a small number of won games) can get a high EAS-scoring, too, when the percent of sac-games in the won games is high (and the number of short wins). Higher pawnunits-sacs give bonus-points: 1 pawnsac = 2x points *** 2 pawnsac = 10x points *** 3 pawnsac = 20x points 4 pawnsac = 40x points *** 5+ pawnsac = 80x points *** 5+ Queensac = 150x points Additionally, very short won games (percent*100) give bonus-points. If the short games ended before endgame (a check for low material is done), then these points are quadrupled (4x): 60 moves= 1x points *** 55 moves= 2x points *** 50 moves= 5x points 45 moves= 10x points *** 40 moves= 30x points. Since V4.1, the tool also searches for "bad draws" and adds points for the good draws ((percent of good draws (out of all draws)) exp3 /15) Bad draws are games, which were drawn before endgame was reached (except the games, were the engine had a material disadvantage) and draws after the engine had a material advantage of at least 1 pawn in the game. ***************************************************************************** *** Evaluated file: a.pgn *** ***************************************************************************** bad Rank EAS-Score sacs shorts draws Engine/player ------------------------------------------------------------- 1 720731 61.78% 90.97% 11.57% "OpenTal 1.0" 2 389453 33.84% 77.98% 16.51% "Stockfish 5 Fury" 3 366802 30.27% 80.27% 16.18% "Velvet 4.0.0" 4 320254 25.03% 74.97% 15.99% "DanaSah 9.0" 5 316163 29.58% 77.34% 14.89% "Velvet 4.1.0" 6 311120 16.07% 72.53% 19.04% "Maverick 1.5" 7 307238 18.62% 78.06% 16.11% "SOS 5" 8 300415 22.08% 76.07% 18.67% "Protej 0.6.3" 9 297864 11.42% 76.90% 23.96% "Leorik 1.0" 10 297571 23.74% 71.40% 21.53% "Stockfish 5" 11 296229 25.28% 73.58% 14.51% "CSTal (10m) R14.1.02" 12 292642 09.65% 81.11% 13.33% "MadChess 1.1" 13 283048 19.20% 75.53% 22.46% "Pedone 3.0" 14 278996 16.41% 79.87% 24.37% "Giraffe 2016-10-23" 15 277228 09.10% 70.28% 24.89% "Leorik 2.0" 16 274102 07.60% 78.56% 18.94% "MadChess 1.2" 17 269510 17.94% 73.71% 15.68% "Prophet 4.2" 18 268219 23.51% 64.68% 16.82% "Expositor 2WN29" 19 267534 12.07% 70.84% 33.51% "MinimalChess 0.6" 20 264831 10.74% 78.28% 29.51% "FrankyGo 1.0.3"
Most aggressive Whittington / Schroeder
Moderator: Ras
-
swami
- Posts: 6662
- Joined: Thu Mar 09, 2006 4:21 am
Re: Most aggressive Whittington / Schroeder
-
michaelnill
- Posts: 9
- Joined: Wed Aug 31, 2022 9:25 pm
- Full name: Michael Nill
Re: Most aggressive Whittington / Schroeder
First of all I want to thank you for your tools and tests you provide us.pohl4711 wrote: ↑Thu Oct 06, 2022 7:13 am
I am sorry, but my EAS-list is calculated out of the testgames played for my SPCC-ratinglist. And this list is focused on the best engines. So, engines like OpenTal or Rodent are much too weak for a testrun.But Pedone 3 and Rebel 14.1 were tested by me. And they are very aggressive.
You can take a look at my full ratinglist, with all engines since 2020 in:
https://www.sp-cc.de/files/spcc_full_list.txt
Scroll down to the end of the list, there you find the EAS-ratinglist of all these engines (more than 1 million games played, each EAS-calculation takes more than 2.5 hours...)... Pedone 3 is on rank 3 and Rebel 14.1 on rank 6.
The only engine, I could test, is ShashChess24 - (HighTal personality). I will do that in the near future. But right now, I am testing Stockfish 221004, then Berserk 10 and Alexandria 3 will follow, so please be patient.
But I would love to see the EAS-ratinglist, you have done, with all the engines, you mentioned here (icluding the second ratinglist with all statistics, the tool builds). Perhaps as a small download, because the second ratinglist is not fitting in a talkchess-thread-window.
By the way: What about Uralochka 3.37c ? And Velvet 4.0.0? Very aggressive (see my full EAS-ratinglist or my Top10 alltime EAS-ratinglist at the bottom of my EAS-site:
https://www.sp-cc.de/eas-ratinglist.htm
Regards - Stefan (SPCC)
Your tests have indeed little bit of a different purpose than mine and look very professionally done.
Before I begin I want to mention that these tests of mine have little to no purpose or rather strong evidence and should be taken as something fun I did in my spare time.
This is the latest test I did. Opening suite used was my 1500+ gambit openings, (30s+ 0.5s inc) under cutechess-cli
Initially there were played total of 15000 games, i.e every engine played against HIARCS14 1500 games.
I filtered out the draws and duplicate games and wins outside move #32.
You can download the games played in PGN format here: https://pixeldrain.com/u/q4eszhRz
You can download the opening suite of 1500+ gambits here: https://pixeldrain.com/u/fwi91Lkq
Just to mention that his was just one of the few tests that I did and plays just a fraction in my ranking of top 10 aggressive engines
Other tests I don't know if I should merge because they were played are under different conditions played without an opening book, under different time control (1m+1s) and different GUI.
inside this code snippet you will find both EAS ratinglists of the games played:
Code: Select all
*****************************************************************************
*** Engine Aggressiveness Tool V4.1 Score points Ratinglist
*****************************************************************************
Score is: (percent*100) of the percent-values of the sacs (1-5+ pawnunits)
calculated out of the won games by the engine, only.
So, a weak engine (with a small number of won games) can get
a high EAS-scoring, too, when the percent of sac-games in the won games
is high (and the number of short wins). Higher pawnunits-sacs give bonus-points:
1 pawnsac = 2x points *** 2 pawnsac = 10x points *** 3 pawnsac = 20x points
4 pawnsac = 40x points *** 5+ pawnsac = 80x points *** 5+ Queensac = 150x points
Additionally, very short won games (percent*100) give bonus-points. If the short
games ended before endgame (a check for low material is done), then these points
are quadrupled (4x):
60 moves= 1x points *** 55 moves= 2x points *** 50 moves= 5x points
45 moves= 10x points *** 40 moves= 30x points.
Since V4.1, the tool also searches for "bad draws" and adds points for the good
draws ((percent of good draws (out of all draws)) exp3 /15)
Bad draws are games, which were drawn before endgame was reached (except the
games, were the engine had a material disadvantage) and draws
after the engine had a material advantage of at least 1 pawn in the game.
*****************************************************************************
*** Evaluated file: 10MostAggressive_noDPL_noDRW_noCMT_clean.pgn ***
*****************************************************************************
bad
Rank EAS-Score sacs shorts draws Engine/player
-------------------------------------------------------------
1 1200000 00.00% 100.00% 00.00% "Hiarcs14WCSC"
2 901556 08.06% 100.00% 00.00% "opental_x64popcnt"
3 894268 16.38% 100.00% 00.00% "rodent-IV - Suicidal"
4 888146 14.46% 100.00% 00.00% "DisasterArea-1.65w64pc"
5 887668 03.57% 100.00% 00.00% "ShashChess24 - HighTal"
6 863176 07.05% 100.00% 00.00% "Zappa Mexico II x64"
7 855784 06.28% 100.00% 00.00% "spark-1.0-win64-mp"
8 845694 06.33% 100.00% 00.00% "Wasp450-x64-modern"
9 826842 03.60% 100.00% 00.00% "Pedone_win_bmi2"
10 816268 03.58% 100.00% 00.00% "Wasp600-windows-avx"
11 775698 03.06% 100.00% 00.00% "Rebel-14.1.02-ChrisW-NNUE-Tal-0-1MinuteBlitzer-AVX2"
*****************************************************************************
*****************************************************************************
*****************************************************************************
*** 2nd Ratinglist with all stats in percent-values *************************
*****************************************************************************
bad
Rank EAS-Score wins all sacs sacsQ sacs5+ sacs4 sacs3 sacs2 sacs1 all shorts short40 short45 short50 short55 short60 draws Engine/player
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1 1200000 10 00.00% =[00.00% + 00.00% + 00.00% + 00.00% + 00.00% + 00.00%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Hiarcs14WCSC"
2 901556 1253 08.06% =[00.00% + 00.64% + 00.48% + 01.36% + 02.15% + 03.43%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "opental_x64popcnt"
3 894268 1233 16.38% =[00.00% + 02.27% + 03.24% + 03.08% + 03.65% + 04.14%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "rodent-IV - Suicidal"
4 888146 1210 14.46% =[00.00% + 01.07% + 02.56% + 02.15% + 04.05% + 04.63%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "DisasterArea-1.65w64pc"
5 887668 1093 03.57% =[00.00% + 00.00% + 00.37% + 00.55% + 00.91% + 01.74%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "ShashChess24 - HighTal"
6 863176 1050 07.05% =[00.00% + 00.10% + 00.67% + 01.71% + 02.10% + 02.48%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Zappa Mexico II x64"
7 855784 1242 06.28% =[00.00% + 00.24% + 00.56% + 00.97% + 02.09% + 02.42%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "spark-1.0-win64-mp"
8 845694 1121 06.33% =[00.00% + 00.36% + 00.80% + 00.80% + 02.05% + 02.32%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Wasp450-x64-modern"
9 826842 1110 03.60% =[00.00% + 00.18% + 00.09% + 00.54% + 01.53% + 01.26%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Pedone_win_bmi2"
10 816268 1060 03.58% =[00.00% + 00.19% + 00.19% + 00.28% + 01.04% + 01.89%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Wasp600-windows-avx"
11 775698 948 03.06% =[00.00% + 00.21% + 00.11% + 00.63% + 01.27% + 00.84%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Rebel-14.1.02-ChrisW-NNUE-Tal-0-1MinuteBlitzer-AVX2"
*****************************************************************************
****************************************************
*** EAS-tool (C) 2022 Stefan Pohl (www.sp-cc.de) ***
****************************************************
-
pohl4711
- Posts: 2813
- Joined: Sat Sep 03, 2011 7:25 am
- Location: Berlin, Germany
- Full name: Stefan Pohl
Re: Most aggressive Whittington / Schroeder
Thank you for posting this result, which reminds me, that I am a complete idiot: When percent gets 100%, the output is one character too long... I fixed this. Now, 100% is printed as 100.0% instead of 100.00%...michaelnill wrote: ↑Thu Oct 06, 2022 1:01 pmFirst of all I want to thank you for your tools and tests you provide us.pohl4711 wrote: ↑Thu Oct 06, 2022 7:13 am
I am sorry, but my EAS-list is calculated out of the testgames played for my SPCC-ratinglist. And this list is focused on the best engines. So, engines like OpenTal or Rodent are much too weak for a testrun.But Pedone 3 and Rebel 14.1 were tested by me. And they are very aggressive.
You can take a look at my full ratinglist, with all engines since 2020 in:
https://www.sp-cc.de/files/spcc_full_list.txt
Scroll down to the end of the list, there you find the EAS-ratinglist of all these engines (more than 1 million games played, each EAS-calculation takes more than 2.5 hours...)... Pedone 3 is on rank 3 and Rebel 14.1 on rank 6.
The only engine, I could test, is ShashChess24 - (HighTal personality). I will do that in the near future. But right now, I am testing Stockfish 221004, then Berserk 10 and Alexandria 3 will follow, so please be patient.
But I would love to see the EAS-ratinglist, you have done, with all the engines, you mentioned here (icluding the second ratinglist with all statistics, the tool builds). Perhaps as a small download, because the second ratinglist is not fitting in a talkchess-thread-window.
By the way: What about Uralochka 3.37c ? And Velvet 4.0.0? Very aggressive (see my full EAS-ratinglist or my Top10 alltime EAS-ratinglist at the bottom of my EAS-site:
https://www.sp-cc.de/eas-ratinglist.htm
Regards - Stefan (SPCC)
Your tests have indeed little bit of a different purpose than mine and look very professionally done.
Before I begin I want to mention that these tests of mine have little to no purpose or rather strong evidence and should be taken as something fun I did in my spare time.
This is the latest test I did. Opening suite used was my 1500+ gambit openings, (30s+ 0.5s inc) under cutechess-cli
Initially there were played total of 15000 games, i.e every engine played against HIARCS14 1500 games.
I filtered out the draws and duplicate games and wins outside move #32.
You can download the games played in PGN format here: https://pixeldrain.com/u/q4eszhRz
You can download the opening suite of 1500+ gambits here: https://pixeldrain.com/u/fwi91Lkq
Just to mention that his was just one of the few tests that I did and plays just a fraction in my ranking of top 10 aggressive engines
Other tests I don't know if I should merge because they were played are under different conditions played without an opening book, under different time control (1m+1s) and different GUI.
inside this code snippet you will find both EAS ratinglists of the games played:
Thank you and I look forward to see ShashChess24 enter your arena.Code: Select all
***************************************************************************** *** Engine Aggressiveness Tool V4.1 Score points Ratinglist ***************************************************************************** Score is: (percent*100) of the percent-values of the sacs (1-5+ pawnunits) calculated out of the won games by the engine, only. So, a weak engine (with a small number of won games) can get a high EAS-scoring, too, when the percent of sac-games in the won games is high (and the number of short wins). Higher pawnunits-sacs give bonus-points: 1 pawnsac = 2x points *** 2 pawnsac = 10x points *** 3 pawnsac = 20x points 4 pawnsac = 40x points *** 5+ pawnsac = 80x points *** 5+ Queensac = 150x points Additionally, very short won games (percent*100) give bonus-points. If the short games ended before endgame (a check for low material is done), then these points are quadrupled (4x): 60 moves= 1x points *** 55 moves= 2x points *** 50 moves= 5x points 45 moves= 10x points *** 40 moves= 30x points. Since V4.1, the tool also searches for "bad draws" and adds points for the good draws ((percent of good draws (out of all draws)) exp3 /15) Bad draws are games, which were drawn before endgame was reached (except the games, were the engine had a material disadvantage) and draws after the engine had a material advantage of at least 1 pawn in the game. ***************************************************************************** *** Evaluated file: 10MostAggressive_noDPL_noDRW_noCMT_clean.pgn *** ***************************************************************************** bad Rank EAS-Score sacs shorts draws Engine/player ------------------------------------------------------------- 1 1200000 00.00% 100.00% 00.00% "Hiarcs14WCSC" 2 901556 08.06% 100.00% 00.00% "opental_x64popcnt" 3 894268 16.38% 100.00% 00.00% "rodent-IV - Suicidal" 4 888146 14.46% 100.00% 00.00% "DisasterArea-1.65w64pc" 5 887668 03.57% 100.00% 00.00% "ShashChess24 - HighTal" 6 863176 07.05% 100.00% 00.00% "Zappa Mexico II x64" 7 855784 06.28% 100.00% 00.00% "spark-1.0-win64-mp" 8 845694 06.33% 100.00% 00.00% "Wasp450-x64-modern" 9 826842 03.60% 100.00% 00.00% "Pedone_win_bmi2" 10 816268 03.58% 100.00% 00.00% "Wasp600-windows-avx" 11 775698 03.06% 100.00% 00.00% "Rebel-14.1.02-ChrisW-NNUE-Tal-0-1MinuteBlitzer-AVX2" ***************************************************************************** ***************************************************************************** ***************************************************************************** *** 2nd Ratinglist with all stats in percent-values ************************* ***************************************************************************** bad Rank EAS-Score wins all sacs sacsQ sacs5+ sacs4 sacs3 sacs2 sacs1 all shorts short40 short45 short50 short55 short60 draws Engine/player -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 1 1200000 10 00.00% =[00.00% + 00.00% + 00.00% + 00.00% + 00.00% + 00.00%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Hiarcs14WCSC" 2 901556 1253 08.06% =[00.00% + 00.64% + 00.48% + 01.36% + 02.15% + 03.43%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "opental_x64popcnt" 3 894268 1233 16.38% =[00.00% + 02.27% + 03.24% + 03.08% + 03.65% + 04.14%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "rodent-IV - Suicidal" 4 888146 1210 14.46% =[00.00% + 01.07% + 02.56% + 02.15% + 04.05% + 04.63%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "DisasterArea-1.65w64pc" 5 887668 1093 03.57% =[00.00% + 00.00% + 00.37% + 00.55% + 00.91% + 01.74%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "ShashChess24 - HighTal" 6 863176 1050 07.05% =[00.00% + 00.10% + 00.67% + 01.71% + 02.10% + 02.48%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Zappa Mexico II x64" 7 855784 1242 06.28% =[00.00% + 00.24% + 00.56% + 00.97% + 02.09% + 02.42%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "spark-1.0-win64-mp" 8 845694 1121 06.33% =[00.00% + 00.36% + 00.80% + 00.80% + 02.05% + 02.32%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Wasp450-x64-modern" 9 826842 1110 03.60% =[00.00% + 00.18% + 00.09% + 00.54% + 01.53% + 01.26%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Pedone_win_bmi2" 10 816268 1060 03.58% =[00.00% + 00.19% + 00.19% + 00.28% + 01.04% + 01.89%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Wasp600-windows-avx" 11 775698 948 03.06% =[00.00% + 00.21% + 00.11% + 00.63% + 01.27% + 00.84%] 100.00% = [100.00% + 00.00% + 00.00% + 00.00% + 00.00%] 00.00% "Rebel-14.1.02-ChrisW-NNUE-Tal-0-1MinuteBlitzer-AVX2" ***************************************************************************** **************************************************** *** EAS-tool (C) 2022 Stefan Pohl (www.sp-cc.de) *** ****************************************************
I made small updates for my EAS-tool, Sacrifice Games Search Tools and Short Games Analyzer Tools...Main change is the percent-routine (percent-calculation is really complex in batch-files, because batch allows integer-varibales, only...), which now prevents possible broken outputs, if the percent-value is 0 or 100. Please redownload the Tools from my website.
-
pohl4711
- Posts: 2813
- Joined: Sat Sep 03, 2011 7:25 am
- Location: Berlin, Germany
- Full name: Stefan Pohl
Re: Most aggressive Whittington / Schroeder
I did a 2500 games quicktest of ShashChess24 HighTal vs. Stockfish 15 (singlethread, 60sec+600ms, UHO openings (otherwise 90% draws, with UHO around 50%). And used my EAS-tool. The result is disappointing for the "High Tal"-setting:michaelnill wrote: ↑Thu Oct 06, 2022 1:16 ambit late to the party, not long ago I tested HIARCS14 at 1600 elo (around 1900 FIDE) versus various engines. Was curious to see which engine can manage to checkmate the fastest i.e which engine goes "straight for the throat" against non-perfect play response. I made them play thousands of bullet games, some tests done without opening book and others using gambit opening suite. I still cannot say that I have come to any particular conclusion, mainly because I don't know how big the correlation between "most aggressive" and engine prioritizing heavy attacks on opponent king is. But according to the statistics these engines scored the best:Werewolf wrote: ↑Mon Sep 19, 2022 3:21 pm I'm doing work on a quantitative "aggression test". It involves beating a 2100 FIDE Elo engine in certain positions and shows an engine's tendencies towards king attacks. It doesn't test for strength.
What is the most aggressive chess engine out there, over 2700 elo? I was wondering if Ed & Chris had cooked something up in this department...is the new CS_Tal still private for instance?
1. DisasterArea-1.65 (tweaked parameters)
2. rodent-IV - (Suicidal Gambit personality)
3. opental 1.1
4. Wasp 4.5 (tweaked parameters)
5. spark 1.0 (tweaked parameters)
6. Zappa Mexico II (tweaked parameters)
7. ShashChess24 - (HighTal personality)
8. Pedone 3.0
9. Wasp 6.0
10. Rebel-14.1.2 CSTal
...
42. ChessMind_082
if anyone is interested I will post full .pgn database and statistics as well as the overall EAS score.
P.S I can provide download links to the engine and parameters used for anyone interested too..
P.P.S I'd love to see if Mr.Pohl is interested to test these engines
Code: Select all
*****************************************************************************
*** Evaluated file: resultsShash.pgn ***
*****************************************************************************
bad
Rank EAS-Score sacs shorts draws Engine/player
-------------------------------------------------------------
1 52494 18.88% 12.45% 17.35% "Stockfish 15 avx2"
2 33417 11.55% 07.59% 29.03% "ShashChess24 HighTal"
*****************************************************************************
*****************************************************************************
*****************************************************************************
*** 2nd Ratinglist with all stats in percent-values *************************
*****************************************************************************
bad
Rank EAS-Score wins all sacs sacsQ sacs5+ sacs4 sacs3 sacs2 sacs1 all shorts short40 short45 short50 short55 short60 draws Engine/player
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1 52494 715 18.88% =[00.00% + 00.00% + 00.56% + 01.82% + 05.59% + 10.91%] 12.45% = [00.00% + 00.00% + 01.40% + 02.80% + 08.25%] 17.35% "Stockfish 15 avx2"
2 33417 580 11.55% =[00.00% + 00.17% + 00.17% + 01.03% + 03.10% + 07.07%] 07.59% = [00.00% + 00.17% + 00.86% + 01.72% + 04.83%] 29.03% "ShashChess24 HighTal"
*****************************************************************************
So, a testrun for my SPCC-ratinglist does not make sense for me...
-
Vinvin
- Posts: 5298
- Joined: Thu Mar 09, 2006 9:40 am
- Full name: Vincent Lejeune
Re: Most aggressive Whittington / Schroeder
IIRC, there were the Kamikaze personality for Shredder who was very aggressive !
michaelnill wrote: ↑Thu Oct 06, 2022 2:15 am 11. Rodent IV 0.33 Tal
12. ShashChess 16
13. dragon-3.1 - Aggressive
14. Critter_1.6a_64bit
15. stockfish-211-64-ja PHQ
16. Fizbo 2
17. Hakkapeliitta 3.0
18. gfruit
19. Capivara LK 0.09b02g
20. gothmog_10b10
21. benjamin2
22. Komodo 1.0 KingHunter
23. Szint Tal
24. Crafty 25ja
25. Delphil 3.2
26. Hannibal 1.7
27. MP-x64-Active--Thinker
28. Little Goliath Evolution
29. Chess Tiger 2007
30. Deep Junior Yokohama
31. Little Goliath Evolution, 3.15.4
32. Velvet 4.0
33. Theking 3.5
34. Glass 2.0 PERSONALITY
35. Amyan 1.72
36. Comet A90
37. Monarch 1.7
38. Gandalf 6
39. Schooner 2.0.34
40. Maverick 1.5
41. FireBird 1.2
42. trappybeowulf-rev11-64-ja
Some of those parameters I found online, others are my invention.
I don't know if I can post link to the engine collection here but anyone feel free to contact me privately.
-
michaelnill
- Posts: 9
- Joined: Wed Aug 31, 2022 9:25 pm
- Full name: Michael Nill
Re: Most aggressive Whittington / Schroeder
Now this I find very interesting, did you make sure to disable NNUE on ShashChess24? My apologies for never mentioning it but with engines that make use of NNUE (such as Dragon, ShashChess) this feature need to be disabled in order for the "personality" to take over.
Other than that I have nothing to add, I am in pure disbelief to see SF 15 be more aggressive than ShashChess with HighTal personality.
-
michaelnill
- Posts: 9
- Joined: Wed Aug 31, 2022 9:25 pm
- Full name: Michael Nill
Re: Most aggressive Whittington / Schroeder
True! How could I forget about it..
Now that you mention it maybe I should add HIARCS Aggressive too
Here are the settings for Kamikaze personality:
Code: Select all
[ENGINE]
Name=Kamikaze Classic
Author=Stefan Meyer-Kahlen
Filename=EngineClassic.exe
[OPTIONS]
Queen=90
Queen (Endgame)=90
Rook=85
Rook (Endgame)=90
Bishop=80
Bishop (Endgame)=85
Knight=80
Knight (Endgame)=85
Pawn=75
Pawn (Endgame)=85
Draw Score=-20
King Safety=150
Center Control=120
Pawn Structure=85
Pawn Structure (Endgame)=90
Passed Pawns=105
Passed Pawns (Endgame)=110
Bishop Pair=105
Bishop Pair (Endgame)=105Code: Select all
[ENGINE]
Name=Gambit Classic
Author=Stefan Meyer-Kahlen
Filename=EngineClassic.exe
[OPTIONS]
Queen=110
Queen (Endgame)=105
Rook=90
Rook (Endgame)=95
Bishop=90
Bishop (Endgame)=95
Knight=90
Knight (Endgame)=95
Pawn=85
Pawn (Endgame)=95
Draw Score=-10
King Safety=130
Center Control=120
Pawn Structure=95
Passed Pawns=105
Passed Pawns (Endgame)=110
Bishop Pair=105
Bishop Pair (Endgame)=105
-
pohl4711
- Posts: 2813
- Joined: Sat Sep 03, 2011 7:25 am
- Location: Berlin, Germany
- Full name: Stefan Pohl
Re: Most aggressive Whittington / Schroeder
Oh. No, I did not disable the nnue-net. My bad (but for my excuse: KomodoDragon switches the net off automatically, if "aggressive" is chosen). I will retry. But then ShashChess will be much weaker. So, vs. SF 15 makes no sense. I will try Stockfish final HCE as opponent.michaelnill wrote: ↑Sat Oct 08, 2022 10:05 amNow this I find very interesting, did you make sure to disable NNUE on ShashChess24? My apologies for never mentioning it but with engines that make use of NNUE (such as Dragon, ShashChess) this feature need to be disabled in order for the "personality" to take over.
Other than that I have nothing to add, I am in pure disbelief to see SF 15 be more aggressive than ShashChess with HighTal personality.
-
Rowen
- Posts: 108
- Joined: Tue Nov 15, 2016 1:19 pm
- Location: Cheshire, England
Re: Most aggressive Whittington / Schroeder
Himichaelnill wrote: ↑Sat Oct 08, 2022 10:05 amNow this I find very interesting, did you make sure to disable NNUE on ShashChess24? My apologies for never mentioning it but with engines that make use of NNUE (such as Dragon, ShashChess) this feature need to be disabled in order for the "personality" to take over.
Other than that I have nothing to add, I am in pure disbelief to see SF 15 be more aggressive than ShashChess with HighTal personality.
To what extent does NNUE enabled affect personality settings (parameters)? I noticed you tested Rebel 14.1.2 CSTAl with NNUE but you changed parameters, Also if using Shaschess personality settings such as TAl (Shashin theory?) does that mean NNUE should not be enabled? I dont remember reading that anywhere, I am just wondering? Thanks
-
michaelnill
- Posts: 9
- Joined: Wed Aug 31, 2022 9:25 pm
- Full name: Michael Nill
Re: Most aggressive Whittington / Schroeder
If I understand correctly when NNUE is enabled the network dictates the evaluation values.Rowen wrote: ↑Sat Oct 08, 2022 7:39 pm
Hi
To what extent does NNUE enabled affect personality settings (parameters)? I noticed you tested Rebel 14.1.2 CSTAl with NNUE but you changed parameters, Also if using Shaschess personality settings such as TAl (Shashin theory?) does that mean NNUE should not be enabled? I dont remember reading that anywhere, I am just wondering? Thanks
In other hand if you want to tweak or rather modify some parts of the evaluation you need to disable the NNUE or if you want to be more technical set it to use regular evaluation. From what I understand personalities are just modified parameters that affect the evaluation, correct me if I am wrong.
I read about this on Komodo website, as mr.Pohl mentioned in his reply the Komodo team has automatic function that switches to regular evaluation whenever there is selected non-default personality for Dragon.