New Rating List ?

Subject: New Rating List ? Fri Oct 06, 2023 10:52 pm

I applaud to Stefan's new setup of his rating list --> less draws among the top engines. I want to promote the idea further using the TCEC positions composed by Jeroen Noomen and GM Matthew Sadler and made a setup, a pilot run gives -

Code:: No. Name Win Draw Loss Unf. Score Games %
----------------------------------------------------------------
  1 SF16 +32 =20 -0 *0 42.0 52 80.8%
  2 Berserk-11.1 +18 =26 -9 *0 31.0 53 58.5%
  3 Komodo-Dragon-2.5 +19 =23 -10 *0 30.5 52 58.7%
  4 Rebel-Dev +15 =28 -9 *0 29.0 52 55.8%
  5 Ethereal-14.00-NNUE +14 =27 -11 *0 27.5 52 52.9%
  6 CST-2.0 +13 =26 -13 *0 26.0 52 50.0%
  7 RubiChess-20230918 +16 =20 -16 *0 26.0 52 50.0%
  8 Koivisto-9.0 +13 =21 -18 *0 23.5 52 45.2%
  9 Caissa-1.13.1 +8 =28 -17 *0 22.0 53 41.5%
10 Rebel-EAS +10 =23 -20 *0 21.5 53 40.6%
11 Clover-6.0 +9 =24 -20 *0 21.0 53 39.6%
12 Revenge-3.0 +7 =28 -18 *0 21.0 53 39.6%
13 Igel-3.5.0 +9 =22 -22 *0 20.0 53 37.7%

Total Games: 341
White Wins: 157 (46.0%)
Black Wins: 26 (7.6%)
Draws: 158 (46.3%)
Unfinished: 0 (0.0%)

Draw rate 46%

Ordo rating list

Code::    # PLAYER : RATING ERROR POINTS PLAYED (%) CFS(%) W D L D(%)
   1 SF16 : 3760.6 61.2 43.0 54 80 100 32 22 0 41
   2 Komodo-Dragon-2.5 : 3584.1 44.2 32.0 54 59 73 20 24 10 44
   3 Berserk-11.1 : 3561.7 54.5 31.5 55 57 77 18 27 10 49
   4 Rebel-Dev : 3531.5 54.7 30.5 54 56 62 16 29 9 54
   5 RubiChess-20230918 : 3519.6 55.7 27.0 54 50 54 16 22 16 41
   6 Ethereal-14.00-NNUE : 3514.3 71.5 29.0 55 53 59 14 30 11 55
   7 CST-2.0 : 3507.1 41.6 27.5 54 51 96 14 27 13 50
   8 Koivisto-9.0 : 3450.7 44.1 23.5 54 44 61 13 21 20 39
   9 Rebel-EAS : 3439.2 57.8 23.5 55 43 63 12 23 20 42
  10 Clover-6.0 : 3427.2 51.8 22.0 55 40 60 10 24 21 44
  11 Caissa-1.13.1 : 3417.0 46.5 23.0 55 42 71 8 30 17 55
  12 Igel-3.5.0 : 3396.8 44.2 20.0 55 36 56 9 22 24 40
  13 Revenge-3.0 : 3390.1 46.2 21.5 54 40 --- 7 29 18 54

Although few games I very much like the big elo difference between SF and Komodo, IMO it is real when push comes to shove.

[Q] - before going for real, I tried to involve the new Rofchade 3.1 but as version 3.0 crashes under cutechess, any idea?

Posts : 42 Join date : 2022-06-08

Nice to see new lists.. i always hope they finally test a engine against all engines with same total games in a list to get more accurate results.. good luck!

About Rofchade ,i'm using on 3 systems Cutechess Gui and have been testing RofChade different versions regular and have not see any crashes yet!

RofChade 3.1 in list: https://ipmanchess.yolasite.com/r9-7945hx.php

Ipman.

Subject: Re: New Rating List ? Sat Oct 07, 2023 9:42 am

Absolutely, robin round, all vs all, small pool is doable.

Regarding Rofchade, tried 2 cutechess versions, can you see what I am doing wrong?

Code:: C:\Cute Chess>ccc -concurrency 5 -tournament gauntlet -engine name=REBEL-EAS cmd=REBEL-EAS.exe proto=uci option.Hash=128 -engine name=rofChade-3.1 cmd=rofChade-3.1.exe proto=uci option.Hash=128 option.NetFile=230620-224.nnue -each restart=on tc=40/10 timemargin=500 -draw movenumber=160 movecount=3 score=100 -resign movecount=3 score=999 -rounds 10 -repeat -pgnout all.pgn -openings file=fq.pgn plies=40
Warning: 2 opening repetitions vs 1 games per encounter
Started game 5 of 10 (REBEL-EAS vs rofChade-3.1)
Started game 3 of 10 (REBEL-EAS vs rofChade-3.1)
Started game 4 of 10 (rofChade-3.1 vs REBEL-EAS)
Started game 2 of 10 (rofChade-3.1 vs REBEL-EAS)
Started game 1 of 10 (REBEL-EAS vs rofChade-3.1)
Terminating process of engine rofChade-3.1(6)
Finished game 4 (rofChade-3.1 vs REBEL-EAS): 0-1 {White's connection stalls}
Score of REBEL-EAS vs rofChade-3.1: 1 - 0 - 0 [1.000] 1
Terminating process of engine rofChade-3.1(9)
Terminating process of engine rofChade-3.1(5)
Terminating process of engine rofChade-3.1(1)
Terminating process of engine rofChade-3.1(2)
Finished game 5 (REBEL-EAS vs rofChade-3.1): * {No result}
Score of REBEL-EAS vs rofChade-3.1: 1 - 0 - 0 [1.000] 1
Finished game 3 (REBEL-EAS vs rofChade-3.1): * {No result}
Score of REBEL-EAS vs rofChade-3.1: 1 - 0 - 0 [1.000] 1
Finished game 1 (REBEL-EAS vs rofChade-3.1): * {No result}
Score of REBEL-EAS vs rofChade-3.1: 1 - 0 - 0 [1.000] 1
Finished game 2 (rofChade-3.1 vs REBEL-EAS): * {No result}
Score of REBEL-EAS vs rofChade-3.1: 1 - 0 - 0 [1.000] 1
... REBEL-EAS playing Black: 1 - 0 - 0 [1.000] 1
... White vs Black: 0 - 1 - 0 [0.000] 1
Elo difference: inf +/- nan, LOS: 84.1 %, DrawRatio: 0.0 %
Finished match

C:\Cute Chess>pause
Press any key to continue . . .

Posts : 159 Join date : 2022-03-01 Location : Berlin

Admin wrote:: Regarding Rofchade, tried 2 cutechess versions, can you see what I am doing wrong?

Put Rofchade binary and the net in a folder. And do not use the uci-option for naming the net, it is not necessary. Only Threads and Hash need a uci-option setting. Right now, my Rofchade 3.1 test using cutechess-cli, is running: More than 10000 games played. No crashes, timelosses, disconnects.

My Rofchade-line looks like this:

-engine name="RofChade 3.1 avx2" dir="C:/Cutechess/Engines/AB_testruns/RofChade_3.1" option.Hash=256 option.Threads=1 cmd="rofChadeAVX2.exe" proto=uci

Posts : 159 Join date : 2022-03-01 Location : Berlin

Admin wrote:: Absolutely, robin round, all vs all, small pool is doable.

Don't forget the additionally Gamepair-Ratinglist and Gamepair-Statistics, which are very interesting and the EAS-Ratinglist calculated out of the played games...
No other Ratinglist has this to offer. And Gamepairs are the most natural way of looking at results of games, played with biased openings. As Vondele (Stockfish Maintainer) mentioned: "Thinking uniquely in game pairs makes sense with the biased openings used these days. While pentanomial makes sense it is a bit complicated so we could simplify and score game pairs only (not games) as W-L-D (a traditional score of 2-0, or 1.5-0.5 is just a W)."

https://www.sp-cc.de/files/uho_top15_gamepair.txt

Example: Gamepair result of the 1000 games of Stockfish 230929 vs. KomodoDragon 3.2:
500 gamepairs (+375, =118, -7) = 86.8%... So KomodoDragon just won 7 out of 500 gamepairs - ouch...

https://www.sp-cc.de/eas-ratinglist.htm

PS: And not to forget: On my website, you can replay some of the most spectacular sac-games from the latest testrun, using the ChessBase pgn-viewer:
https://www.sp-cc.de/view-games-with-sacs.htm

Example: This crazy game by Stockfish Dev vs. KomodoDragon 3.2:

[pgn]
[Event "UHO Ratinglist"]
[Site "?"]
[Date "2023.10.05"]
[Round "352"]
[White "Stockfish 230929 avx2"]
[Black "KomodoDragon 3.2 avx2"]
[Result "1-0"]
[ECO "A45"]
[Opening "Trompovsky attack (Ruth, Opovcensky Opening)"]
[TimeControl "180+1"]
[PlyCount "79"]
[GameDuration "00"]
[GameEndTime "2023-10-05T09.221 Mitteleuropäische Sommerzeit"]
[GameStartTime "2023-10-05T09.592 Mitteleuropäische Sommerzeit"]

1. d4 { book } 1... Nf6 { book } 2. Bg5 { book } 2... c5 { book } 3. d5 {
book } 3... Qb6 { book } 4. Nc3 { book } 4... Qxb2 { book } 5. Bd2 { book }
5... Qb6 { book } 6. e4 { book } 6... e5 { book } 7. g4 { +1.08/27 13s }
7... d6 { -0.74/23 16s } 8. g5 { +1.03/29 3.5s } 8... Ng8 { -0.81/21 3.1s }
9. Rb1 { +1.13/28 32s } 9... Qd8 { -0.98/23 7.6s } 10. f4 { +1.26/25 4.8s }
10... exf4 { -0.96/21 5.9s } 11. Nf3 { +1.32/25 14s } 11... Nd7 { -0.86/21
5.3s } 12. Bxf4 { +1.45/22 2.0s } 12... h6 { -1.01/21 3.4s } 13. g6 {
+1.50/25 7.5s } 13... fxg6 { -0.95/22 14s } 14. Bg2 { +1.57/26 21s } 14...
Ngf6 { -0.78/20 4.4s } 15. O-O { +1.65/22 3.0s } 15... Nh5 { -1.00/21 8.4s
} 16. Bg3 { +1.32/24 6.8s } 16... Qa5 { -1.29/24 17s } 17. Qe1 { +1.94/21
2.8s } 17... Nxg3 { -1.18/21 11s } 18. hxg3 { +2.19/21 2.2s } 18... Ne5 {
-1.03/21 3.4s } 19. Nxe5 { +3.85/22 2.5s } 19... dxe5 { -2.14/18 2.0s } 20.
Qf2 { +4.00/21 2.6s } 20... Qxc3 { 0.00/24 2.6s } 21. Qf7+ { +8.68/22 2.4s
} 21... Kd8 { 0.00/5 0.001s } 22. Rxb7 { +9.80/23 2.6s } 22... Bxb7 {
-7.78/28 11s } 23. Qxb7 { +M43/39 2.2s } 23... Rc8 { -11.34/22 1.6s } 24.
Rf7 { +M39/41 2.2s } 24... Qd4+ { -15.62/25 1.6s } 25. Kh2 { +M33/41 2.5s }
25... Qa4 { -M74/23 0.35s } 26. Bh3 { +M31/41 2.5s } 26... Qxc2+ { -M70/27
0.51s } 27. Kg1 { +M29/39 2.5s } 27... Qd1+ { -M74/28 0.58s } 28. Rf1 {
+M27/39 2.8s } 28... Qxf1+ { -29.04/22 1.9s } 29. Kxf1 { +M25/39 2.3s }
29... Rc7 { -M40/27 0.72s } 30. Qb8+ { +M23/41 3.7s } 30... Ke7 { -M38/27
0.38s } 31. Qxc7+ { +M21/38 3.8s } 31... Kf6 { -M36/27 0.38s } 32. d6 {
+M19/38 2.2s } 32... a6 { -M20/26 1.3s } 33. Qd8+ { +M17/39 2.2s } 33...
Kf7 { -M18/22 0.53s } 34. d7 { +M15/40 2.1s } 34... Kg8 { -M16/23 0.81s }
35. Qe8 { +M11/59 2.2s } 35... Kh7 { -M10/21 0.69s } 36. d8=Q { +M9/78 2.1s
} 36... Rg8 { -M8/20 0.48s } 37. Be6 { +M7/198 2.1s } 37... Rh8 { -M6/23
0.49s } 38. Qf7 { +M5/245 0.18s } 38... h5 { -M4/32 0.58s } 39. Qg5 {
+M3/245 0.013s } 39... a5 { -M2/99 0.012s } 40. Qgxg6# { +M1/245 0.008s,
White mates } 1-0
[/pgn]

Subject: Re: New Rating List ? Sat Oct 07, 2023 11:23 am

pohl4711 wrote:: Example: This crazy game by Stockfish Dev vs. KomodoDragon 3.2:

[pgn]
[Event "UHO Ratinglist"]
[Site "?"]
[Date "2023.10.05"]
[Round "352"]
[White "Stockfish 230929 avx2"]
[Black "KomodoDragon 3.2 avx2"]
[Result "1-0"]
[ECO "A45"]
[Opening "Trompovsky attack (Ruth, Opovcensky Opening)"]
[TimeControl "180+1"]
[PlyCount "79"]
[GameDuration "00"]
[GameEndTime "2023-10-05T09.221 Mitteleuropäische Sommerzeit"]
[GameStartTime "2023-10-05T09.592 Mitteleuropäische Sommerzeit"]

1. d4 { book } 1... Nf6 { book } 2. Bg5 { book } 2... c5 { book } 3. d5 {
book } 3... Qb6 { book } 4. Nc3 { book } 4... Qxb2 { book } 5. Bd2 { book }
5... Qb6 { book } 6. e4 { book } 6... e5 { book } 7. g4 { +1.08/27 13s }
7... d6 { -0.74/23 16s } 8. g5 { +1.03/29 3.5s } 8... Ng8 { -0.81/21 3.1s }
9. Rb1 { +1.13/28 32s } 9... Qd8 { -0.98/23 7.6s } 10. f4 { +1.26/25 4.8s }
10... exf4 { -0.96/21 5.9s } 11. Nf3 { +1.32/25 14s } 11... Nd7 { -0.86/21
5.3s } 12. Bxf4 { +1.45/22 2.0s } 12... h6 { -1.01/21 3.4s } 13. g6 {
+1.50/25 7.5s } 13... fxg6 { -0.95/22 14s } 14. Bg2 { +1.57/26 21s } 14...
Ngf6 { -0.78/20 4.4s } 15. O-O { +1.65/22 3.0s } 15... Nh5 { -1.00/21 8.4s
} 16. Bg3 { +1.32/24 6.8s } 16... Qa5 { -1.29/24 17s } 17. Qe1 { +1.94/21
2.8s } 17... Nxg3 { -1.18/21 11s } 18. hxg3 { +2.19/21 2.2s } 18... Ne5 {
-1.03/21 3.4s } 19. Nxe5 { +3.85/22 2.5s } 19... dxe5 { -2.14/18 2.0s } 20.
Qf2 { +4.00/21 2.6s } 20... Qxc3 { 0.00/24 2.6s } 21. Qf7+ { +8.68/22 2.4s
} 21... Kd8 { 0.00/5 0.001s } 22. Rxb7 { +9.80/23 2.6s } 22... Bxb7 {
-7.78/28 11s } 23. Qxb7 { +M43/39 2.2s } 23... Rc8 { -11.34/22 1.6s } 24.
Rf7 { +M39/41 2.2s } 24... Qd4+ { -15.62/25 1.6s } 25. Kh2 { +M33/41 2.5s }
25... Qa4 { -M74/23 0.35s } 26. Bh3 { +M31/41 2.5s } 26... Qxc2+ { -M70/27
0.51s } 27. Kg1 { +M29/39 2.5s } 27... Qd1+ { -M74/28 0.58s } 28. Rf1 {
+M27/39 2.8s } 28... Qxf1+ { -29.04/22 1.9s } 29. Kxf1 { +M25/39 2.3s }
29... Rc7 { -M40/27 0.72s } 30. Qb8+ { +M23/41 3.7s } 30... Ke7 { -M38/27
0.38s } 31. Qxc7+ { +M21/38 3.8s } 31... Kf6 { -M36/27 0.38s } 32. d6 {
+M19/38 2.2s } 32... a6 { -M20/26 1.3s } 33. Qd8+ { +M17/39 2.2s } 33...
Kf7 { -M18/22 0.53s } 34. d7 { +M15/40 2.1s } 34... Kg8 { -M16/23 0.81s }
35. Qe8 { +M11/59 2.2s } 35... Kh7 { -M10/21 0.69s } 36. d8=Q { +M9/78 2.1s
} 36... Rg8 { -M8/20 0.48s } 37. Be6 { +M7/198 2.1s } 37... Rh8 { -M6/23
0.49s } 38. Qf7 { +M5/245 0.18s } 38... h5 { -M4/32 0.58s } 39. Qg5 {
+M3/245 0.013s } 39... a5 { -M2/99 0.012s } 40. Qgxg6# { +M1/245 0.008s,
White mates } 1-0
[/pgn]

20... Qxc3 { 0.00/24 2.6s }, quite some horizon effect there.

Subject: Re: New Rating List ? Sat Oct 07, 2023 11:31 am

pohl4711 wrote:

Admin wrote:: Regarding Rofchade, tried 2 cutechess versions, can you see what I am doing wrong?

Put Rofchade binary and the net in a folder. And do not use the uci-option for naming the net, it is not necessary. Only Threads and Hash need a uci-option setting. Right now, my Rofchade 3.1 test using cutechess-cli, is running: More than 10000 games played. No crashes, timelosses, disconnects.

My Rofchade-line looks like this:

-engine name="RofChade 3.1 avx2" dir="C:/Cutechess/Engines/AB_testruns/RofChade_3.1" option.Hash=256 option.Threads=1 cmd="rofChadeAVX2.exe" proto=uci

How odd, with the folder trick it works that way....never seen that before. And typing uci the net is empty. Always something with those Dutch clown

Subject: Re: New Rating List ? Sat Oct 07, 2023 12:21 pm

pohl4711 wrote:

Admin wrote:: Absolutely, robin round, all vs all, small pool is doable.

Don't forget the additionally Gamepair-Ratinglist and Gamepair-Statistics, which are very interesting and the EAS-Ratinglist calculated out of the played games...
No other Ratinglist has this to offer. And Gamepairs are the most natural way of looking at results of games, played with biased openings. As Vondele (Stockfish Maintainer) mentioned: "Thinking uniquely in game pairs makes sense with the biased openings used these days. While pentanomial makes sense it is a bit complicated so we could simplify and score game pairs only (not games) as W-L-D (a traditional score of 2-0, or 1.5-0.5 is just a W)."

https://www.sp-cc.de/files/uho_top15_gamepair.txt

Example: Gamepair result of the 1000 games of Stockfish 230929 vs. KomodoDragon 3.2:
500 gamepairs (+375, =118, -7) = 86.8%... So KomodoDragon just won 7 out of 500 gamepairs - ouch...

Yep, SF finally gets what it deserves. SF notable won the last 7 TCEC tournaments.

Tried your gamepair tool, is it possible it doesn't work with FEN pgns?

Code:: [Event "?"]
[Site "?"]
[Date "2023.10.06"]
[Round "1"]
[White "Revenge-3.0"]
[Black "Caissa-1.13.1"]
[Result "1/2-1/2"]
[FEN "1n1qkb1r/rp2pp1p/p1p1bnp1/2PpN3/3P4/1QN1P3/PP1B1PPP/R3KB1R b KQk - 0 1"]
[PlyCount "94"]
[SetUp "1"]
[TimeControl "40/40"]

1... Bg7 {-0.72/20 1.5s} 2. Bd3 {+0.60/19 1.1s} Nbd7 {-0.90/22 1.1s}
3. Nxd7 {+0.69/20 1.6s} Bxd7 {-1.05/20 1.0s} 4. h3 {+0.75/19 0.84s}
Qb8 {-1.18/21 1.9s} 5. Na4 {+0.94/21 2.5s} e5 {-1.39/22 3.5s}
6. dxe5 {+0.78/21 1.9s} Qxe5 {-1.11/23 1.7s} 7. Bc3 {+0.79/22 0.96s}

Get the following error -

Processing head-to-head file 77 of 78
Processing head-to-head file 78 of 78
************************************************************
Step 4: Elo-calculating with ORDO
ERROR: Input file contains no games
************************************************************
-------------------------------------------------------------------
--- Number of all Gamepairs : 0
--- Number of drawn Gamepairs overall: 0 (= 00.00%)
--- Number of 1:1 drawn Gamepairs : 0 (= 00.00%)
--- Number of 2-draws drawn Gamepairs: 0 (= 00.00%)
-------------------------------------------------------------------
************************************************************
Press any key to continue . . .

Subject: Re: New Rating List ? Sat Oct 07, 2023 12:35 pm

I let the pilot run overnight, created the head-to-head

Code::    # PLAYER : RATING ERROR POINTS PLAYED (%) CFS(%) W D L D(%)
   1 SF16 : 3724.6 43.1 127.0 166 77 100 90 74 2 45
   2 Komodo-Dragon-2.5 : 3561.3 42.8 95.0 165 58 61 56 78 31 47
   3 Berserk-11.1 : 3553.2 35.7 95.0 167 57 87 53 84 30 50
   4 Rebel-Dev : 3521.6 31.4 88.0 166 53 67 47 82 37 49
   5 RubiChess-20230918 : 3511.3 41.0 85.5 165 52 77 42 87 36 53
   6 Ethereal-14.00-NNUE : 3490.3 34.5 81.5 167 49 60 36 91 40 54
   7 Koivisto-9.0 : 3482.7 32.5 79.5 165 48 54 39 81 45 49
   8 Clover-6.0 : 3480.9 27.1 79.0 165 48 60 41 76 48 46
   9 CST-2.0 : 3473.3 45.6 76.5 165 46 83 38 77 50 47
  10 Rebel-EAS : 3445.9 42.4 71.0 165 43 55 28 86 51 52
  11 Caissa-1.13.1 : 3442.3 46.4 70.5 165 43 76 32 77 56 47
  12 Revenge-3.0 : 3420.1 34.3 66.0 165 40 87 25 82 58 50
  13 Igel-3.5.0 : 3392.5 32.9 61.5 166 37 --- 24 75 67 45

163 elo difference Exclamation

Head-to-Head

Code:: Head to head statistics:

1) SF16 3724.6 : 166 (+90,=74,-2), 76.5 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   Komodo-Dragon-2.5 : 14 ( 8, 6, 0), 78.6 : +163.3, 34.3, 100.0
   Berserk-11.1 : 14 ( 7, 7, 0), 75.0 : +171.4, 33.6, 100.0
   Rebel-Dev : 14 ( 8, 5, 1), 75.0 : +203.1, 31.6, 100.0
   RubiChess-20230918 : 14 ( 8, 6, 0), 78.6 : +213.3, 32.5, 100.0
   Ethereal-14.00-NNUE : 14 ( 5, 9, 0), 67.9 : +234.3, 28.5, 100.0
   Koivisto-9.0 : 14 ( 8, 6, 0), 78.6 : +241.9, 26.0, 100.0
   Clover-6.0 : 14 ( 9, 5, 0), 82.1 : +243.7, 24.2, 100.0
   CST-2.0 : 14 ( 8, 6, 0), 78.6 : +251.3, 34.8, 100.0
   Rebel-EAS : 14 ( 5, 9, 0), 67.9 : +278.7, 33.3, 100.0
   Caissa-1.13.1 : 13 ( 7, 6, 0), 76.9 : +282.3, 34.9, 100.0
   Revenge-3.0 : 13 ( 7, 5, 1), 73.1 : +304.5, 17.0, 100.0
   Igel-3.5.0 : 14 ( 10, 4, 0), 85.7 : +332.1, 31.0, 100.0

2) Komodo-Dragon-2.5 3561.3 : 165 (+56,=78,-31), 57.6 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 6, 8), 21.4 : -163.3, 34.3, 0.0
   Berserk-11.1 : 14 ( 3, 9, 2), 53.6 : +8.2, 28.0, 61.5
   Rebel-Dev : 14 ( 3, 6, 5), 42.9 : +39.8, 28.0, 92.2
   RubiChess-20230918 : 14 ( 7, 5, 2), 67.9 : +50.1, 15.9, 99.9
   Ethereal-14.00-NNUE : 14 ( 6, 6, 2), 64.3 : +71.1, 27.8, 99.5
   Koivisto-9.0 : 14 ( 4, 9, 1), 60.7 : +78.6, 33.8, 99.0
   Clover-6.0 : 13 ( 4, 6, 3), 53.8 : +80.5, 27.6, 99.8
   CST-2.0 : 14 ( 4, 8, 2), 57.1 : +88.0, 31.6, 99.7
   Rebel-EAS : 13 ( 5, 7, 1), 65.4 : +115.5, 37.5, 99.9
   Caissa-1.13.1 : 14 ( 7, 4, 3), 64.3 : +119.0, 34.0, 100.0
   Revenge-3.0 : 14 ( 7, 7, 0), 75.0 : +141.2, 31.8, 100.0
   Igel-3.5.0 : 13 ( 6, 5, 2), 65.4 : +168.9, 29.7, 100.0

3) Berserk-11.1 3553.2 : 167 (+53,=84,-30), 56.9 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 7, 7), 25.0 : -171.4, 33.6, 0.0
   Komodo-Dragon-2.5 : 14 ( 2, 9, 3), 46.4 : -8.2, 28.0, 38.5
   Rebel-Dev : 14 ( 4, 7, 3), 53.6 : +31.6, 28.5, 86.6
   RubiChess-20230918 : 14 ( 3, 6, 5), 42.9 : +41.9, 30.4, 91.6
   Ethereal-14.00-NNUE : 14 ( 7, 4, 3), 64.3 : +62.9, 21.6, 99.8
   Koivisto-9.0 : 14 ( 6, 7, 1), 67.9 : +70.5, 27.9, 99.4
   Clover-6.0 : 14 ( 4, 9, 1), 60.7 : +72.3, 18.5, 100.0
   CST-2.0 : 13 ( 6, 7, 0), 73.1 : +79.9, 31.3, 99.5
   Rebel-EAS : 14 ( 6, 4, 4), 57.1 : +107.3, 29.0, 100.0
   Caissa-1.13.1 : 14 ( 7, 6, 1), 71.4 : +110.8, 32.6, 100.0
   Revenge-3.0 : 14 ( 2, 11, 1), 53.6 : +133.0, 23.6, 100.0
   Igel-3.5.0 : 14 ( 6, 7, 1), 67.9 : +160.7, 23.5, 100.0

4) Rebel-Dev 3521.6 : 166 (+47,=82,-37), 53.0 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 1, 5, 8), 25.0 : -203.1, 31.6, 0.0
   Komodo-Dragon-2.5 : 14 ( 5, 6, 3), 57.1 : -39.8, 28.0, 7.8
   Berserk-11.1 : 14 ( 3, 7, 4), 46.4 : -31.6, 28.5, 13.4
   RubiChess-20230918 : 14 ( 2, 7, 5), 39.3 : +10.3, 22.8, 67.4
   Ethereal-14.00-NNUE : 13 ( 3, 6, 4), 46.2 : +31.3, 23.5, 90.8
   Koivisto-9.0 : 14 ( 2, 9, 3), 46.4 : +38.8, 25.1, 93.9
   Clover-6.0 : 14 ( 4, 7, 3), 53.6 : +40.7, 24.1, 95.4
   CST-2.0 : 14 ( 5, 7, 2), 60.7 : +48.2, 22.3, 98.5
   Rebel-EAS : 14 ( 6, 8, 0), 71.4 : +75.7, 20.4, 100.0
   Caissa-1.13.1 : 14 ( 4, 9, 1), 60.7 : +79.2, 29.7, 99.6
   Revenge-3.0 : 13 ( 7, 3, 3), 65.4 : +101.4, 30.5, 100.0
   Igel-3.5.0 : 14 ( 5, 8, 1), 64.3 : +129.1, 25.4, 100.0

5) RubiChess-20230918 3511.3 : 165 (+42,=87,-36), 51.8 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 6, 8), 21.4 : -213.3, 32.5, 0.0
   Komodo-Dragon-2.5 : 14 ( 2, 5, 7), 32.1 : -50.1, 15.9, 0.1
   Berserk-11.1 : 14 ( 5, 6, 3), 57.1 : -41.9, 30.4, 8.4
   Rebel-Dev : 14 ( 5, 7, 2), 60.7 : -10.3, 22.8, 32.6
   Ethereal-14.00-NNUE : 14 ( 2, 9, 3), 46.4 : +21.0, 29.0, 76.6
   Koivisto-9.0 : 14 ( 2, 11, 1), 53.6 : +28.6, 33.8, 80.1
   Clover-6.0 : 13 ( 2, 9, 2), 50.0 : +30.4, 31.0, 83.7
   CST-2.0 : 14 ( 3, 8, 3), 50.0 : +38.0, 23.5, 94.7
   Rebel-EAS : 13 ( 4, 6, 3), 53.8 : +65.4, 33.1, 97.6
   Caissa-1.13.1 : 13 ( 6, 5, 2), 65.4 : +69.0, 35.3, 97.4
   Revenge-3.0 : 14 ( 6, 7, 1), 67.9 : +91.1, 34.0, 99.6
   Igel-3.5.0 : 14 ( 5, 8, 1), 64.3 : +118.8, 29.6, 100.0

6) Ethereal-14.00-NNUE 3490.3 : 167 (+36,=91,-40), 48.8 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 9, 5), 32.1 : -234.3, 28.5, 0.0
   Komodo-Dragon-2.5 : 14 ( 2, 6, 6), 35.7 : -71.1, 27.8, 0.5
   Berserk-11.1 : 14 ( 3, 4, 7), 35.7 : -62.9, 21.6, 0.2
   Rebel-Dev : 13 ( 4, 6, 3), 53.8 : -31.3, 23.5, 9.2
   RubiChess-20230918 : 14 ( 3, 9, 2), 53.6 : -21.0, 29.0, 23.4
   Koivisto-9.0 : 14 ( 4, 5, 5), 46.4 : +7.6, 30.1, 59.9
   Clover-6.0 : 14 ( 3, 8, 3), 50.0 : +9.4, 22.6, 66.1
   CST-2.0 : 14 ( 3, 8, 3), 50.0 : +17.0, 35.7, 68.3
   Rebel-EAS : 14 ( 3, 10, 1), 57.1 : +44.4, 32.3, 91.6
   Caissa-1.13.1 : 14 ( 2, 9, 3), 46.4 : +47.9, 25.7, 96.9
   Revenge-3.0 : 14 ( 4, 10, 0), 64.3 : +70.1, 24.4, 99.8
   Igel-3.5.0 : 14 ( 5, 7, 2), 60.7 : +97.8, 22.8, 100.0

7) Koivisto-9.0 3482.7 : 165 (+39,=81,-45), 48.2 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 6, 8), 21.4 : -241.9, 26.0, 0.0
   Komodo-Dragon-2.5 : 14 ( 1, 9, 4), 39.3 : -78.6, 33.8, 1.0
   Berserk-11.1 : 14 ( 1, 7, 6), 32.1 : -70.5, 27.9, 0.6
   Rebel-Dev : 14 ( 3, 9, 2), 53.6 : -38.8, 25.1, 6.1
   RubiChess-20230918 : 14 ( 1, 11, 2), 46.4 : -28.6, 33.8, 19.9
   Ethereal-14.00-NNUE : 14 ( 5, 5, 4), 53.6 : -7.6, 30.1, 40.1
   Clover-6.0 : 14 ( 6, 4, 4), 57.1 : +1.8, 19.2, 53.8
   CST-2.0 : 14 ( 5, 2, 7), 42.9 : +9.4, 29.2, 62.6
   Rebel-EAS : 13 ( 3, 8, 2), 53.8 : +36.8, 24.5, 93.4
   Caissa-1.13.1 : 13 ( 3, 8, 2), 53.8 : +40.4, 25.9, 94.1
   Revenge-3.0 : 13 ( 4, 7, 2), 57.7 : +62.6, 18.4, 100.0
   Igel-3.5.0 : 14 ( 7, 5, 2), 67.9 : +90.2, 19.6, 100.0

  Clover-6.0 3480.9 : 165 (+41,=76,-48), 47.9 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 5, 9), 17.9 : -243.7, 24.2, 0.0
   Komodo-Dragon-2.5 : 13 ( 3, 6, 4), 46.2 : -80.5, 27.6, 0.2
   Berserk-11.1 : 14 ( 1, 9, 4), 39.3 : -72.3, 18.5, 0.0
   Rebel-Dev : 14 ( 3, 7, 4), 46.4 : -40.7, 24.1, 4.6
   RubiChess-20230918 : 13 ( 2, 9, 2), 50.0 : -30.4, 31.0, 16.3
   Ethereal-14.00-NNUE : 14 ( 3, 8, 3), 50.0 : -9.4, 22.6, 33.9
   Koivisto-9.0 : 14 ( 4, 4, 6), 42.9 : -1.8, 19.2, 46.2
   CST-2.0 : 13 ( 4, 6, 3), 53.8 : +7.6, 31.0, 59.7
   Rebel-EAS : 14 ( 3, 8, 3), 50.0 : +35.0, 26.8, 90.4
   Caissa-1.13.1 : 14 ( 5, 6, 3), 57.1 : +38.6, 29.9, 90.1
   Revenge-3.0 : 14 ( 7, 3, 4), 60.7 : +60.8, 17.4, 100.0
   Igel-3.5.0 : 14 ( 6, 5, 3), 60.7 : +88.4, 20.5, 100.0

9) CST-2.0 3473.3 : 165 (+38,=77,-50), 46.4 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 6, 8), 21.4 : -251.3, 34.8, 0.0
   Komodo-Dragon-2.5 : 14 ( 2, 8, 4), 42.9 : -88.0, 31.6, 0.3
   Berserk-11.1 : 13 ( 0, 7, 6), 26.9 : -79.9, 31.3, 0.5
   Rebel-Dev : 14 ( 2, 7, 5), 39.3 : -48.2, 22.3, 1.5
   RubiChess-20230918 : 14 ( 3, 8, 3), 50.0 : -38.0, 23.5, 5.3
   Ethereal-14.00-NNUE : 14 ( 3, 8, 3), 50.0 : -17.0, 35.7, 31.7
   Koivisto-9.0 : 14 ( 7, 2, 5), 57.1 : -9.4, 29.2, 37.4
   Clover-6.0 : 13 ( 3, 6, 4), 46.2 : -7.6, 31.0, 40.3
   Rebel-EAS : 14 ( 6, 4, 4), 57.1 : +27.4, 28.5, 83.2
   Caissa-1.13.1 : 14 ( 2, 8, 4), 42.9 : +31.0, 42.5, 76.7
   Revenge-3.0 : 14 ( 4, 7, 3), 53.6 : +53.2, 34.4, 93.9
   Igel-3.5.0 : 13 ( 6, 6, 1), 69.2 : +80.8, 28.6, 99.8

10) Rebel-EAS 3445.9 : 165 (+28,=86,-51), 43.0 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 9, 5), 32.1 : -278.7, 33.3, 0.0
   Komodo-Dragon-2.5 : 13 ( 1, 7, 5), 34.6 : -115.5, 37.5, 0.1
   Berserk-11.1 : 14 ( 4, 4, 6), 42.9 : -107.3, 29.0, 0.0
   Rebel-Dev : 14 ( 0, 8, 6), 28.6 : -75.7, 20.4, 0.0
   RubiChess-20230918 : 13 ( 3, 6, 4), 46.2 : -65.4, 33.1, 2.4
   Ethereal-14.00-NNUE : 14 ( 1, 10, 3), 42.9 : -44.4, 32.3, 8.4
   Koivisto-9.0 : 13 ( 2, 8, 3), 46.2 : -36.8, 24.5, 6.6
   Clover-6.0 : 14 ( 3, 8, 3), 50.0 : -35.0, 26.8, 9.6
   CST-2.0 : 14 ( 4, 4, 6), 42.9 : -27.4, 28.5, 16.8
   Caissa-1.13.1 : 14 ( 5, 4, 5), 50.0 : +3.5, 29.1, 54.8
   Revenge-3.0 : 14 ( 4, 9, 1), 60.7 : +25.7, 31.3, 79.4
   Igel-3.5.0 : 14 ( 1, 9, 4), 39.3 : +53.4, 33.8, 94.3

11) Caissa-1.13.1 3442.3 : 165 (+32,=77,-56), 42.7 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 13 ( 0, 6, 7), 23.1 : -282.3, 34.9, 0.0
   Komodo-Dragon-2.5 : 14 ( 3, 4, 7), 35.7 : -119.0, 34.0, 0.0
   Berserk-11.1 : 14 ( 1, 6, 7), 28.6 : -110.8, 32.6, 0.0
   Rebel-Dev : 14 ( 1, 9, 4), 39.3 : -79.2, 29.7, 0.4
   RubiChess-20230918 : 13 ( 2, 5, 6), 34.6 : -69.0, 35.3, 2.6
   Ethereal-14.00-NNUE : 14 ( 3, 9, 2), 53.6 : -47.9, 25.7, 3.1
   Koivisto-9.0 : 13 ( 2, 8, 3), 46.2 : -40.4, 25.9, 5.9
   Clover-6.0 : 14 ( 3, 6, 5), 42.9 : -38.6, 29.9, 9.9
   CST-2.0 : 14 ( 4, 8, 2), 57.1 : -31.0, 42.5, 23.3
   Rebel-EAS : 14 ( 5, 4, 5), 50.0 : -3.5, 29.1, 45.2
   Revenge-3.0 : 14 ( 3, 7, 4), 46.4 : +22.2, 31.3, 76.1
   Igel-3.5.0 : 14 ( 5, 5, 4), 53.6 : +49.8, 28.2, 96.2

12) Revenge-3.0 3420.1 : 165 (+25,=82,-58), 40.0 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 13 ( 1, 5, 7), 26.9 : -304.5, 17.0, 0.0
   Komodo-Dragon-2.5 : 14 ( 0, 7, 7), 25.0 : -141.2, 31.8, 0.0
   Berserk-11.1 : 14 ( 1, 11, 2), 46.4 : -133.0, 23.6, 0.0
   Rebel-Dev : 13 ( 3, 3, 7), 34.6 : -101.4, 30.5, 0.0
   RubiChess-20230918 : 14 ( 1, 7, 6), 32.1 : -91.1, 34.0, 0.4
   Ethereal-14.00-NNUE : 14 ( 0, 10, 4), 35.7 : -70.1, 24.4, 0.2
   Koivisto-9.0 : 13 ( 2, 7, 4), 42.3 : -62.6, 18.4, 0.0
   Clover-6.0 : 14 ( 4, 3, 7), 39.3 : -60.8, 17.4, 0.0
   CST-2.0 : 14 ( 3, 7, 4), 46.4 : -53.2, 34.4, 6.1
   Rebel-EAS : 14 ( 1, 9, 4), 39.3 : -25.7, 31.3, 20.6
   Caissa-1.13.1 : 14 ( 4, 7, 3), 53.6 : -22.2, 31.3, 23.9
   Igel-3.5.0 : 14 ( 5, 6, 3), 57.1 : +27.7, 24.2, 87.3

13) Igel-3.5.0 3392.5 : 166 (+24,=75,-67), 37.0 %

   vs. : games ( +, =, -), (%) : Diff, SD, CFS (%)
   SF16 : 14 ( 0, 4, 10), 14.3 : -332.1, 31.0, 0.0
   Komodo-Dragon-2.5 : 13 ( 2, 5, 6), 34.6 : -168.9, 29.7, 0.0
   Berserk-11.1 : 14 ( 1, 7, 6), 32.1 : -160.7, 23.5, 0.0
   Rebel-Dev : 14 ( 1, 8, 5), 35.7 : -129.1, 25.4, 0.0
   RubiChess-20230918 : 14 ( 1, 8, 5), 35.7 : -118.8, 29.6, 0.0
   Ethereal-14.00-NNUE : 14 ( 2, 7, 5), 39.3 : -97.8, 22.8, 0.0
   Koivisto-9.0 : 14 ( 2, 5, 7), 32.1 : -90.2, 19.6, 0.0
   Clover-6.0 : 14 ( 3, 5, 6), 39.3 : -88.4, 20.5, 0.0
   CST-2.0 : 13 ( 1, 6, 6), 30.8 : -80.8, 28.6, 0.2
   Rebel-EAS : 14 ( 4, 9, 1), 60.7 : -53.4, 33.8, 5.7
   Caissa-1.13.1 : 14 ( 4, 5, 5), 46.4 : -49.8, 28.2, 3.8
   Revenge-3.0 : 14 ( 3, 6, 5), 42.9 : -27.7, 24.2, 12.7

2 SF losses only, Rebel-Dev and Revenge.

Komodo 31 losses.

Posts : 159 Join date : 2022-03-01 Location : Berlin

Admin wrote:: Tried your gamepair tool, is it possible it doesn't work with FEN pgns?

No, sadly not. The Tool needs pgn from move 1 and (important) all used openings have to have the same length (like my UHO-Openings-sets - in theses sets, each line has the same length), which is not the case in the TCEC openings for example...

From the ReadMe:
The tool works only, when:
1) The pgn-gamefile contains only games with not more than 995 different head-to-head engine-competitions
2) Each opening is played exactly 2 times in each head-to-head (each engine plays it with white and black)
3) Each opening-line has the same length/depth (in plies)
4) The files pgn-extract.exe, textReplace.exe, ordo-win64.exe, pairSplit.exe and nameList.exe are in the
bin folder, which has to be in the working-folder - they are needed for processing...KUDOS to the authors
and (C): D.Barnes, T.Zipproth, M.Ballicora, N.Pollock

In my tool (Auto_Gamepairs_Rescorer_V1.bat), you have to set the 4 parameters at the top (just open the
file with any texteditor):
For example:
set opening_plies=12 (Each opening-line is 6 moves (=12 plies) deep)
set games=testmatch.pgn (name of the pgn-gamefile)
set reference_eng="Stockfish 14 avx2" (name of the engine, Ordo shall use for reference)
set /A reference_elo=3700 (Elo-number, which Ordo shall use for the reference-engine)

Subject: Re: New Rating List ? Sat Oct 07, 2023 4:54 pm

Yep, I suspected that, tried set opening_plies=0 to no avail.

But the head-to-head is part of ORDO, I am using -

ordo-win64 -P pgn.txt -W -U "0,1,2,3,4,5,6,7,8,9,10" -o output.txt -a 3500 -C -J -s10 -V -D -j head-to-head.txt
start ouput.txt

No anchors needed doing an initial robin-round.

» Speedy rating list
» Rating List Experiment
» First Gambit Rating List
» SSDF Rating List 23-05-24
» Gambit Rating List - May 20, 2021