Разные матчи и турниры движков

[Владимир001]

Raubfisch X48c2 wins Fritz 15 New Engines Tournament, 2022.01.19-24

Time 15' + 10" computer: Lenovo Z710 i7-4710MG 8GB RAM, GUI Fritz 15
182 tournament games download, All games CEDR:13.7454 games

[Владимир001]

CONGRATULATIONS TO THE WINNER!
The New NimaTiv PRO CTG Is Very Strong And Well Tuned!!!
Vitamin CTG Is An Old And Strong Book!

Settings Used For The Tourney:
GUI Used= Fritz
Engine Used= SF PolyGlot (Provided To Mee By Massimiliano Goi, Thank You Friend)(Thanking To The Author Too!)
Time Control= 3min+1sec increment
Permanent Brain= OFF
Threads Used Per Engine= 2
Hash= 128mb

Games Played In This Tourney:
https://pixeldrain.com/u/V741dWAo

[Владимир001]

Hardware: i5-8400 CPU @ 2.80GHz, 6 cores, 32GB, gpu RTX 2060, 6GB
Gauntlet: SF vs. LC0 192x15b nets
Network IDs: 753723, 761352, 771721
LC0-version: v0.28.2
LC0-options: minibatch-size=48, default or zz4032 tunes https://discordapp.com/channels/4254...05545290526770
SF-version: 22010217
SF-options: 1 Thread, 1024MB Hash
Software: Cutechess-CLI
Time control: 6"+0.1"
Book: book-ply8-unifen-Q-0.0-0.25 (jhorthos), order=sequential
Tablebases: 6-men
Adjudication: -draw movenumber=35 movecount=5 score=8 -resign movecount=5 score=650
Ordo: -a 0 -A "Stockfish.dev" -W -D -U "0,1,2,3,4,5,6,7,8,9,10" -s 2000 -n 6
Comment: The aim of the test is to compare T75, T76 and T77 with best tune. The order is considered sure with 99% CFS. CFS matrix and detailed results in 7z file.

   # PLAYER               :  RATING  ERROR   POINTS  PLAYED   (%)  CFS(%)      W      D     L  D(%)
   1 Stockfish.dev        :     0.0   ----  45496.5   80000    57     100  17766  55461  6773    69
-  2 lc0.net.753723       :   -43.8    2.6   8771.0   20000    44      98   1764  14014  4222    70
   3 lc0.net.761352       :   -47.5    2.6   8667.5   20000    43      89   1757  13821  4422    69
   4 lc0.net.771721       :   -49.8    2.6   8603.5   20000    43     100   1652  13903  4445    70
+  5 lc0.net.753723_zz    :   -55.0    2.6   8461.5   20000    42     ---   1600  13723  4677    69

White advantage = 37.34 +/- 0.68
Draw rate (equal opponents) = 73.06 % +/- 0.17

PGN:

https://pixeldrain.com/u/RUNsnU9V

[Владимир001]

Test: Hanse Lc0 A/B Helper Engine vs Lc0
Hardware: GPU RTX 3060; CPU AMD 3960X - underclocked @3.8 Mhz, 1.15 volts
System: Windows 11 Pro Insider Preview
Net: T60_611062
Tournament: Gauntlet vs. Lc0
Book: UHOXXL+0.90_+1.19.epd
Time control: 200” + 1”; legacy time management
Game Duration: mean = 6.9 minutes; median = 7.0 minutes
Tablebases: syzygy 6-man
Adjudication: -draw movenumber=34 movecount=4 score=20 -resign movecount=3 score=400
Software: cutechess-cli 1.2.0
lc0 A/B Helper Engine Parameters:
GIT branch: “helper_and_q_based_move” through commit 12f1e5c (1/23/2022)
See: https://github.com/hans-ekbrand/lc0/...-helper-engine

--backend=cuda-fp16
--backend-opts=multi_stream=true
--auxengine-file=stockfish.exe (latest dev version from Git; native, COMP=clang)
--auxengine-options=Hash=512;Threads=1;Ponder=off
--auxengine-instances=15
--auxengine-options-onroot="Hash=2048;Threads=22"
--two-fold-draws=false
--solid-tree-threshold=2000000000
--q-based-move-selection=true
--move-selection-visits-scaling-power=0.5
--auxengine-max-depth=2
--auxengine-threshold=100
--auxengine-time=800

Comment: Hanse continues to enhance the A/B helper approach, but even now the combination of root node and tree node helper engines appears quite strong relative to Lc0.

   # PLAYER    :  RATING  ERROR  POINTS  PLAYED   (%)  CFS(%)    W    D    L  D(%)
   1 lcABhelper:    51.7   39.6    93.5     170    55      99   56   75   39    44
   2 lc0       :     0.0   ----    76.5     170    45     ---   39   75   56    44

White advantage = 225.47 +/- 20.20
Draw rate (equal opponents) = 100.00 % +/- 1.67

[Владимир001]

Capablanca - Nimzowitsch match simulation

https://www.chess2u.com/t17916-capab...tch-simulation

[Владимир001]

Slowchess 2.8 wins CEDR - Fritz 17 Tournament, 2022.01.23-26

Time 15'+10" HP ENVY i-5 i5-8250U 8GB RAM GUI Fritz 17
90 games P-15'+10" games CEDR 13.7454 games Table: Scid vs PC

[Владимир001]

policy_tournament: 780001 vs 780048
LC0: lc0_selfplay (The LCZero Authors), --backend=cuda-fp16
Hardware: RTX 2060
Software: multiordo (Tilps)
Time control: 1 nodes/move
Book: book-ply8-unifen-Q-0.0-0.25 (82713 openings)
Comment: without tablebases, cutechess-cli.

policy_tournament.exe selfplay --player1.weights=780001
--player2.weights=780048
--openings-pgn=book-ply8-unifen-Q-0.0-0.25.pgn
--policy-mode-size=64 --mirror-openings=true --tournament
-results-file=output.pgn --visits=1 --games=-1 --parallelism=48
--backend=multiplexing --backend-opts="backend=cuda-fp16"

 # PLAYER : RATING  ERROR    POINTS  PLAYED       W       L       D  D(%)  CFS(%)
-1 780001 :   0.00   ----  167258.5  330852  105287  101622  123943  37.5     100
 2 780800 :  -3.18   1.50   81964.0  165426   50993   52491   61942  37.4      91
+3 780048 :  -4.60   1.46   81629.5  165426   50629   52796   62001  37.5     ---

White advantage = 13.77 +/- 1.47
Draw rate (equal opponents) = 37.51 % +/- 0.40

[Владимир001]

policy_tournament: 512x40-s2-swa-200000 vs 512x40-s2-swa-400000 / 512x40-s2-swa-600000
LC0: policy_tournament.exe, --backend=cuda-fp16
Hardware: RTX 2060
Software: multiordo (Tilps)
Time control: 1 nodes/move
Book: book-ply8-unifen-Q-0.0-0.25.pgn
Tablebases: Syzygy 5-men
Comment: without cutechess-cli

# PLAYER               :  RATING  ERROR   POINTS  PLAYED      W      L      D  D(%)  CFS(%)
1 512x40-s2-swa-600000 :  222.23   3.03  39009.5   50000  32450   4431  13119  26.2     100
2 512x40-s2-swa-400000 :   18.48   2.64  26315.5   50000  16651  14020  19329  38.7     100
3 512x40-s2-swa-200000 :    0.00   ----  34675.0  100000  18451  49101  32448  32.4     ---

White advantage = 12.22 +/- 1.76
Draw rate (equal opponents) = 37.49 % +/- 0.47

https://pixeldrain.com/l/1mRD8WkK

[Владимир001]

policy_tournament: 780001 vs 512x40-s2-swa-600000
LC0: policy_tournament.exe, --backend=cuda-fp16
Hardware: RTX 2060
Software: multiordo (Tilps)
Time control: 1 nodes/move
Book: book-ply8-unifen-Q-0.0-0.25.pgn
Tablebases: Syzygy 5-men
Comment: without cutechess-cli

# PLAYER               :  RATING  ERROR   POINTS  PLAYED      W      L      D  D(%)  CFS(%)
1 780001               :    0.00   ----  41443.5   50000  36166   3279  10555  21.1     100
2 512x40-s2-swa-600000 : -277.21   3.47   8556.5   50000   3279  36166  10555  21.1     ---

White advantage = 19.90 +/- 2.27

[Владимир001]

policy_tournament: 780001 vs 780856
LC0: policy_tournament.exe, --backend=cuda-fp16
Hardware: RTX 2060
Software: multiordo (Tilps)
Time control: 1 nodes/move
Book: openings-10ply-100k.pgn
Tablebases: Syzygy 5-men
Comment: tournamentstatus P1: +30992 -30690 =38318 Win: 50.15% Elo: 1.05 LOS: 88.80% P1-W: +17253 -13908 =18839
P1-B: +13739 -16782 =19479 npm -nan(ind) nodes 0 moves 0

# PLAYER : RATING  ERROR   POINTS  PLAYED      W      L      D  D(%)  CFS(%)
1 780001 :   0.00   ----  50151.0  100000  30992  30690  38318  38.3      86
2 780856 :  -1.06   1.93  49849.0  100000  30690  30992  38318  38.3     ---

White advantage = 22.42 +/- 1.72
Draw rate (equal opponents) = 38.45 % +/- 0.44

[Владимир001]

Match: pr1686 (new a0 tm) vs master (original a0 tm)
SF version: stockfish_21122716_x64_bmi2
LC0 version (both): local
SF options: defaults
LC0 options (both): threads=1, minibatch-size=48
LC0 options (master): alphazero(alphazero-time-pct=23.0629), smart-pruning-factor=3.6141
LC0 options (pr1686): alphazero(min-pct=2.52,time-mult=2.17,ply-mult=2.72)
SF Net: nn-ac07bd334b62
LC0 Net: 744501
Book: custom-6k (1-3 ply)
Hardware: gtx970 + i5-4570
Tablebases: syzygy 5-man
Adjudication: -draw movenumber=40 movecount=2 score=10 -resign movecount=1 score=500 twosided=true
Software: cutechess-cli
Comment: Reworked the formula in pr1686 and tuned it vs sf @ vstc. Ran tests vs original a0 tm using multiple types of time control (no-increment, increment-only, vstc @ standard ratio, ltc @ sufi ratio). The no-increment match was stopped early because the original a0 tm did so bad. Despite the good results, Im marking this as draft until I can do more testing/tuning. The min-pct is too low IMO, and it could be because I use a 3-ply opening book which may require less thinking time on the first move than a longer opening book would.

 6s+0.1s
   # PLAYER        :  RATING  ERROR  POINTS  PLAYED   (%)  CFS(%)    W    D    L  D(%)
   1 sf            :     0.0   ----   794.0    1500    53      86  382  824  294    55
   2 lc0-pr1686    :    -9.4   16.8   365.0     750    49      97  160  410  180    55
   3 lc0-master    :   -31.9   16.8   341.0     750    45     ---  134  414  202    55


0s+0.2s
   # PLAYER        :  RATING  ERROR  POINTS  PLAYED   (%)  CFS(%)    W    D    L  D(%)
   1 lc0-pr1686    :    18.4   15.3   263.0     500    53      99   81  364   55    73
   2 lc0-master    :     0.0   ----   237.0     500    47     ---   55  364   81    73


12s+0s
   # PLAYER        :  RATING  ERROR  POINTS  PLAYED   (%)  CFS(%)    W    D    L  D(%)
   1 lc0-pr1686    :   163.6   33.9   143.5     200    72     100   94   99    7    50
   2 lc0-master    :     0.0   ----    56.5     200    28     ---    7   99   94    50


144s+0.2s
   # PLAYER        :  RATING  ERROR  POINTS  PLAYED   (%)  CFS(%)    W    D    L  D(%)
   1 lc0-pr1686    :    15.5   15.1   130.5     250    52      98   21  219   10    88
   2 lc0-master    :     0.0   ----   119.5     250    48     ---   10  219   21    88

[Владимир001]

CONGRATULATIONS TO THE WINNER!!!
NimaTiv PRO Is A Strong And PowerFul CTG Book!
And Mustafa Is A Strong Competitor!

Settings Used For The Tourney:
GUI Used= Fritz
Engine Used= SF PolyGlot (Provided To Mee By Massimiliano Goi, Thank You Friend)(Thanking To The Author Too!)
Time Control= 3min+1sec increment
Permanent Brain= OFF
Threads Used Per Engine= 2
Hash= 128mb

Games Played In This Tourney:
https://pixeldrain.com/u/uv9ZkNx1

[Владимир001]

Match: 780001 vs 780856
LC0 version: v0.28
LC0 options: --smart-pruning-factor=1 --minibatch-size=64 --cpuct=1.473 --cpuct-factor=3.973 --cpuct-base=45669 --fpu-value=0.279 --policy-softmax-temp=1.206
Hardware: EPYC 74F3 + A100
Time control: 1k npm
Book: book-ply8-unifen-Q-0.25-0.40.pgn
Tablebases: no
Adjudication: -resign movecount=2 score=500 twosided=true -draw movenumber=50 movecount=20 score=10
Software: cutechess-cli
Comment:

   # PLAYER        :  RATING  ERROR  POINTS  PLAYED   (%)  CFS(%)     W     D     L  D(%)
   1 Lc0.780856    :    11.1    4.0  7192.0   14000    51     100  3597  7190  3213    51
   2 Lc0.780001    :     0.0   ----  6808.0   14000    49     ---  3213  7190  3597    51

White advantage = 132.91 +/- 2.02
Draw rate (equal opponents) = 61.30 % +/- 0.51

https://pixeldrain.com/u/BWfqoM1q

[Владимир001]

Fisherov 0.98i wins CEDR - Fritz 17 Tournament Chess Engines Diary, 2022.01.18-22

Time 15'+10" HP Pavilion i5-1035G1 8GB RAM GUI Fritz 17
132 games P-15'+10" games CEDR 13.7454 games Table: Scid vs PC

[Владимир001]

Blue Marlin 14.7 wins Banksia Tournament - 10-27.01.2022

Time 15' + 10" computer: Lenovo Ideapad 330 i3-8130U 12GB RAM GUI Banksia
182 tournament games download, All games CEDR:13.7454 games,

[Владимир001]

Test: Hanse Lc0 A/B w/Berserk as helper Vs. Lc0
Hardware: GPU RTX 3060; CPU AMD 3960X - underclocked @3.8 Mhz, 1.15 volts

Развернуть для просмотра

System: Windows 11 Pro Insider Preview
Net: T60_611062
Tournament: Gauntlet vs. Lc0
Book: UHOXXL+0.90_+1.19.epd
Time control: 200” + 1”; legacy time management
Game Duration: mean = 7.2 minutes; median = 7.5 minutes
Tablebases: syzygy 6-man
Adjudication: -draw movenumber=34 movecount=4 score=20 -resign movecount=3 score=400
Software: cutechess-cli 1.2.0
lc0 A/B Helper Engine Parameters:
GIT branch: “helper_and_q_based_move” through commit 9a53066 (1/26/2022)
See: https://github.com/hans-ekbrand/lc0/...-helper-engine

--backend=cuda-fp16
--backend-opts=multi_stream=true
--auxengine-file=berserk.exe (compiled from latest Git code)
--auxengine-options=Hash=512;Threads=1;Ponder=off
--auxengine-instances=15
--auxengine-options-onroot="Hash=2048;Threads=22"
--two-fold-draws=false
--solid-tree-threshold=2000000000
--q-based-move-selection=true
--move-selection-visits-scaling-power=0.5
--auxengine-max-depth=2
--auxengine-threshold=100
--auxengine-time=800

Comment: Berserk as helper engine not as strong as similar test with Stockfish as helper (https://discordapp.com/channels/4254...61560832667649)

[свернуть]

   # PLAYER             :  RATING  ERROR  POINTS  PLAYED   (%)  CFS(%)    W    D    L  D(%)
   1 lcab w/berserk     :    26.2   36.7    84.5     160    53      92   44   81   35    51
   2 lc0                :     0.0   ----    75.5     160    47     ---   35   81   44    51

White advantage = 190.58 +/- 18.54
Draw rate (equal opponents) = 100.00 % +/- 1.14

[Владимир001]

91ST AMATEUR SERIES (Division 3)

Intel i7 Quad
ChessGUI
256mb hash each
3-4-5 piece tablebases
Ponder off
GM2700+.cgb book
40 moves in 16 minutes repeating (adapted for the CCRL)
4 cycles 44 rounds
All engines 64-bit 1CPU (where available)

Цитата:

The top engine will gain automatic promotion and the bottom engine will gain automatic relegation. Any other changes of division will be at my discretion and may not necessarily involve playoff matches.

Final Standings

29.0 - Combusken 2.0.0 64-bit
27.5 - Roc 1.8 64-bit
26.5 - Stash 32.0 64-bit
23.0 - Dark Toga 1.1 64-bit
22.5 - Halogen 10 64-bit
22.0 - Invictus r382 64-bit
21.0 - Mr Bob 1.1.0 64-bit
20.0 - Velvet 3.1.0 64-bit
19.0 - Demolito 2021-10-04 64-bit
18.5 - chess22k 1.14 64-bit
18.5 - Orion 0.8 64-bit
16.5 - Nirvanachess 2.5 64-bit


The complete tournament pgn (zipped) can be downloaded here:
http://kirill-kryukov.com/chess/disc...e.php?id=49895

[Владимир001]

91ST AMATEUR SERIES (Division 4)

Intel i7 Quad
ChessGUI
256mb hash each
3-4-5 piece tablebases
Ponder off
GM2700+.cgb book
40 moves in 16 minutes repeating (adapted for the CCRL)
4 cycles 44 rounds
All engines 64-bit 1CPU (where available)

Цитата:

The top engine will gain automatic promotion and the bottom engine will gain automatic relegation. Any other changes of division will be at my discretion and may not necessarily involve playoff matches.

Final Standings

27.5 - Mantissa 3.3.0 64-bit
27.5 - Counter 4.0 64-bit
26.0 - Black Marlin 4.0 64-bit
24.5 - Topple 0.8.1 64-bit
23.0 - Drofa 3.2.0 64-bit
21.5 - Pirarucu 3.3.5 64-bit
21.5 - Cheng 4.41 64-bit
20.5 - Bit-Genie 9.0 64-bit
19.5 - Amoeba 3.4 64-bit
18.5 - FabChess 1.16 64-bit
18.0 - Monolith 2.0 64-bit
16.0 - Bagatur 3.0 64-bit


The complete tournament pgn (zipped) can be downloaded here:
http://kirill-kryukov.com/chess/disc...e.php?id=49896

[svoitsl]

Цитата: Сообщение от Viktor

Где скачать сеть 512x25-s2-swa-400000? Хотел потестировать.

можете ещё и эти потестировать
512x25-s2-swa-600000.pb.gz
512x40-s2-swa-600000.pb.gz

[Владимир001]

Capablanca - Nimzowitsch match simulation

https://www.chess2u.com/t17916-capab...ulation#113051

[Владимир001]

Zeus 22.5 AVX2 vs. Stockfish Polyglot 260122 x64,AVX2

Processor AMD A10 – 9700 R7,10 Compute Cores 4C + 6G - 3.50 GHz - RAM 8.00 GB - 64-bit operating system - x64 processor - Windows 10 Pro - 1024 mb hash – Fritz 16 GUI – Permanent brain: ON – opening book: BiyuSuite 8 Equallium 2022 ctg - final tablebases: Syzygy 3-4-5 pieces - both engines play with 2 threads and large pages - NNUE: Zeus 22.5 AVX2: nn.zeus – Stockfish Polyglot: nn-ac07bd334b62.nnue

Book Settings:

1.Tournament ON
2.Min games 1-8
3.Variety of play 300
4.Influence of learn 100
5.Learning strength 500

DESKTOP-TORM2PL, Blitz 5min 0

1 Stockfish Polyglot 260122 x64,AVX2 +4 +14/=136/-12 50.62% 82.0/162
2 Zeus 22.5 -4 +12/=136/-14 49.38% 80.0/162

Author: Olyprog frm Outskirt chess forum.

PGN GAMES

https://www.mediafire.com/file/3rjcb...-1421.pgn/file

[Владимир001]

Tucano 10.00 wins CEDR - Test NNUE Light Engines, 2022.01.20-23

Time 15'+10" HP Pavilion i5-1035G1 8GB RAM GUI Fritz 17
112 games P-15'+10" games CEDR 13.7454 games Table: Scid vs PC

[Владимир001]

Wasp 5.20 second compiles wins Banksia Crazy Tournament, 28.01.2022

Level - Depth 10 :computer: Dell G15 SE, i7-11800H @ 2.30GHz RAM 16,0 GB, GUI Banksia

[Владимир001]

policy_tournament: 512x25-s2-swa-400000 vs 512x25-s2-swa-600000
LC0: policy_tournament.exe, --backend=cuda-fp16
Hardware: RTX 2060
Software: multiordo (Tilps)
Time control: 1 nodes/move
Book: book-ply8-unifen-Q-0.0-0.25.pgn
Tablebases: Syzygy 5-men
Comment:

tournamentstatus final P1: +5739 -30959 =13302 Win: 24.78% Elo: -192.89 LOS:  0.00% P1-W: +3070 -14929 =7001 P1-B: +2669 -16030 =6301 npm -nan(ind) nodes 0 moves 0

# PLAYER               :  RATING  ERROR   POINTS  PLAYED      W      L      D  D(%)  CFS(%)
1 512x25-s2-swa-600000 :  194.87   2.96  37610.0   50000  30959   5739  13302  26.6     100
2 512x25-s2-swa-400000 :    0.00   ----  12390.0   50000   5739  30959  13302  26.6     ---

White advantage = 14.13 +/- 2.05
Draw rate (equal opponents) = 33.33 % +/- 0.49

[Владимир001]

Hardware: AMD Ryzen 9 5900X CPU @ 3.70GHz, 12 cores, 64GB, gpu RTX 3080, 10GB
Gauntlet: SF-dev vs. LC0 512x40b nets
Network IDs: 780001, 780048
LC0-version: v0.28.2
LC0-options: default
SF-version: SF-dev 22011322
SF-options: 8 Threads, 1024MB Hash
Software: Cutechess-CLI
Time control: LC0: 1,500 nodes/move, SF-dev: 500,000 nodes/move
Book: Chad's 8-ply book, book-ply8-unifen-Q-0.0-0.25.pgn and book-ply8-unifen-Q-0.25-0.40.pgn (10K openings in each book), order=sequential
Tablebases: 6-men
Adjudication: -draw movenumber=50 movecount=5 score=8 -resign movecount=5 score=1000
Ordo: -a 0 -A "Stockfish.dev" -W -D -U "0,1,2,3,4,5,6,7,8,9,10" -s 2000 -n 1
Comment: The aim of the test is to order T78 nets. The order is considered sure with 99% CFS (a bit more than 3.5 Elo).

 # PLAYER            :  RATING  ERROR   POINTS  PLAYED   (%)  CFS(%)      W      D      L  D(%)
   1 Stockfish.dev     :     0.0   ----  65590.0  120000    55     100  30773  69634  19593    58
+  2 lc0.net.780048    :   -32.5    1.8  27360.5   60000    46     100   9737  35247  15016    59
   3 lc0.net.780001    :   -36.3    1.8  27049.5   60000    45     ---   9856  34387  15757    57

White advantage = 78.45 +/- 0.64
Draw rate (equal opponents) = 62.57 % +/- 0.15

https://pixeldrain.com/u/G8TxcwAh

[Владимир001]

policy_tournament: 780001 vs 512x25-s2-swa-600000
LC0: policy_tournament.exe, --backend=cuda-fp16
Hardware: RTX 2060
Software: multiordo (Tilps)
Time control: 1 nodes/move
Book: book-ply8-unifen-Q-0.0-0.25.pgn
Tablebases: Syzygy 5-men
Comment:

tournamentstatus final P1: +37119 -3071 =9810 Win: 84.05% Elo: 288.68 LOS: 100.00% P1-W: +19076 -1376 =4548 P1-B: +18043 -1695 =5262 npm -nan(ind) nodes 0 moves 0

# PLAYER               :  RATING  ERROR   POINTS  PLAYED      W      L      D  D(%)  CFS(%)
1 780001               :    0.00   ----  83467.5  100000  73285   6350  20365  20.4     100
2 512x40-s2-swa-600000 : -277.13   3.41   8556.5   50000   3279  36166  10555  21.1     100
3 512x25-s2-swa-600000 : -291.91   3.57   7976.0   50000   3071  37119   9810  19.6     ---

White advantage = 18.79 +/- 1.89
Draw rate (equal opponents) = 32.16 % +/- 0.52

[Владимир001]

PGN - http://kirill-kryukov.com/chess/disc...e.php?id=49927

Blunder 7.6.0 64-bit 2640

[Владимир001]

PGN - http://kirill-kryukov.com/chess/disc...e.php?id=49929

[Владимир001]

PGN - http://kirill-kryukov.com/chess/disc...e.php?id=49928

Развернуть для просмотра

CCRL 40/15 Rating List - Custom engine selection
1331708 games played by 3005 programs, run by 25 testers
Ponder off, General books (up to 12 moves), 3-4-5 piece EGTB
Time control: Equivalent to 40 moves in 15 minutes on an Intel i7-4770k.
Computed on January 21, 2022 with Bayeselo based on 1'331'708 games
Tested by CCRL team, 2005-2022, http://ccrl.chessdom.com/ccrl/4040/

Rank                Engine                  Elo  +    -  Score  AvOp  Games
1 Dragon by Komodo 2.6 64-bit            3496  +17  -17  65.8%  -96.2  966
  Dragon by Komodo 2.5 64-bit            3486  +15  -15  66.0%  -96.8  1348
  Dragon by Komodo 64-bit                3461  +12  -12  67.5% -113.1  2199
  Dragon by Komodo 2 64-bit              3456  +16  -16  66.1%  -97.6  1097

[свернуть]

[Владимир001]

PGN - http://kirill-kryukov.com/chess/disc...e.php?id=49930

Развернуть для просмотра

CCRL 40/15 Rating List - Custom engine selection
1331708 games played by 3005 programs, run by 25 testers
Ponder off, General books (up to 12 moves), 3-4-5 piece EGTB
Time control: Equivalent to 40 moves in 15 minutes on an Intel i7-4770k.
Computed on January 21, 2022 with Bayeselo based on 1'331'708 games
Tested by CCRL team, 2005-2022, http://ccrl.chessdom.com/ccrl/4040/

Rank                Engine                  Elo  +    -  Score  AvOp  Games
1 Fritz 17 64-bit                        3189  +16  -16  47.2%  +17.6  1104
  Fritz 18 64-bit                        3185  +23  -23  47.4%  +15.7  524
  Fritz 16 64-bit                        3099  +11  -11  45.6%  +33.8  2998
  Fritz 15 64-bit                        3091  +9  -9  45.3%  +33.2  3749
  Deep Fritz 14 64-bit 1CPU              3016  +11  -11  47.2%  +18.3  2672
  Fritz 13                                2958  +11  -11  45.7%  +26.1  2532
  Fritz 12                                2885  +11  -11  50.8%  -4.6  2838
  Fritz 11                                2851  +10  -10  53.3%  -21.6  3263
  Fritz 10                                2778  +15  -15  52.6%  -18.2  1442

[свернуть]