CCRL 40/40, 40/4 and FRC lists updated (21st May 2017)

[Graham Banks]

The latest CCRL Rating Lists and Statistics are available for viewing from the following links:
http://computerchess.org.uk/ccrl/4040/ (40/40)
http://www.computerchess.org.uk/ccrl/404/ (40/4)
http://www.computerchess.org.uk/ccrl/404FRC/ (FRC 40/4)

Please note that the three lists are often updated separately to each other. The FRC list is only updated when a new engine or engine version is being/has been tested.

40/40 testing this week that I'm aware of will include (with live broadcast port for TLCV noted where applicable):

64th Amateur Series Division 4 Tournament (finishing Wednesday 16002)
64th Amateur Series Division 5 Tournament (continuing 16001)
64th Amateur Series Division 6 Tournament (starting Thursday 16002)
Burnt And Shredded Tournament (continuing 16053)
Laser 1.4 64-bit Gauntlet (continuing 16063)
Jumbo 0.4.34 64-bit Gauntlet (continuing 16064)
Zurichess Luzern 64-bit Gauntlet (continuing 16066)
Booot 6.2 64-bit Gauntlet (continuing)
Andscacs 0.91 64-bit Gauntlet (continuing)
Arasan 20.1 64-bit Gauntlet (continuing)
Wasp 2.01 64-bit Gauntlet (continuing)
Various 1CPU Tournaments (being run by Charles)
Various 4CPU Gauntlets (being run by Charles)
Komodo 10.4 64-bit 4CPU Gauntlet (continuing 16083)

40/4 testing since the last update has included:

Fischerle 0.9.80 SE 64-bit
Embla 0.9.9 64-bit
Schooner 1.7.0 64-bit
Booot 6.2 64-bit
Laser 1.4 64-bit
4CPU Tournament (continuing 16084)

[Nordlandia]

What is the exact reason why CCRL avoid stockfish dev versions?

[CheckersGuy]

What is the exact reason why CCRL avoid stockfish dev versions?

There is amost a new stockfish dev version every week. Wouldn`t really make sense to me to test a new dev version every week. Just wait for stockfish 9 in the (hopefully) near future and we will see how much elo the last few patches gained.

[Norm Pollock]

[Event "CCRL 40/40"]
[Site "CCRL"]
[Date "2015.12.23"]
[Round "490.1.931"]
[White "Ecce 1.0 (rev. 508) 64-bit"]
[Black "Clarabit 1.00 64-bit"]
[Result "1/2-1/2"]
[ECO "B10"]
[Opening "Caro-Kann defence"]
[PlyCount "137"]
[WhiteElo "2049"]
[BlackElo "2101"]

1. e4 c6 2. Ne2 d5 3. e5 d4 4. c3 c5 5. Qa4+ Nc6 6. cxd4 cxd4 7. Na3 e6 8. Nb5
Bd7 9. Nexd4 a6 10. Nc3 Nxe5 11. Qc2 Qh4 12. Qe4 Qxe4+ 13. Nxe4 O-O-O 14. Nc2
Bc6 15. Ng5 h6 16. d4 hxg5 17. dxe5 Rd5 18. Bxg5 Rxe5+ 19. Be3 Nf6 20. O-O-O
Bc5 21. Bxc5 Rxc5 22. f3 Nd5 23. Kb1 Ba4 24. Bd3 a5 25. Rd2 Nf4 26. b3 Rd5 27.
bxa4 Rxd3 28. Rxd3 Nxd3 29. Ne3 Rh4 30. Rd1 Ne5 31. Rc1+ Kb8 32. Rc5 f6 33.
Rxa5 Rxh2 34. Rb5 Rh1+ 35. Kc2 Ra1 36. Rb2 Re1 37. Rb3 Re2+ 38. Kb1 g5 39. a5
Ka7 40. Ra3 f5 41. Nc2 Rxg2 42. Nd4 g4 43. fxg4 fxg4 44. Re3 Nc4 45. Rxe6 Na3+
46. Ka1 Rd2 47. Nb3 Rf2 48. Rg6 Rg2 49. Nd4 g3 50. Rg5 Ka6 51. Nb3 Nb5 52. Nc5+
Ka7 53. Na4 Na3 54. Nc3 Ka6 55. Nd5 Nc2+ 56. Kb1 Ka7 57. Nc3 Nd4 58. Rg6 Nc6
59. a3 Nd4 60. Rg8 Rg1+ 61. Kb2 b5 62. Ne4 Nf5 63. Rg5 g2 64. Nc5 Ne3 65. Rg7+
Kb8 66. Nd7+ Kb7 67. Ne5+ Ka6 68. Nc6 Rh1 69. Ra7# 1/2-1/2

[Graham Banks]

[Event "CCRL 40/40"]
[Site "CCRL"]
[Date "2015.12.23"]
[Round "490.1.931"]
[White "Ecce 1.0 (rev. 508) 64-bit"]
[Black "Clarabit 1.00 64-bit"]
[Result "1/2-1/2"]
[ECO "B10"]
[Opening "Caro-Kann defence"]
[PlyCount "137"]
[WhiteElo "2049"]
[BlackElo "2101"]

1. e4 c6 2. Ne2 d5 3. e5 d4 4. c3 c5 5. Qa4+ Nc6 6. cxd4 cxd4 7. Na3 e6 8. Nb5
Bd7 9. Nexd4 a6 10. Nc3 Nxe5 11. Qc2 Qh4 12. Qe4 Qxe4+ 13. Nxe4 O-O-O 14. Nc2
Bc6 15. Ng5 h6 16. d4 hxg5 17. dxe5 Rd5 18. Bxg5 Rxe5+ 19. Be3 Nf6 20. O-O-O
Bc5 21. Bxc5 Rxc5 22. f3 Nd5 23. Kb1 Ba4 24. Bd3 a5 25. Rd2 Nf4 26. b3 Rd5 27.
bxa4 Rxd3 28. Rxd3 Nxd3 29. Ne3 Rh4 30. Rd1 Ne5 31. Rc1+ Kb8 32. Rc5 f6 33.
Rxa5 Rxh2 34. Rb5 Rh1+ 35. Kc2 Ra1 36. Rb2 Re1 37. Rb3 Re2+ 38. Kb1 g5 39. a5
Ka7 40. Ra3 f5 41. Nc2 Rxg2 42. Nd4 g4 43. fxg4 fxg4 44. Re3 Nc4 45. Rxe6 Na3+
46. Ka1 Rd2 47. Nb3 Rf2 48. Rg6 Rg2 49. Nd4 g3 50. Rg5 Ka6 51. Nb3 Nb5 52. Nc5+
Ka7 53. Na4 Na3 54. Nc3 Ka6 55. Nd5 Nc2+ 56. Kb1 Ka7 57. Nc3 Nd4 58. Rg6 Nc6
59. a3 Nd4 60. Rg8 Rg1+ 61. Kb2 b5 62. Ne4 Nf5 63. Rg5 g2 64. Nc5 Ne3 65. Rg7+
Kb8 66. Nd7+ Kb7 67. Ne5+ Ka6 68. Nc6 Rh1 69. Ra7# 1/2-1/2

Thanks Norm.

[Sven]

I wonder why Jumbo 0.4.34 in 40/4 has the same rating (after 400 games) as version 0.4.17 - both at 2316 - while in 40/40 version 0.4.34 is clearly above 0.4.17 (currently about +100 points after roughly 200 games in the running gauntlet - +89 after 130 games in the CCRL list). I see no good reason why the improvement of the newer version would shrink to zero with shorter TC. In my own tests with ultra-fast games the improvement over 0.4.17 was about 50-60 Elo points.

Could someone from the CCRL group check, please, whether the version that played those 400 games for the 40/4 list was really Jumbo 0.4.34?

[Sergio Martinez]

I wonder why Jumbo 0.4.34 in 40/4 has the same rating (after 400 games) as version 0.4.17 - both at 2316 - while in 40/40 version 0.4.34 is clearly above 0.4.17 (currently about +100 points after roughly 200 games in the running gauntlet - +89 after 130 games in the CCRL list). I see no good reason why the improvement of the newer version would shrink to zero with shorter TC. In my own tests with ultra-fast games the improvement over 0.4.17 was about 50-60 Elo points.

Could someone from the CCRL group check, please, whether the version that played those 400 games for the 40/4 list was really Jumbo 0.4.34?

Hi Sven

Right now I am away from home but the games in the list are correct: they correspond to version 0.4.34.

In the list, version 0.4.34 has only played 400 games against 8 opponents, in my opinion are few games.

The version 0.4.17 has played more than 1000 games against more than 30 opponents.
The tests are not finished so I think that when we run another 500 or 600 games against other opponents we will have an approximate estimate strengh

[Sven]

I wonder why Jumbo 0.4.34 in 40/4 has the same rating (after 400 games) as version 0.4.17 - both at 2316 - while in 40/40 version 0.4.34 is clearly above 0.4.17 (currently about +100 points after roughly 200 games in the running gauntlet - +89 after 130 games in the CCRL list). I see no good reason why the improvement of the newer version would shrink to zero with shorter TC. In my own tests with ultra-fast games the improvement over 0.4.17 was about 50-60 Elo points.

Could someone from the CCRL group check, please, whether the version that played those 400 games for the 40/4 list was really Jumbo 0.4.34?

Hi Sven

Right now I am away from home but the games in the list are correct: they correspond to version 0.4.34.

In the list, version 0.4.34 has only played 400 games against 8 opponents, in my opinion are few games.

The version 0.4.17 has played more than 1000 games against more than 30 opponents.
The tests are not finished so I think that when we run another 500 or 600 games against other opponents we will have an approximate estimate strengh

Thanks Sergio, sounds plausible!