Strelka 5.5 x32 is released....

[Dr.Wael Deeb]

ouch ! This post by Robert will calm down many people

Gab

Robert is a seller man and every seller man must defend his product by any means necessary....
Dr.D

[gleperlier]

Code: Select all

i7-2670QM, Blitz 5 minutes KO

                                  
1   Strelka 5.5               +6  +15/=33/-14 50.81%   31.5/62
2   Komodo64 SSE Version 4    -6  +14/=33/-15 49.19%   30.5/62

[Houdini]

At move/0.05s the result of 2,000 games match is:

Code: Select all

    Program                            Score     %       Elo    +   -    Draws

  1 Strelka 5.5                    : 1145.5/2000  57.3   3225   12  12   38.6 %
  2 Houdini 1.5a w32               :  854.5/2000  42.7   3175   12  12   38.6 %

+51 Elo points for Strelka. Also, for Arasan14 test suite (5s+1ply / position) the results are so much in favour of Strelka 5.5 that I am suspecting some oddity. Maybe I have to restart my comp and check for GUI bugs and such, but as of now Strelka seems clearly stronger.

Kai

Can you run a test at a real TC like 5"+0.05"?
At 0.05"/move you're using a fringe case of the TM of the engines, I wouldn't recommend this for testing engines with different TM strategies (as is the case).

Robert

[Albert Silver]

ouch ! This post by Robert will calm down many people

Gab

Robert is a seller man and every seller man must defend his product by any means necessary....
Dr.D

Well, after the results of +45, I figured that would make it about as strong as Houdini 1.5 x64, and started a g/5 match. After 30 games, Strelka was under 25%, so... If Strelka 5.5 is the second coming, I did not see it.

[Arpad Rusz]

This mate in 9 is too difficult for Strelka:

[d]8/8/5N2/8/8/8/4Kpb1/5Qbk w - - 0 1

Ng4!

Probably Strelka has a known Houdini 1.5 bug.

[gleperlier]

Hello,

Here is a game I just played with Vitruvius (2 CPU) versus Strelka on Playchess :


[Event "Rated game, 60 m + 15s"]
[Site "Engine Room"]
[Date "2012.04.24"]
[Round "?"]
[White "Strelka 5.5"]
[Black "Vitruvius_1.11_HEM"]
[Result "0-1"]
[ECO "C13"]
[WhiteElo "2498"]
[Annotator "0.09;0.00"]
[PlyCount "51"]
[EventDate "2012.04.24"]
[TimeControl "3600+15"]

{Strelka 5.5: 20.9 plies; 1 957kN/s Intel(R) Core(TM) i7 CPU X 980 @ 3.
33GHz, (6 CPUs),} 1. e4 {B 0} e6 {B 0} 2. d4 {B 0
} d5 {B 0} 3. Nc3 {B 0} Nf6 {B 0} 4. Bg5 {B 0} dxe4 {B 0} 5. Nxe4 {B 0} Be7 {
B 0} 6. Bxf6 {B 0} Bxf6 {B 0} 7. Nf3 {B 0} O-O {B 0} 8. Bc4 {B 0} Nc6 {B 0} 9.
c3 {B 0} e5 {B 0} 10. d5 {B 0} Ne7 {B 0} 11. O-O {B 0} Ng6 {0.00/23 115} 12. a4
{B 0 (Ne4xf6+)} Be7 {-0.10/24 295} 13. a5 {B 0} Kh8 {-0.08/23 49} 14. b4 {0.09/
21 130} Rb8 {-0.12/24 0} 15. Nc5 {-0.15/20 185 (Rf1-e1)} b6 {-0.46/23 182
(Lxc5)} 16. axb6 {-0.90/20 84 (Nc5-e4)} axb6 {-0.50/23 84} 17. Nd3 {-0.90/21
58 (Nc5-a6)} e4 {-0.60/23 95 (Ld6)} 18. Nd4 {-1.28/21 85 (Nf3-e5)} Bb7 {-0.89/
24 270} 19. Nb2 {-1.87/22 123} Bxd5 {-0.89/23 0} 20. Qc2 {-1.93/21 78 (Qd1-b3)}
Bb7 {-0.93/22 141} 21. Rfd1 {-1.96/21 31 (Bc4-a6)} Qc8 {-1.15/23 118} 22. b5 {
-2.01/22 0} Ra8 {-1.18/23 53} 23. Rxa8 {-2.09/22 99} Qxa8 {-1.18/24 0} 24. Bb3
{-1.94/22 82} Nf4 {-1.22/24 0 (Se5)} 25. Nc4 {-1.99/22 95} Nd3 {-1.13/24 0} 26.
Ne3 {-2.12/21 60,Strelka 5.5 gibt auf} 0-1

Not that impressive ....

[ernest]

Probably Strelka has a known Houdini 1.5 bug.

Indeed Houdini 1.5/Houdini 1.5a does not find without EGTB
Houdini 2.0c does find.

[Dr.Wael Deeb]

ouch ! This post by Robert will calm down many people

Gab

Robert is a seller man and every seller man must defend his product by any means necessary....
Dr.D

Well, after the results of +45, I figured that would make it about as strong as Houdini 1.5 x64, and started a g/5 match. After 30 games, Strelka was under 25%, so... If Strelka 5.5 is the second coming, I did not see it.

I didn't test it yet at longer time controls so we'll see after a while I guess

[rvida]

Also, for Arasan14 test suite (5s+1ply / position) the results are so much in favour of Strelka 5.5 that I am suspecting some oddity.
Kai

Arasan suite / 30s:
Solved 140/175 in 23m (Critter MP 127 in 37m)

Strelka 5.5 has a database of best-moves stored in the executable file for many popular test positions. (29643 positions in total).

When it reaches depth 5 in such position, it excludes all other moves and searches only the stored best-move.

[A Distel]

ouch ! This post by Robert will calm down many people

Gab

Robert is a seller man and every seller man must defend his product by any means necessary....
Dr.D

People like to feel they are buying of their own good judgment as a result of the information the salesman has given them