Has Stockfish 1.6s DC Finally Dethroned Rybka?!!

[slobo]

It appears so! Perhaps I'm being "over-exuberant" just like Dann was.

Go! Test it now!!

Thanks to Dann Corbit/Jim Ablett/ Marco Costalba /Joona Kiiski/ Tord Romstad!!

And good bye to naysayers who tried hard to put Stockfish development down just because of the naming convention!!

It blowed Rybka with 4:0 here.
It is now probably the second best engine of the world, and for sure better than Rybka.

Great to hear that.

Why 2nd best? Have you pitted it against Robbolito yet?

Half of SF games have been played with TM 5 min for game,
other half with TM 2 min + 2 sec.

[mariaclara]

doncha jst lv Unc Dann..

HAPPY NEW YEAR !!!

Just one query.
If somebody other than the engine author/s makes code changes to an engine rather than just a different compile, can that engine be released under the same name?
I'm sure there are examples where this has occurred previously.

Regardless, well done to Dann if he's achieved a huge gain.

It's not a new engine. It's a tiny tweak with a screwdriver to the carb on the Ferrari.

It's not a fully developed idea yet. The entire thing was just to illustrate a simple concept. I guess that other engine authors will improve the idea and find better ways to do the same thing with their engines.

The most exciting thing is having a branch factor less than 2.

[Edward German]

It appears so! Perhaps I'm being "over-exuberant" just like Dann was.

Go! Test it now!!

Thanks to Dann Corbit/Jim Ablett/ Marco Costalba /Joona Kiiski/ Tord Romstad!!

And good bye to naysayers who tried hard to put Stockfish development down just because of the naming convention!! :lol:

Hi!

Why only Rybka?

See here, Stockfish 1.6s 32-Bit (Compiled by Dann Corbit) beat RobboLito on Core Duo 2.66 GHz, my Nick here is "Wolkenlos".

Game on playchess before only few minutes:

[Event "Wertungspartie, 3m + 0s"]
[Site "Maschinenraum"]
[Date "2009.12.30"]
[Round "?"]
[White "Wolkenlos, Stockfish 1.6s 32-Bit"]
[Black "Aggressive king, RobboLito 0.085g3 x"]
[Result "1-0"]
[ECO "A81"]
[WhiteElo "2411"]
[BlackElo "2643"]
[Annotator "0.52;??"]
[PlyCount "105"]
[EventDate "2009.12.30"]
[TimeControl "180"]

{RobboLito 0.085g3 x64: 19.1 ply; 2.821kN/s Intel(R) Core(TM) i7 CPU
975 @ 3.33GHz 3340MHz, (4 threads), en-chess.ctg, 128 MB} 1. d4 {B/0 0} f5 {0}
2. g3 {B/0 0} Nf6 {0} 3. Bg2 {B/0 0} d6 {0} 4. Nf3 {0.52/17 12} c6 {(e6) 0} 5.
O-O {0.64/15 5} Qc7 {(e6) 0} 6. c4 {1.09/15 12} e5 {5} 7. Nc3 {1.01/15 7} Be7 {
0} 8. Bh3 {0.92/16 9} Na6 {(0-0) 9} 9. c5 {0.96/15 4} e4 {0} 10. Nh4 {0.76/16 5
} g6 {0} 11. Bf4 {0.88/16 6} Qd8 {0} 12. cxd6 {0.96/15 4} Bxd6 {0} 13. Bh6 {
1.01/15 4} Qe7 {(Lf8) 0} 14. f3 {1.25/15 5} exf3 {0} 15. Nxf3 {1.29/15 2} Bd7 {
(Sc7) 0} 16. e4 {1.41/16 7} Nxe4 {0} 17. Nxe4 {0.88/15 3} fxe4 {0} 18. Bxd7+ {
1.01/16 4} Qxd7 {7} 19. Ng5 {0.88/19 0} Be7 {(0-0-0) 0} 20. Nxe4 {0.92/15 4}
O-O-O {0} 21. Bg7 {1.05/15 4} Rhg8 {0} 22. Rf7 {1.01/15 6} Qe6 {0} 23. Qf3 {
0.88/12 0} h5 {(Sb4) 0} 24. Rf1 {0.76/14 9} Nc7 {(Tge8) 29} 25. a4 {0.92/15 6}
Ne8 {(Tge8) 4} 26. Be5 {0.28/13 2} Nd6 {0} 27. Rf4 {0.08/15 8} Qa2 {(Sxe4) 0}
28. Nf6 {0.52/13 3} Rgf8 {4} 29. Qd3 {0.40/15 0} g5 {(Sf7) 0} 30. R4f2 {
0.48/14 3} Qxa4 {0} 31. Qg6 {0.88/15 3} Bxf6 {0} 32. Rxf6 {0.96/15 2} Rxf6 {4}
33. Bxf6 {0.92/15 1} Rd7 {4} 34. Qg8+ {0.88/15 0} Kc7 {3} 35. Qxg5 {0.56/16 1}
Qb5 {0} 36. Be5 {0.44/14 2} Qxb2 {0} 37. Qxh5 {0.44/12 1} Qc3 {0} 38. Qg6 {
0.40/12 2} a5 {1} 39. h4 {0.40/12 0} a4 {3} 40. h5 {0.32/13 0} a3 {(Kb6) 0} 41.
Qe6 {2.18/14 2} Qe3+ {9} 42. Kg2 {2.82/18 0} Qe2+ {0} 43. Rf2 {2.82/16 1} Qe4+
{0} 44. Kh2 {3.07/17 1} Qd5 {0} 45. Qxd5 {2.94/16 1} cxd5 {0} 46. Ra2 {
3.23/16 1} Kc6 {(Kd8) 0} 47. Bxd6 {2.90/17 2} Kxd6 {0} 48. Rxa3 {3.31/15 0} Rd8
{0} 49. g4 {4.16/16 1} Rb8 {0} 50. g5 {4.48/17 1} b5 {0} 51. Kg3 {4.84/18 1}
Ke6 {0} 52. Kg4 {6.74/17 1} b4 {0} 53. Re3+ {Aggressive king,RobboLito 0.085g3
x6 bietet Remis Aggressive king,RobboLito 0.085g3 x6 bietet Remis Aggressive
king,RobboLito 0.085g3 x6 bietet Remis Aggressive king,RobboLito 0.085g3 x6
bietet Remis Aggressive king,RobboLito 0.085g3 x6 bietet Remis Aggressive king,
RobboLito 0.085g3 x6 bietet Remis Aggressive king,RobboLito 0.085g3 x6 bietet
Remis Aggressive king,RobboLito 0.085g3 x6 gibt auf (Lag: Av=0.31s, max=2.6s)
9.17/17 3} 1-0

Stockfish stronger here than Robbo...! :)))

My Best,
Eduard

[solis]

This is what I got today on dual core.The same Rybka that was taking the beating from Robbolito.
Small number of games but still a win for Rybka.
Engine Score Ry St S-B
1: Rybka 3.x64 7.0/10 ····· 4-0-6 21.00
2: Stockfish-16s-x64-ja 3.0/10 0-4-6 ····· 21.00

10 games played / Tournament finished
Name of the tournament: Arena tournament
Site/ Country: Milwaukee,USA
Level: Blitz 10/0
Hardware: Pentium(R) Dual-Core CPU T4300 @ 2.10GHz with 3,032 MB Memory
Operating system: Windows 7 Home Premium Home Edition (Build 7600)

[kingliveson]

This is what I got today on dual core.The same Rybka that was taking the beating from Robbolito.
Small number of games but still a win for Rybka.
Engine Score Ry St S-B
1: Rybka 3.x64 7.0/10 ····· 4-0-6 21.00
2: Stockfish-16s-x64-ja 3.0/10 0-4-6 ····· 21.00

10 games played / Tournament finished
Name of the tournament: Arena tournament
Site/ Country: Milwaukee,USA
Level: Blitz 10/0
Hardware: Pentium(R) Dual-Core CPU T4300 @ 2.10GHz with 3,032 MB Memory
Operating system: Windows 7 Home Premium Home Edition (Build 7600)

Just keep in mind that when the results come out and Rybka is not ahead, you will be accused of using a weakened pirated fake no good purposely mis-labeled version of Rybka.

[gerold]

This is what I got today on dual core.The same Rybka that was taking the beating from Robbolito.
Small number of games but still a win for Rybka.
Engine Score Ry St S-B
1: Rybka 3.x64 7.0/10 ····· 4-0-6 21.00
2: Stockfish-16s-x64-ja 3.0/10 0-4-6 ····· 21.00

10 games played / Tournament finished
Name of the tournament: Arena tournament
Site/ Country: Milwaukee,USA
Level: Blitz 10/0
Hardware: Pentium(R) Dual-Core CPU T4300 @ 2.10GHz with 3,032 MB Memory
Operating system: Windows 7 Home Premium Home Edition (Build 7600)

Just keep in mind that when the results come out and Rybka is not ahead, you will be accused of using a weakened pirated fake no good purposely mis-labeled version of Rybka.

They are going to skin me alive. Left it running all day with
no other programs running on my dual core. HP
Level.10/0
Tankist 1.2 vs.Stockfish 1.6 JC. 33-8-29.
.
68% - 32%

Best,

Gerold.

.

[gerold]

IN the above post that was Stockfish 1.6 DC.

Will run the same test tomorrow all day and see what happens.

Best,

Gerold.

[M ANSARI]

Stockfish 1.6s is scoring around 37% in my 5 min +1sec match against R3. Hardware is Octa 8 core @4Ghz with Perfect book truncated at 10 moves. This is about 5% better score than JA version. Although that is a nice improvement, I was hoping the scaling on more cores would be more prominent.

[Jouni]

No the 1. challenge is Naum4:

Nunn openings, 32b, 60/3

Code: Select all

                    
1   Naum 4           ½11111½1½½½½01110½1½  14.0/20
2   Stockfish 1.6s   ½00000½0½½½½10001½0½   6.0/20

Jouni

[solis]

This is what I got today on dual core.The same Rybka that was taking the beating from Robbolito.
Small number of games but still a win for Rybka.
Engine Score Ry St S-B
1: Rybka 3.x64 7.0/10 ····· 4-0-6 21.00
2: Stockfish-16s-x64-ja 3.0/10 0-4-6 ····· 21.00

10 games played / Tournament finished
Name of the tournament: Arena tournament
Site/ Country: Milwaukee,USA
Level: Blitz 10/0
Hardware: Pentium(R) Dual-Core CPU T4300 @ 2.10GHz with 3,032 MB Memory
Operating system: Windows 7 Home Premium Home Edition (Build 7600)

Just keep in mind that when the results come out and Rybka is not ahead, you will be accused of using a weakened pirated fake no good purposely mis-labeled version of Rybka.

You are right,
this is the same so called pirated and weakened no good version I bought from Convecta in addition to version I bought from CB.Now nobody said a word but when this Rybka was destroyed by Robbolito even on the 4 cores same engine was fake and who knows what else.Always excuses insted to wait for Vas to release Rybka 4 engine and if he is capable to regain the primat in chess again.
One has to say that SF team has done great job with this new engine as resultas are showing.
Also if some of the improvement ideas came from Robbo that would reinforce what I have said before that Robbo as a new engine could help programers to develop stronger new engines
One more thing that I have noticed playing SF 16 is evaluation that appears to be much higher that the position is showing.I had one example to show but when I tried to copy it all the move evaluation was there so the text would be very long ans it would be impractical to put it here.
But once again my congratulations to SF team and thanks for the pleasure of playing that engine