Stockfish 1.7.1 - Rybka 3 (both single core)

[alpha123]

Intel Core 2 Quad Q8300 @ 2.50gHz
6GB DDR2
Windows 7 Home Premium 64-bit

40/3 repeating (CCRL blitz adapted to my comp)
Silver Suite
Shredder 12 GUI
Both engines single core 64 bit (I don't own Deep Rybka)

Stockfish 1.7.1 JA 64bit single-core - Rybka 3 1-cpu 29.0 - 21.0 (50: +19, =20, -11) 58.00% Perf = +56 Elo

Games


Impressive performance by Stockfish! (Small number of games though)

Don't give me the "But it wasn't contempt 0!" because 15 is Rybka's default, and it makes no sense to test at non-default settings.


Peter[/url]

[alpha123]

Oh yeah, ponder was ON.

Peter

[ernest]

Don't give me the "But it wasn't contempt 0!"

Then why don't you try it again, this time with contempt 0 (that was Vasik's recommendation for close strength engines) to see if that makes any difference?...

[alpha123]

Don't give me the "But it wasn't contempt 0!"

Then why don't you try it again, this time with contempt 0 (that was Vasik's recommendation for close strength engines) to see if that makes any difference?...

When I'm done testing the new SMP Critter, I will, but I am fundamentally opposed to testing engines with non-default settings.


Peter

[Sven]

Don't give me the "But it wasn't contempt 0!"

Then why don't you try it again, this time with contempt 0 (that was Vasik's recommendation for close strength engines) to see if that makes any difference?...

When I'm done testing the new SMP Critter, I will, but I am fundamentally opposed to testing engines with non-default settings.

After having followed some former threads about that "contempt" topic I think it is possible that you misinterpret what the "default" really means. It has been made clear by others that the consequences of setting "contempt" depend on the strength of the opponents. This implies that, although there is only one value that is formally the "default" value, this default is suitable either for opponents close in strength or for significantly weaker (stronger?!) engines, but not both at the same time. The program author had to decide to which of the two groups he wanted to dedicate that default value, but he also gave an information about which other value were better for the other group.

So I consider it to be immediately evident that in reality there are two different "default values" given, where only one is set automatically and the other is given as an information, which could also have been a user guide document. I see no point in ignoring that additional information.

If it were possible for an engine to determine the opponent's strength in advance somehow, and adapt its settings based on that, then there would probably be two "real" defaults in the code. But you don't have that.

Sven

[Dr.Wael Deeb]

Don't give me the "But it wasn't contempt 0!"

Then why don't you try it again, this time with contempt 0 (that was Vasik's recommendation for close strength engines) to see if that makes any difference?...

If Vasik was so smart,why didn't he leave it as a default option

[Sven]

Don't give me the "But it wasn't contempt 0!"

Then why don't you try it again, this time with contempt 0 (that was Vasik's recommendation for close strength engines) to see if that makes any difference?...

If Vasik was so smart,why didn't he leave it as a default option

If you were so smart, why didn't you notice what I wrote above about that?

Sven

[Dr.Wael Deeb]

Don't give me the "But it wasn't contempt 0!"

Then why don't you try it again, this time with contempt 0 (that was Vasik's recommendation for close strength engines) to see if that makes any difference?...

If Vasik was so smart,why didn't he leave it as a default option

If you were so smart, why didn't you notice what I wrote above about that?

Sven

Got me with pants down this time
If any excuse,returning home after a long night shift type nightmare

Seriously,all this talk about the contempt factor isssue is trivial to say the least....I presonaly will be pleased to see Rybka 3 a$$ kicked with it's default settings....changing anything in the setings is a personality creation.Period.
Dr.D

[Sven]

changing anything in the setings is a personality creation.Period.

So all immutable personal opinions have now been exchanged on that topic.Period.

Sven

[alpha123]

Don't give me the "But it wasn't contempt 0!"

Then why don't you try it again, this time with contempt 0 (that was Vasik's recommendation for close strength engines) to see if that makes any difference?...

When I'm done testing the new SMP Critter, I will, but I am fundamentally opposed to testing engines with non-default settings.

After having followed some former threads about that "contempt" topic I think it is possible that you misinterpret what the "default" really means. It has been made clear by others that the consequences of setting "contempt" depend on the strength of the opponents. This implies that, although there is only one value that is formally the "default" value, this default is suitable either for opponents close in strength or for significantly weaker (stronger?!) engines, but not both at the same time. The program author had to decide to which of the two groups he wanted to dedicate that default value, but he also gave an information about which other value were better for the other group.

So I consider it to be immediately evident that in reality there are two different "default values" given, where only one is set automatically and the other is given as an information, which could also have been a user guide document. I see no point in ignoring that additional information.

If it were possible for an engine to determine the opponent's strength in advance somehow, and adapt its settings based on that, then there would probably be two "real" defaults in the code. But you don't have that.

Sven

Sigh. You just don't get it, do you.....

Anyway, the match is running, after 32 games Stockfish is down 15.0-17.0 (+7 =16 -9), but we still have lots of games to go.....

Peter