Rybka 4 vs Stockfish 1.8 - Long TC Match - LIVE BROADCAST!

[Martin Thoresen]

Thanks for the interest, everyone!

Of course all settings are default (or standard). Also, these matches has nothing to do with CCRL testing.

Best Regards,
Martin

[gerold]

Thanks for the interest, everyone!

Of course all settings are default (or standard). Also, these matches has nothing to do with CCRL testing.

Best Regards,
Martin

Thanks Martin.

Best,
Gerold.

[Martin Thoresen]

Thanks Martin.

Best,
Gerold.

No problem!

This is just to clarify to Ignacio Garcia:

All engines I test in these matches are using default settings (or standard settings, or "out of the box settings" if you like). I never test engines with modifications or with changed parameters.

To S. Taylor:
"Go both!
It's not about strength testing, it's only for entertainment!"

Well, I am a huge Stockfish fan because Glaurung is an original engine created by the Norwegian Tord Romstad. I am a Norwegian my self, so this should explain a lot. But you are right, go both!

Best Regards,
Martin

[IGarcia]

Good to hear about "out the box settings". This is very important.

I do think its not very hard to "tune" a engine if its going to play a match against another (only one) engine. For example, you can use STS tests to spot strong and weak points for an engine and then tune your favorite engine to avoid opposite strong positions and try to push the game into a weak opposite game. This can be done by giving different values to Knights, bishops or bishop pair bonus, etc...


To be honest, I dislike SF game play in G3, currently at move 32.Ne1 0.68/26 1:51 ... I don't like white bishop at b5, now cant do much from there. But because my chess level (I play ugly bad) this may be a good SF game!! Lets see.

[LaurenceChen]

It looks very difficult for White to win in this position after the 32. Ne1. Black should be able to equalize the game and draw without any issues. This game should end in a draw. I don't trust the PV scores shown by the engines.

[IGarcia]

It looks very difficult for White to win in this position after the 32. Ne1. Black should be able to equalize the game and draw without any issues. This game should end in a draw. I don't trust the PV scores shown by the engines.

G3 Shock. Not only black equalized.. R4 manage to get the full point. And the bishop at b5 was like a veggetal

[LaurenceChen]

Well, Rybka won because Stockfish overstretched itself. Black had a slight pull from move 32 onwards. Although the evaluation from the position from the engines point of view showed that White had the small advantage. The loss by Stockfish doesn't surprise me... This is a great game which shows why evaluation, PV, by engines must be seen critically and never taken at face value... Perhaps, this is why centaurs, man + machine, have an advantage in CC games over opponents who solely rely on the engines evaluation.

[S.Taylor]

SF almost saved this game (G2), which would have shown greatness.
My (early) prediction is that SF will win atleast 1.5 more games from 48 than Houdini did. (but hopefully more) Unless, SF has a problem in choosing wrong endgames.

In this game, after all the terrible attack and near mates SF was faced with, the only thing wrong at the end, when it had all been thwarted, was that Rybka could get his pawns at the end of it, which means he could then march his own down to queen. I don't know if SF had another way withOUT letting Rybka do this at the end of it.

Thanks for you remarks Taylor.
In fast TC Houdini beats up on SF.
Houdini in fast TC is about equal with R4.

Gerold, thanking me for my remarks seems to be a polite way of not saying they are great. Indeed, in these computer games, you are not actually commenting on the games, as if on principal.

Indeed, i withdraw my prediction (of SF doing better vs Rybka 4, than Houdini did).

Martin,
Are you allowed to change details after you say you will do something? If so, can i say what _I_ would vote for?
That is, that IF by game 12, Rybka has earned a +5 score, then the outcome is certain enough, and you need not show us how well Rybka crushes SF... unless that was what interests us.
IF SF has won one game, and you want to see it happen again, in order to observe things a bit more, then I would give it to the very most, till game 24.
Next match?
If new Houdini is out by July 15, and seems to be a clear improvement over Houdini 1.02, then that can go vs. Rybka. If not, do other things (Hiarcs/SF?) and leave Rybka for later.

[IGarcia]

Taylor: For me it has to be played at least half match (24 games) to give an evaluation of decision on match change. I like we having less draws, and game 5 won by SF brings some hope about this will not be SF doomed

[S.Taylor]

Taylor: For me it has to be played at least half match (24 games) to give an evaluation of decision on match change. I like we having less draws, and game 5 won by SF brings some hope about this will not be SF doomed

Indeed, we will get a better idea over the next few games.
It is now much less likely that Rybka will be +5 over SF by game 12.

(But if the next 6 games [7-12] DID score 5-1 to Rybka, then still, I'd feel that it is a cruel competition)