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

[Martin Thoresen]

Wasn't game 28 adjudicated earlier than it should (according to the "early" definition)? There were no 6 consecutive plies above 4.8.

Hello Peter,

I checked the game and you are right. I can't explain why, as I am just following what Chessbase have written about this topic.

http://www.chessbase.com/workshop2.asp?id=4615

If you're interested in the technical details behind the Fritz software's decision to resign, it really has everything to do with the engine's numerical evaluation. With "Early" selected, the Fritz program will resign if it sees it's behind by 4.8 pawn units (or more) three moves in a row. So you'll need to be ahead by about a Rook, or two minor pieces, or a minor and a couple of pawns, or some combination of material and positional criteria exceeding 4.8 pawns (and maintain that advantage for at least three consecutive turns) before Fritz calls it quits with the "Early" setting selected. With "Late" selected, you'll need to be ahead by a larger margin (although still for just three consecutive turns); the margin must be at least 6.5 pawns (or the material/positional equivalent) when you've clicked the radio button beside "Late".

[pacifist]

It seems Rybka is in good form . Still possible for SF to do similar score as Houdini but it will be very hard now. I was hoping better result from SF.

[Martin Thoresen]

It seems Rybka is in good form . Still possible for SF to do similar score as Houdini but it will be very hard now. I was hoping better result from SF.

Well I kind of agree. SF hasn't won in 12 games or so. But it's not over yet!

[S.Taylor]

It seems Rybka is in good form . Still possible for SF to do similar score as Houdini but it will be very hard now. I was hoping better result from SF.

Well I kind of agree. SF hasn't won in 12 games or so. But it's not over yet!

If it did catch up, it would look like something went wrong.

[pacifist]

Something is wrong : 10/13 the last 13 games? Settings are the same?

[Martin Thoresen]

Something is wrong : 10/13 the last 13 games? Settings are the same?

Of course, this is an automated match which means I don't ever touch any settings during the match.

It's simply statistical noise IMO.

[S.Taylor]

Something is wrong : 10/13 the last 13 games? Settings are the same?

I saw near the beginning, that SF finds it hard to cope with Rybka 4. Any win, was just at great stress, not like it had the technique and was on top of it.
According to the way it appeared to me, when seeing early games, it is not surprising what has happened in the past 13 games. It's just that I HOPED it would be otherwise.

[pacifist]

Match is interrupted? I saw next match is already planned in 2 days so this one is over?

[Martin Thoresen]

Match is interrupted? I saw next match is already planned in 2 days so this one is over?

Not over yet, but Stockfish crashed in game 42 so progress was halted because this happened during the night when I was asleep (of course).

Game 42 has been restarted and the match should finish during tomorrow.

[pacifist]

Ah ok then:). But I remember the final position was winning for Rybka, Stockfish then receives a gift from the gods to redeem himself .