This Year is going to be hard for Rybka to win the Chess960

[pichy]

http://www.chesstigers.de/live/ccm8_cwc/live.htm

[Tony Thomas]

We havent heard from you in a while Jorge. I guess Rybka doesnt have much of a chance unless Shredder or Naum messes up.

[S.Taylor]

We havent heard from you in a while Jorge. I guess Rybka doesnt have much of a chance unless Shredder or Naum messes up.

goes against my understanding of high elo advantage in playing strength.
strength must mean more than books

[Tony Thomas]

We havent heard from you in a while Jorge. I guess Rybka doesnt have much of a chance unless Shredder or Naum messes up.

goes against my understanding of high elo advantage in playing strength.
strength must mean more than books

In a 3 game tournament, anything can happen.

[Mike S.]

In a 3 game tournament, anything can happen.

IF the finals will be four games each as announced (I'm not sure), then each engine will play sixteen games total. The discripition on the web page is outdated: Actually they played six rounds today, not three and this is also planned for tomorrow.

There was a strange CPU problem which is at least one of the reasons which caused Rybka's first loss. I have read that Vista SP1 forces the installation of a bugfix for AMD Phenom cpus, which slows down the cpu. The Rybka people found a fix which removes this, bringing the Phenoms to full speed. (That is only my simplified view of this matter.)

I didn't read all the details of the technical stuff which was discussed in the Rybka forum about this. People who are interested in this particular topic, should take a look into the two, or more "Mainz" threads there (Mainz is the city where this tournament is played, as a part of a big chess event with Anand, Kosteniuk, and many others).

[pichy]

We havent heard from you in a while Jorge. I guess Rybka doesnt have much of a chance unless Shredder or Naum messes up.

goes against my understanding of high elo advantage in playing strength.
strength must mean more than books

Also the high elo rating of these four computer programs is NOT equal to the FIDE IPS Rating of the best human chess960 players. Simply because the computer rating seems to be a little bit inflated when compared to the human rating at slow time control.

http://www.chesstigers.de/ccm8_index_ne ... ng=1&kat=6

This statement needs some clarification, as far as being the strongest chess tournament…ever! Fide-category 29. You can NOT simply mix inflated computer rating with Fide Human Rating. In my opinion it is hard to predict the outcome of these four computer programs versus the best four human Chess960 players at slow time control. It would be interesting to have these four programs play against..........

1 Ivanchuk,Vassily 2818
2 Anand,Viswanathan 2815
3 Kamsky,Gata 2815
4 Aronian,Levon 2800

http://www.hessischer-schachverband.de/ ... nking.html

[Marc MP]

...

There was a strange CPU problem which is at least one of the reasons which caused Rybka's first loss. I have read that Vista SP1 forces the installation of a bugfix for AMD Phenom cpus, which slows down the cpu. The Rybka people found a fix which removes this, bringing the Phenoms to full speed. (That is only my simplified view of this matter.)

...

Hi Mike,

The bug you talk about is probably the one described there (see TLB errata):

http://en.wikipedia.org/wiki/Phenom_(processor)

I believe that each participant gets the same hardware, so it possible other teams were affected too.

[Tony Thomas]

In a 3 game tournament, anything can happen.

IF the finals will be four games each as announced (I'm not sure), then each engine will play sixteen games total. The discripition on the web page is outdated: Actually they played six rounds today, not three and this is also planned for tomorrow.

There was a strange CPU problem which is at least one of the reasons which caused Rybka's first loss. I have read that Vista SP1 forces the installation of a bugfix for AMD Phenom cpus, which slows down the cpu. The Rybka people found a fix which removes this, bringing the Phenoms to full speed. (That is only my simplified view of this matter.)

I didn't read all the details of the technical stuff which was discussed in the Rybka forum about this. People who are interested in this particular topic, should take a look into the two, or more "Mainz" threads there (Mainz is the city where this tournament is played, as a part of a big chess event with Anand, Kosteniuk, and many others).

I didnt see any descriptions on the page Jorge posted, may be I over looked them. For some reason, I thought that the last invitational was 1RR. I just noticed that Naum won the match against Rybka 1.5-0.5, you go Alex .
Edit : I finally found the correct link by visiting the homepage..If I had to take a guess, the final will be played between Naum and Rybka..

[Uri Blass]

We havent heard from you in a while Jorge. I guess Rybka doesnt have much of a chance unless Shredder or Naum messes up.

goes against my understanding of high elo advantage in playing strength.
strength must mean more than books

Also the high elo rating of these four computer programs is NOT equal to the FIDE IPS Rating of the best human chess960 players. Simply because the computer rating seems to be a little bit inflated when compared to the human rating at slow time control.

http://www.chesstigers.de/ccm8_index_ne ... ng=1&kat=6

This statement needs some clarification, as far as being the strongest chess tournament…ever! Fide-category 29. You can NOT simply mix inflated computer rating with Fide Human Rating. In my opinion it is hard to predict the outcome of these four computer programs versus the best four human Chess960 players at slow time control. It would be interesting to have these four programs play against..........

1 Ivanchuk,Vassily 2818
2 Anand,Viswanathan 2815
3 Kamsky,Gata 2815
4 Aronian,Levon 2800

http://www.hessischer-schachverband.de/ ... nking.html

I think that it seems clear that these GM's are going to lose.
No evidence for inflation in the rating and the rating of ccrl are on slower hardware then the hardware of the tournament.

I can imagine that programs perform better on a quad and not on
Athlon 64 X2 4600+ (2.4 GHz)

Uri

[pichy]

We havent heard from you in a while Jorge. I guess Rybka doesnt have much of a chance unless Shredder or Naum messes up.

goes against my understanding of high elo advantage in playing strength.
strength must mean more than books

Also the high elo rating of these four computer programs is NOT equal to the FIDE IPS Rating of the best human chess960 players. Simply because the computer rating seems to be a little bit inflated when compared to the human rating at slow time control.

http://www.chesstigers.de/ccm8_index_ne ... ng=1&kat=6

This statement needs some clarification, as far as being the strongest chess tournament…ever! Fide-category 29. You can NOT simply mix inflated computer rating with Fide Human Rating. In my opinion it is hard to predict the outcome of these four computer programs versus the best four human Chess960 players at slow time control. It would be interesting to have these four programs play against..........

1 Ivanchuk,Vassily 2818
2 Anand,Viswanathan 2815
3 Kamsky,Gata 2815
4 Aronian,Levon 2800

http://www.hessischer-schachverband.de/ ... nking.html

I think that it seems clear that these GM's are going to lose.
No evidence for inflation in the rating and the rating of ccrl are on slower hardware then the hardware of the tournament.

I can imagine that programs perform better on a quad and not on
Athlon 64 X2 4600+ (2.4 GHz)

Uri

Uri, I also believe that the computer will beat the GM's, but it will NOT be a total disaster for the GM's, specially when they play with the White Pieces.