Terry McCracken wrote:If you can afford it you can get them now. Call Cray for a personal supercomputer or build it yourself and you'll get billions of nodes per sec.
By 2035 machines will be so fast it will be impossible to know how fast they'll be able to search. They will be at least a billion times faster than today!!!
Modern chess programs are not designed to be massively parallel.
But I'm sure that in the day someone will figure out how to write a chess program that can run on a supercomputer as fast as Cray or even the Roadrunner we'll have chess more or less solved. It will play like God or like something never seen before in the history of chess.
Still even if someone figured out how to design such a massively parallel chess program, I'm sure it will have some major bugs and other flaws in that in some positions no matter how long it thinks it still can't find the best move. For example Fritz 8, an old program, has some serious bugs which in some repetitive positions where one side is leading in material no matter how long it thinks it still thinks it's a draw.
But surely such a thing will beat the best chess players of our time rather easily like Kramnik, Toplaov, Radjabov and Ivanchuk.
Last edited by Uri on Tue Feb 24, 2009 9:40 pm, edited 1 time in total.
Terry McCracken wrote:If you can afford it you can get them now. Call Cray for a personal supercomputer or build it yourself and you'll get billions of nodes per sec.
By 2035 machines will be so fast it will be impossible to know how fast they'll be able to search. They will be at least a billion times faster than today!!!
Modern chess programs are not designed to be massively parallel.
But I'm sure that in the day someone will figure out how to write a chess program that can run on a supercomputer as fast as Cray or even the Roadrunner we'll have chess more or less solved. It will play like God or like something never seen before in the history of chess.
Still even if someone figured out how to design such a massively parallel chess program, I'm sure it will have some bugs and other flaws in that in some positions no matter how long it thinks it still can't find the best move.
But surely such a thing will beat the best chess players of our time rather easily like Kramnik, Toplaov, Radjabov and Ivanchuk.
As if the current software on the current hardware can't do it now
Did you read Anand's statement in another thread confessing that his computer if stronger than him Mr.Goldy,this is what Zlatnik means,right
Cyrillic regards,
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
Dr.Wael Deeb wrote:Did you read Anand's statement in another thread confessing that his computer if stronger than him Mr.Goldy,this is what Zlatnik means,right
Cyrillic regards,
Dr.D
Yes Zlatnik means gold in slovene and in Czech it means goldworker.
Dr.Wael Deeb wrote:Did you read Anand's statement in another thread confessing that his computer if stronger than him Mr.Goldy,this is what Zlatnik means,right
Cyrillic regards,
Dr.D
Yes Zlatnik means gold in slovene and in Czech it means goldworker.
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
Terry McCracken wrote:If you can afford it you can get them now. Call Cray for a personal supercomputer or build it yourself and you'll get billions of nodes per sec.
By 2035 machines will be so fast it will be impossible to know how fast they'll be able to search. They will be at least a billion times faster than today!!!
Modern chess programs are not designed to be massively parallel.
Dr.Wael Deeb wrote:Did you read Anand's statement in another thread confessing that his computer if stronger than him Mr.Goldy,this is what Zlatnik means,right
Cyrillic regards,
Dr.D
I'm not sure if Anand said that in order to please the weak player audience or if he really meant it when he said that the computer is stronger than him. I think he didn't quite mean it because many positions Anand clearly understands better than the computer.
For example the Sicilian Najdorf, the Sicilian Dragon and various Gambits are openings which many GMs and even IMs play much better than any chess program. The openings is a phase of the game where the computer is weak without its opening database. Also I have to admit that compared to Anand I'm a very weak player, both tactically and strategically but especially tactically since I have studied chess strategy quite deeply.
I'm just not convinced that the computer is unbeatable. I'm sure that if it played in Fide with the 100 best GMs it will get crushed and then we will probably know it's true rating. The top 100 GMs can definitely beat it, maybe not in an 8 game match but they can still beat it.
Roman Dzindzichashvili and Milov Vadim managed to beat Rybka several times meaning it's not invincible yet.
Dr.Wael Deeb wrote:Did you read Anand's statement in another thread confessing that his computer if stronger than him Mr.Goldy,this is what Zlatnik means,right
Cyrillic regards,
Dr.D
I'm not sure if Anand said that in order to please the weak player audience or if he really meant it when he said that the computer is stronger than him. I think he didn't quite mean it because many positions Anand clearly understands better than the computer.
For example the Sicilian Najdorf, the Sicilian Dragon and various Gambits are openings which many GMs and even IMs play much better than any chess program. The openings is a phase of the game where the computer is weak without its opening database. Also I have to admit that compared to Anand I'm a very weak player, both tactically and strategically but especially tactically since I have studied chess strategy quite deeply.
I'm just not convinced that the computer is unbeatable. I'm sure that if it played in Fide with the 100 best GMs it will get crushed and then we will probably know it's true rating. The top 100 GMs can definitely beat it, maybe not in an 8 game match but they can still beat it.
Roman Dzindzichashvili and Milov Vadim managed to beat Rybka several times meaning it's not invincible yet.
I think you should remember that the games that Roman and Milov won were handicapped games ... I don't recall any of them winning a non handicapped game.
Here is a letter that one of the world's top GM's wrote that should more than convince anyone how strong they think Rybka has become, they certainly think that it is almost unbeatable.
Dear Sirs
Looking at various chess sites, I noticed that there are still many aspects of GM Kurnosov’s games at the Aeroflot Open which remain unclear, and I should like to bring some clarity to some of these. After the tournament, I carefully analysed again Kurnosov’s games, played up to round six and onwards. Certain facts emerge from this.
Prior to round six, he played three games as Black. In round two, we have the game Onischuk-Kurnosov. In this game, Onischuk played the novelty 13.Bb2 and from then until the end of the game, i.e. move 27, Kurnosov played every single move strictly in accordance with the first choice of Rybka, and won in great style. Two rounds later, he was again Black, and again played the Grunfeld Defence in the game Moiseenko-Kurnosov. Moiseenko played the novelty 12.Nd4 and once again, Kurnosov followed Rybka’s first choice line down to move 25.
However, in this game, he was not able to win, as Moiseenko obtained a fairly simple position, without risk. But even so, it seems to me that Black could still have played for a win in the final position. In round six, against me, Kurnosov again followed Rybka’s first choice line and won.
In all three of these games, very few of Rybka’s second-choice moves were played, and it was precisely these moves which were the weakest ones played. In the games against Onischuk and myself, we each sacrificed a pawn, and in each case, within a few moves, the computer constructs a good defence and wins with a counter-attack.
Thus it was that in round eight, when he was under the closest scrutiny from the arbiters and spectators, Kurnosov played extremely weakly and lost. In my mind, this is a clear demonstration that Kurnosov had been using the help of a computer program, leaving the playing hall after every move, and having such confidence in himself, that he turned down a draw against me on move 14, with the black pieces, in an absolutely equal position.
I hope that all of the above details will be studied, and will be understood by all chess lovers around the world. I have no doubt that future tournament organisers will take account of these details, which impact badly on the image of chess.
Respectfully,
International Grandmaster
S. Mamedyarov.
07.03.09.
M ANSARI wrote:I think you should remember that the games that Roman and Milov won were handicapped games ... I don't recall any of them winning a non handicapped game.
I guess you are right but take note that Oligarckh from Playchess managed to beat Deep Fritz 11 in the endgame. It might be 1 game in 30 though.
Chess programs playing style reminds me a lot of Paul Morphy. Paul Morphy was a strong tactical player and master of the open game. He liked sacrificial play a lot. He often liked to sacrifice pieces just to have a mating attack.
I'm sure Kasparov in his glory days could beat Rybka. Take this game against Deep Thought for example:
Deep Thought made 4 moves with his knight in the opening capturing two kingside pawns instead of developing his pieces and castling early which is a complete violation of modern opening theory and Kasparov punished him. Surprisingly I analysed the game with Rybka 3 and it made the same mistakes as Deep Thought and couldn't find Kasparov's moves.
A computer often doesn't make the difference between a good pawn and a bad pawn. To him all pawns are the same and that is sometimes a big mistake in evaluation.
M ANSARI wrote:I think you should remember that the games that Roman and Milov won were handicapped games ... I don't recall any of them winning a non handicapped game.
I guess you are right but take note that Oligarckh from Playchess managed to beat Deep Fritz 11 in the endgame. It might be 1 game in 30 though.
Chess programs playing style reminds me a lot of Paul Morphy. Paul Morphy was a strong tactical player and master of the open game. He liked sacrificial play a lot. He often liked to sacrifice pieces just to have a mating attack.
I'm sure Kasparov in his glory days could beat Rybka. Take this game against Deep Thought for example:
Deep Thought made 4 moves with his knight in the opening capturing two kingside pawns instead of developing his pieces and castling early which is a complete violation of modern opening theory and Kasparov punished him. Surprisingly I analysed the game with Rybka 3 and it made the same mistakes as Deep Thought and couldn't find Kasparov's moves.
A computer often doesn't make the difference between a good pawn and a bad pawn. To him all pawns are the same and that is sometimes a big mistake in evaluation.
No question Kasparov could then and no doubt today if he put his mind to the task.
I'm certain Anand could as well. I just missed a win at G/5, playing the Black pieces over the net against Rybka 3 running on a Q6600 with 2gigs of hash the other day. I was very tired and mixed my move order which cost the game.
If I can get that beast into a lost position, at blitz with Black no less, then no question a top GM could as well and catch his limit!
Terry McCracken wrote:No question Kasparov could then and no doubt today if he put his mind to the task.
I'm certain Anand could as well. I just missed a win at G/5, playing the Black pieces over the net against Rybka 3 running on a Q6600 with 2gigs of hash the other day. I was very tired and mixed my move order which cost the game.
If I can get that beast into a lost position, at blitz with Black no less, then no question a top GM could as well and catch his limit!
What does G/5 stand for? Was it on Playchess?
Also computers can't play blindfold chess, only humans can.
Anyway any chess program is a great beauty regardless if it's stronger than the world champion or not and to write a strong chess program is a difficult task only relatively few people are capable of.
