|
< Earlier Kibitzing · PAGE 135 OF 136 ·
Later Kibitzing> |
| Dec-09-06 | | barbababa: <lunacyfrog> Can you find the moves they played somewhere? |
|
| Dec-09-06 | | sharpnova: <acirce> it was 4/6 so rethink your ratings.
Kramnik was destroyed. Yes that blunder was a bit out of character, butit was Kramnik who made the blunder was it not? A performance rating based on six games does mean something. More than you are all making it out to mean. On top of all this, Kramnik getting to prepare with fritz 10 and its opening book weeks in advance is a HUGE decisive factor. He got to know what openings it would play and what kind of moves it would make in ANY situation. All you kramnik fanboys are downplaying fritzes performance WAY too much. I've never seen so much illogical unfounded bias in my life. |
|
| Dec-09-06 | | sharpnova: Kramnik is a bit stronger than Adams. Fritz is a good bit weaker than Hydra. And still the computer trounced him. This was a hard trouncing. Time to rethink all you little "human" in the "human-computer duel" fanboys. Time to rethink your awefully low and ungenerous performance ratings of Hydra's performance. Now we can really say Hydra was at least a 3100 performance based on 6 games. |
|
| Dec-09-06 | | percyblakeney: The Deep Fritz 10 tests are slowly getting more reliable, with tournament time controls it is 2829 at CEGT at the moment, ranked as the 7th strongest available engine (Hydra not included): http://www.husvankempen.de/nunn/40_... |
|
| Dec-09-06 | | TIMER: <percyblakeney> If those ratings are comparable with FIDE, only Rybka has made a breakthrough above the best human level achieved (by Kasparov) of the mid 2800s. It would be interesting to have it play a human, and see if it performs that much better than Fritz 10. |
|
| Dec-09-06 | | percyblakeney: <TIMER> Yes, apart from Rybka the other engines aren't too far ahead. Sometimes a tournament winner has a performance close to 2850, so the best humans aren't without chances. Rybka does seem to play in another division though, preliminary reports say it's at least 150 points stronger than Deep Fritz and I think it would show in a match against human opposition. |
|
| Dec-09-06 | | Eyal: <TIMER: <It would be interesting to have it play a human, and see if it performs that much better than Fritz 10.> Certainly, but I doubt if after what happened to Adams any other top player would agree to the experiment. |
|
| Dec-09-06 | | botevist: <lunacyfrog: ... You are really suggesting that Kramnik knows that he will lose to the computer know matter what he does and is willing to be humiliated on a public stage, just for a paycheck?> Of course he would be willing. Is there anyone who doesn't think so? |
|
| Dec-09-06 | | Larsker: <You are really suggesting that Kramnik knows that he will lose to the computer know matter what he does and is willing to be humiliated on a public stage, just for a paycheck?> Humiliated? He made one terrible blunder but apart from that he achieved 4 draws which is ok. And 500.000 Euros are $665.000 which is pretty nifty for a week's work + preparation. Chess players are generally badly paid. |
|
| Dec-09-06 | | chessmoron: <Larsker> Not 500.000 Euros, it's US $500,000. A 2841 computer beating a 2750 human player is not humiliating at all. |
|
| Dec-09-06 | | Larsker: <chessmoron> It's Euros. http://www.chessbase.com/newsdetail... (Last paragraph). They played in Germany so it's only natural that way. |
|
| Dec-09-06 | | chessmoron: OK! Not bad...a extra $165,000. |
|
| Dec-09-06 | | lunacyfrog: "Kramnik is a bit stronger than Adams. Fritz is a good bit weaker than Hydra. And still the computer trounced him. This was a hard trouncing." What is this assessment based on? Facts? Logic? Anything??? Take away just one bad move and in all probability the match is drawn. How exactly is that a trouncing? Thats an uphill argument, if ive ever seen one. Tragedy for Kramnik that he played the worst move of his career? Yes. Trouncing? I don't think so. If Kramnik gives me a mate in one I will take it too, that didn't take all that much on Fritz's part. Out of the six games Fritz was able to demonstrate superiority in only one: the last game in which a draw was as bad as a loss for Kramnik in any case. And I don't know about other people on here, but I don't have a bias towards Kramnik. Did you see me on here taking Kramnik's side during the whole Elista nonsense? I don't think so. I suspect instead that a lot of people have a bias towards the computer programs because of their own inferiority complexes: they think if the grandmasters are beaten by programs it somehow brings the gms down to their level. Seems kind of sad to me. |
|
| Dec-09-06 | | square dance: <lunacyfrog> you might want to read <sharpnova>'s profile before you bother responding. |
|
| Dec-09-06 | | lunacyfrog: "<lunacyfrog> Can you find the moves they played somewhere?" I haven't looked, but maybe if you could get a copy of game 2's audio commentary? That's where Seirawan discusses it. Also, there was someone on playchess during game 2 who was analysing the position with Rybka. Apparently Rybka had Kramnik nearly a pawn to the good at around move 28. Can anyone confirm this? |
|
| Dec-09-06 | | lunacyfrog: "Sometimes a tournament winner has a performance close to 2850, so the best humans aren't without chances." Didn't Morozevich have over a 2900 performance rating in some tournament in 2004? Not that Moro would be a very good candidate against computers... |
|
| Dec-09-06 | | lunacyfrog: "square dance: <lunacyfrog> you might want to read <sharpnova>'s profile before you bother responding." Oh. Thanks for the heads up. |
|
| Dec-10-06 | | su24: <barbababa: <lunacyfrog> Can you find the moves they played somewhere?> "After: 30.e3! Bc5 31.Kf3!,
the win is as clear as a sunny sky. White's King is making
a beeline to the b5-square, where the b6-pawn is captured
and the a-pawn is escorted to coronation. -- Seirawan" |
|
| Dec-10-06 | | barbababa: <su24><"After: 30.e3! Bc5 31.Kf3!,
the win is as clear as a sunny sky. White's King is making a beeline to the b5-square, where the b6-pawn is captured and the a-pawn is escorted to coronation. -- Seirawan"> I used that for my analysis but it didn't work against my Fritz 8. |
|
| Dec-10-06 | | su24: barbababa: <su24><"After: 30.e3! Bc5 <31.Kf3!, the win is as clear as a sunny sky. White's King is making a beeline to the b5-square, where the b6-pawn is captured and the a-pawn is escorted to coronation. -- Seirawan"> I used that for my analysis but it didn't work against my Fritz 8.> Works just fine against mine. |
|
| Dec-10-06 | | barbababa: <su24><Works just fine against mine.> Where does your Fritz make different move than mine? |
|
| Dec-11-06 | | actinia: I'm not giving up yet on humans in the man vs machine battle. I have three reasons for this.
1) we could allow humans to have one takeback per game. this would mainly take away the huge psychological advantage computers have - that you can't make a single bad move against them
2) we could allow humans to team up... then it would be men vs machine(s) but the interesting dynamics would still be there
3) computers still have an obvious weakness: not being able to sense danger past a certain number of moves. I don't think this weakness has been fully exploited yet. In this sense computers may fall for a line that no human ever would. |
|
| Dec-11-06 | | code13: "Didn't Morozevich have over a 2900 performance rating in some tournament in 2004? " Well Zsophia Polgar had a 2928 performance at Rome in 1989 when she was 14. So human players can have "spikes" of peak performance. |
|
| Dec-11-06 | | TrueBlue: mankind is OK, don't worry, it's just Kramnik who doesn't know how to play without going to the washroom. I would bet my house any day (if I had one) that Topalov can win against Fritz. |
|
Dec-11-06
 | | chancho: <Trueblue:I would bet my house any day (if I had one) that Topalov can win against Fritz.> Didn't Vesko get beat by Fritz 6 in 2004? Or was it Fritz 5? |
|
 |
|
|
< Earlier Kibitzing · PAGE 135 OF 136 ·
Later Kibitzing> |
Kibitzing
