Who is testing Genius 7?

[S.Taylor]

but I wonder how strong the engine is?

Disappointingly little different to Genius 6.5 from 1999.
It is easy enough to test using 2 PC's because its GUI supports Autoplayer
The playing strength reflects no engine development for the last 8 years.
Recently I played a match between G7 and Movei00_8_438 coupled with the S9 book. Movei had little trouble trouncing Genius
If you can get the big opening book that came with G6.5 it is still a useful engine to play against.
AK

Indeed and it plays roughly around 2200 Elo and in no way at the level of 2400 elo as some testers claims....

I do not know about comp-comp games but in comp-human games genius was clearly better than 2200.

Old Genius3 on old p90 even could beat kasparov 1.5:0.5 at 25 minutes per game and certainly was at least at the level of 2600 at 25 minutes/game.

I believe that genius is at least at the level of 2400 against humans
at 120/40 even after considering the fact that humans are relatively better at longer time control.

Uri

It's hard for me to forget genius 3, on a pentium 90mhz.
Not many people seem to realize what went on then. (It beat Kasparov by 1.5-0.5 at 25 min per game!!!). That was a LONG time ago. Fritz 3 seemed to me to have been getting most of the publicity at the time, however, even though it seemed clear that Genius 3 was stronger than Fritz 3.
Uri!
you still say, however, that today, even at real blitz (5 min per game..., MUCH more advanced for the human than 25 per game, AND on more recent hardware which could be unbelievably higher elo from that alone) an average 2000 player can beat Genius 3, which Kasparov couldn't even do at 25 min per game?

[Dr.Wael Deeb]

but I wonder how strong the engine is?

Disappointingly little different to Genius 6.5 from 1999.
It is easy enough to test using 2 PC's because its GUI supports Autoplayer
The playing strength reflects no engine development for the last 8 years.
Recently I played a match between G7 and Movei00_8_438 coupled with the S9 book. Movei had little trouble trouncing Genius
If you can get the big opening book that came with G6.5 it is still a useful engine to play against.
AK

Indeed and it plays roughly around 2200 Elo and in no way at the level of 2400 elo as some testers claims....

I do not know about comp-comp games but in comp-human games genius was clearly better than 2200.

Old Genius3 on old p90 even could beat kasparov 1.5:0.5 at 25 minutes per game and certainly was at least at the level of 2600 at 25 minutes/game.

I believe that genius is at least at the level of 2400 against humans
at 120/40 even after considering the fact that humans are relatively better at longer time control.

Uri

It's hard for me to forget genius 3, on a pentium 90mhz.
Not many people seem to realize what went on then. (It beat Kasparov by 1.5-0.5 at 25 min per game!!!). That was a LONG time ago. Fritz 3 seemed to me to have been getting most of the publicity at the time, however, even though it seemed clear that Genius 3 was stronger than Fritz 3.
Uri!
you still say, however, that today, even at real blitz (5 min per game..., MUCH more advanced for the human than 25 per game, AND on more recent hardware which could be unbelievably higher elo from that alone) an average 2000 player can beat Genius 3, which Kasparov couldn't even do at 25 min per game?

I agree with you in general Taylor....
I paly the newest Genius myself and I can say that a well prepared human player at the level of 2400 Elo will demolish it even at a long time control using a decent hardware....

[S.Taylor]

[I agree with you in general Taylor....
I paly the newest Genius myself and I can say that a well prepared human player at the level of 2400 Elo will demolish it even at a long time control using a decent hardware....

It surprises me that you say "even" at a long time control, as I thought that a human stands MUCH better chances against computer at long time controls. What does a computer need time for? Most of its work is done quite quick, whereas a human needs time like air to breath, to work things out and not make mistakes.

btw, maybe I exagerated/made some mistake. Uri believes that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers night removed.

But still, Kasparov vs Genius 3 on P90 at 25 min a game should have been MUCH MUCH easier for Kasparov to beat than the above.

And what about more recently when Kasparov could only get a tie in a public match against a computer, as did Kramnik soon after....same result, a tie, over a series of games.

Oh, I HAVE seen that computers CAN be overpowered if the human is very wary, but I havn't spent too much time on these things since Hiarcs 7 times. I used to get good results vs genius 2, on 386, 486, but genius 3 seemed to have been an inprovement, and wasn't falling in for the same play as I had seen success with with vs genius 2. Hiarcs seemed very seldom for me to get a win from, too. Although I DID once have a real brilliancy, but that was one of the rare occasions, and with SUCH a brilliancy too (in fact, it was a short time control too, and _I_ was in an almost dream state. [on faster computer than original pentiums]).

[S.Taylor]

[I agree with you in general Taylor....
I paly the newest Genius myself and I can say that a well prepared human player at the level of 2400 Elo will demolish it even at a long time control using a decent hardware....

It surprises me that you say "even" at a long time control, as I thought that a human stands MUCH better chances against computer at long time controls. What does a computer need time for? Most of its work is done quite quick, whereas a human needs time like air to breath, to work things out and not make mistakes.

btw, maybe I exagerated/made some mistake. Uri believes that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers night removed.

But still, Kasparov vs Genius 3 on P90 at 25 min a game should have been MUCH MUCH easier for Kasparov to beat than the above.

And what about more recently when Kasparov could only get a tie in a public match against a computer, as did Kramnik soon after....same result, a tie, over a series of games.

Oh, I HAVE seen that computers CAN be overpowered if the human is very wary, but I havn't spent too much time on these things since Hiarcs 7 times. I used to get good results vs genius 2, on 386, 486, but genius 3 seemed to have been an inprovement, and wasn't falling in for the same play as I had seen success with with vs genius 2. Hiarcs seemed very seldom for me to get a win from, too. Although I DID once have a real brilliancy, but that was one of the rare occasions, and with SUCH a brilliancy too . hmmm, a highly sacrificial sequence... I'm normally wrong about these things, but this time I was right!!!

[Uri Blass]

[I agree with you in general Taylor....
I paly the newest Genius myself and I can say that a well prepared human player at the level of 2400 Elo will demolish it even at a long time control using a decent hardware....

It surprises me that you say "even" at a long time control, as I thought that a human stands MUCH better chances against computer at long time controls. What does a computer need time for? Most of its work is done quite quick, whereas a human needs time like air to breath, to work things out and not make mistakes.

btw, maybe I exagerated/made some mistake. Uri believes that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers night removed.

But still, Kasparov vs Genius 3 on P90 at 25 min a game should have been MUCH MUCH easier for Kasparov to beat than the above.

And what about more recently when Kasparov could only get a tie in a public match against a computer, as did Kramnik soon after....same result, a tie, over a series of games.

Oh, I HAVE seen that computers CAN be overpowered if the human is very wary, but I havn't spent too much time on these things since Hiarcs 7 times. I used to get good results vs genius 2, on 386, 486, but genius 3 seemed to have been an inprovement, and wasn't falling in for the same play as I had seen success with with vs genius 2. Hiarcs seemed very seldom for me to get a win from, too. Although I DID once have a real brilliancy, but that was one of the rare occasions, and with SUCH a brilliancy too . hmmm, a highly sacrificial sequence... I'm normally wrong about these things, but this time I was right!!!

Knight is a lot in chess.

I believe that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers knight removed.

It does not mean that 2400 can score 50% against genius in equal conditions.

Maybe there is some weakness of genius that they can learn to use but
assuming that they did not do special preperation against Genius I expect Genius on good hardware to get clearly more than 2400 fide rating.

Uri

[Dr.Wael Deeb]

[I agree with you in general Taylor....
I paly the newest Genius myself and I can say that a well prepared human player at the level of 2400 Elo will demolish it even at a long time control using a decent hardware....

It surprises me that you say "even" at a long time control, as I thought that a human stands MUCH better chances against computer at long time controls. What does a computer need time for? Most of its work is done quite quick, whereas a human needs time like air to breath, to work things out and not make mistakes.

btw, maybe I exagerated/made some mistake. Uri believes that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers night removed.

But still, Kasparov vs Genius 3 on P90 at 25 min a game should have been MUCH MUCH easier for Kasparov to beat than the above.

And what about more recently when Kasparov could only get a tie in a public match against a computer, as did Kramnik soon after....same result, a tie, over a series of games.

Oh, I HAVE seen that computers CAN be overpowered if the human is very wary, but I havn't spent too much time on these things since Hiarcs 7 times. I used to get good results vs genius 2, on 386, 486, but genius 3 seemed to have been an inprovement, and wasn't falling in for the same play as I had seen success with with vs genius 2. Hiarcs seemed very seldom for me to get a win from, too. Although I DID once have a real brilliancy, but that was one of the rare occasions, and with SUCH a brilliancy too . hmmm, a highly sacrificial sequence... I'm normally wrong about these things, but this time I was right!!!

Knight is a lot in chess.

I believe that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers knight removed.

It does not mean that 2400 can score 50% against genius in equal conditions.

Maybe there is some weakness of genius that they can learn to use but
assuming that they did not do special preperation against Genius I expect Genius on good hardware to get clearly more than 2400 fide rating.

Uri

Exactly my point Uri,a well prepared player particulary against Genius will defeat it without problems....

[Dr.Wael Deeb]

[I agree with you in general Taylor....
I paly the newest Genius myself and I can say that a well prepared human player at the level of 2400 Elo will demolish it even at a long time control using a decent hardware....

It surprises me that you say "even" at a long time control, as I thought that a human stands MUCH better chances against computer at long time controls. What does a computer need time for? Most of its work is done quite quick, whereas a human needs time like air to breath, to work things out and not make mistakes.

btw, maybe I exagerated/made some mistake. Uri believes that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers night removed.

But still, Kasparov vs Genius 3 on P90 at 25 min a game should have been MUCH MUCH easier for Kasparov to beat than the above.

And what about more recently when Kasparov could only get a tie in a public match against a computer, as did Kramnik soon after....same result, a tie, over a series of games.

Oh, I HAVE seen that computers CAN be overpowered if the human is very wary, but I havn't spent too much time on these things since Hiarcs 7 times. I used to get good results vs genius 2, on 386, 486, but genius 3 seemed to have been an inprovement, and wasn't falling in for the same play as I had seen success with with vs genius 2. Hiarcs seemed very seldom for me to get a win from, too. Although I DID once have a real brilliancy, but that was one of the rare occasions, and with SUCH a brilliancy too . hmmm, a highly sacrificial sequence... I'm normally wrong about these things, but this time I was right!!!

Todays long time controls are also a big advantage for the chess engines especially when it comes to the ones using multiprocessoring....
You did watch Kramnik playing Fritz,right?,do you still think that the computers can be overpowered just like that?
Personaly i don't think so....

[S.Taylor]

I mean I didn't think so. But I do admit that there MIGHT still be that possibility, but I had thought those days were already finished

[Dann Corbit]

[I agree with you in general Taylor....
I paly the newest Genius myself and I can say that a well prepared human player at the level of 2400 Elo will demolish it even at a long time control using a decent hardware....

It surprises me that you say "even" at a long time control, as I thought that a human stands MUCH better chances against computer at long time controls. What does a computer need time for? Most of its work is done quite quick, whereas a human needs time like air to breath, to work things out and not make mistakes.

btw, maybe I exagerated/made some mistake. Uri believes that humans close to as low as 2000 have good chances against the best of computers on best hardware at 30 min a game, with the computers night removed.

But still, Kasparov vs Genius 3 on P90 at 25 min a game should have been MUCH MUCH easier for Kasparov to beat than the above.

And what about more recently when Kasparov could only get a tie in a public match against a computer, as did Kramnik soon after....same result, a tie, over a series of games.

Oh, I HAVE seen that computers CAN be overpowered if the human is very wary, but I havn't spent too much time on these things since Hiarcs 7 times. I used to get good results vs genius 2, on 386, 486, but genius 3 seemed to have been an inprovement, and wasn't falling in for the same play as I had seen success with with vs genius 2. Hiarcs seemed very seldom for me to get a win from, too. Although I DID once have a real brilliancy, but that was one of the rare occasions, and with SUCH a brilliancy too . hmmm, a highly sacrificial sequence... I'm normally wrong about these things, but this time I was right!!!

Todays long time controls are also a big advantage for the chess engines especially when it comes to the ones using multiprocessoring....
You did watch Kramnik playing Fritz,right?,do you still think that the computers can be overpowered just like that?
Personaly i don't think so....

It is true that computers can beat the best players at slow time control.
However, they do even better at faster time controls.
Look at computer Elo on chess lists like FICS and the like for blitz verses bullet verses standard and you will see that the faster the game, the higher the Elo.
The only arena where humans stand a chance against the strongest programs now is in correspondence chess.

[Olivier Deville]

Who is testing Richard Lang's Genius 7? The latest ChessGenius Classic is compatible with Windows Vista, reads CBF files, etc., but I wonder how strong the engine is. http://www.chessgenius.com/pc/index.html

Hi Marek

Genius 7 did not do too well in my last C division : http://loirechecs.chez-alice.fr/chesswa ... 011CSt.htm

I use Auto232 to play the games (Genius 6.5 GUI needed).

Olivier