Houdini wrote:That's an interesting perspective, I've made an entry on the Houdini Facebook page about it.
From Shredder 7's point of view, Houdini 3 is playing virtually perfect chess.
Will it be possible to sustain the 50 Elo/year software improvement over the next 10 years?
Robert
lets say you get a 50 elo/year software improvement over the next 10 years
what is the most you can hope for against a world champion
nearly always a pawn up by move 20/22 ?
what handicap can it offer a world champion yet still win?
Laskos wrote:The beast of the early 2003, undisputed SSDF leader, undisputed Playchess Engine Room favorite Shredder 7.04 stands up against the current leader:
[D]q7/5pkp/3p4/p2P3r/5P2/P1Q5/1PP5/2KR4 b - - 0 29[/D]
Observing the games, Houdini outplays Shredder in the opening, if not finished, in the middlegame, and if not finished, then surely in the endgame. Tremendous achievement in these less than 10 years.
I suggest you try a match between the same opponents under the same conditions, but at knight odds! Houdini always gets White but remove b1 knight. You need some opening book of course to get variety, so may use just a few moves of any normal book, omitting lines where the b1 knight moves. I think the match will be competitive. It seems ridiculous to talk about giving knight odds to an engine that was more or less on a par with players like Kasparove and Kramnik, but this is bullet chess, and the engines don't know to simplify just because they are up a piece, or to avoid unnecessary complications.
modolief wrote:Kai, I'm very fascinated with this post and am wondering if you would be willing to run another experiment. Jonathan Lee had replied that "If you did 20 years of Computer Chess, try Mephisto Genius version 2 or version 3." Would you be willing to run 100 or maybe 300 games of Shredder 7.04 against one of those chess programs?
Though I do have to add the caveats that I'm not sure how hard those versions are to find and whether or not it's a fair comparison because that software might not take full advantage of the hardware that it's run on whereas Shredder 7.04 may. Of course a match between Houdini 3 and Genius version 2 or 3 could be interesting, but that could take on the order of 10,000 games to produce a draw!!
One smaller difficulty is that Mephisto Genius 2 needs DOSBox to run. The bigger difficulty is that I cannot play games in automatic mode, I need to input the moves by hand, which is unrealistic.
Laskos wrote:The beast of the early 2003, undisputed SSDF leader, undisputed Playchess Engine Room favorite Shredder 7.04 stands up against the current leader:
[D]q7/5pkp/3p4/p2P3r/5P2/P1Q5/1PP5/2KR4 b - - 0 29[/D]
Observing the games, Houdini outplays Shredder in the opening, if not finished, in the middlegame, and if not finished, then surely in the endgame. Tremendous achievement in these less than 10 years.
I suggest you try a match between the same opponents under the same conditions, but at knight odds! Houdini always gets White but remove b1 knight. You need some opening book of course to get variety, so may use just a few moves of any normal book, omitting lines where the b1 knight moves. I think the match will be competitive. It seems ridiculous to talk about giving knight odds to an engine that was more or less on a par with players like Kasparove and Kramnik, but this is bullet chess, and the engines don't know to simplify just because they are up a piece, or to avoid unnecessary complications.
Nice idea, 100 games 1'+1'', 10 positions with knight on b1 removed for white, which is always Houdini 3.
Houdini 3 vs Shredder 7.04 knight odds:
+59 -25 =16
Houdini 3 scores 67%
+123 Elo points
Even at knight odds Houdini 3 beats Shredder 7.04 at bullet TC.
Sigh ... yes of course, running a DOSBox for something like this would be pretty ridiculous. Well, perhaps not as a 100% correct comparison, but what about versus Chess Genius 7.2? This might be a reasonable comparison to 1993 because, according to wikipedia, "It has been suggested that the reason the current PC version of ChessGenius (7.2) is marketed as 'ChessGenius Classic' is that apart from speed and efficiency enhancements, and updates to its openings book, the program has not changed dramatically since 1995."
And, along the lines suggested by Larry Kaufman, perhaps it would be fun to see ChessGenius 7.2 versus Houdini 3 at Knight (or Rook?) odds.
BTW, Larry, thanks! I have always been a big fan of your postings on talkchess and the rybka forum. What a great idea to try this again at Knight odds, and it is thrilling to see the result of that contest. Kai, can you make any comments as to what characterized Houdini's wins? Were they strategic crushes somehow? Was it because the opening play was so strong?
Anyway, thanks for the discussion, much appreciated.
