Originally posted by vipiuIt's better than Crafty's 13-ply-deep (6.5 moves) 1. Re3 with '+81'. These problems are composed with ideas that lie in a 'deep' understanding of the position, and even when computers 'see' them, what would the use for them/us be, since they don't 'get' the ideas involved? For us, such problems are nice because of the ideas that are displayed [heavy material imbalance compensated by lack of moves of the opposite side], not because the computer cannot find the solution of it within one second.
yes, it is correct(actually this was also my solution weeks back so I suppose it is correct)
Originally posted by heinzkatwhat is the idea? 1 c6 Rb8 2 Nb5 gxh6 3 Khx somehow to force zwgzwang ? I do not understand 5Kh3 vs 5Kh1...
Segenreich 1989
[fen]k1r5/prp3pp/Pp5P/2P5/3N4/8/8/6K1[/fen]
Even on move 5 Crafty would suggest (20 halfmoves deep) 5. Kh3 with '-1002' while 5. Kh1 mates on move 12.
Originally posted by vipiu3 Khx? Yes, you somehow force zugzwang.
what is the idea? 1 c6 Rb8 2 Nb5 gxh6 3 Khx somehow to force zwgzwang ? I do not understand 5Kh3 vs 5Kh1...
I do not understand 5Kh3 vs 5Kh1...
Neither do I now, but I thought there was something with it, probably my mistake.
My fault; 5. Kh3 is actually the move that is given in the solution. 5. Kh1 seems to be working too, though. The confusion arises since the variation given is 5. Kh3 h4 6. Kxh4?? h5 and Black wins [7. Kh3 h4 8. Kh2 h3 9. Kh1 h2 10. Kxh2 Rh8+ 11. Kg2 Rh7 -+].
The solution as given by Laszlo Polgar's 'Chess Endgames' (with slight edits):
1. c6! +- Rbb8 2. Nb5 gxh6 3. Kh1!! [3. Kh2? h5 4. Kh3 h4 5. Kh2 h3 6. Kh1 (6. Kxh3 h6) 6. ... h2 7. Kh2 h5 8. Kh3 h4] 3. ... h5 4. Kh2 h4 5. Kh3 h6 6. Kh2 h3 7. Kxh3 h5 8. Kh4 Tb7 9. cxb7+ Kb8 10. Nxa7 [threat 11. Nc6#] Kxa7 11. bxc8=Q with 12. Qb7#
1 edit
Originally posted by SwissGambitRybka demo and ChessMaster 7000 could not find the mate in two. Fritz 5.32 did find it.
That's got to be a programming bug if Fritz 10 can't see it.
1.Re1! [zugzwang]1...Bg2 2.Qh4#A simple brute-force search should easily find this.
1...B~ 2.Qg1#
1...Kxe1 2.Qd2#
Edit: I just saw MadRook's post, and I think he has the answer. Fritz on "infinite analysis" does not always find the fastest mate. He is often content just t ...[text shortened]... at there [b]is a forced mate and leave it at that. I have Fritz 9 and it does this too.[/b]
Originally posted by vipiuI know this is off topic, but just wanted to say: I really don't understand the popularity behind crafty. it's not the ultimate standard for computer chess. there are many much stronger free engines out there.
what is the idea? 1 c6 Rb8 2 Nb5 gxh6 3 Khx somehow to force zwgzwang ? I do not understand 5Kh3 vs 5Kh1...
Originally posted by diskamylIt's like the Mona Lisa. Robert Hyatt created a masterpiece that everyone else can only try to copy. Yes, I agree it is not nearly one of the strongest freeware chess engines, but it has so many features that other engines don't have that you just keep coming back to it. hannotate, for example. It has an almost endless supply of options that other engines don't have -- or don't have published.
I know this is off topic, but just wanted to say: I really don't understand the popularity behind crafty. it's not the ultimate standard for computer chess. there are many much stronger free engines out there.
It is still a measuring stick. It may measure yards while some of the other engines measure chains and rods, but it is still an index.
2 edits
Originally posted by SwissGambitChessMaster 7000 does have mate in two and will brute force if I know what you are talking about. Rybka is in arena which would only download in German for me. I do not know what it will do. I did try ChessMaster in brute force and it failed. I think it had two hours a move and it came up with a mate in three in less than a second. P.S. ChessMaster found the mate in two when it was told to search for a mate in two. As soon as I told it to search for it it had the answer.
Do either of them have a 'brute-force' search option? Or mate solver?
Originally posted by gambit3Off topic, but I've never used ChessMaster, Rybka, and haven't had Fritz since the MS/DOS version. What features do any of these program has that are unique? Besides their rating -- I've read a lot about Rybka.
ChessMaster 7000 does have mate in two and will brute force if I know what you are talking about. Rybka is in arena which would only download in German for me. I do not know what it will do. I did try ChessMaster in brute force and it failed. I think it had two hours a move and it came up with a mate in three in less than a second.
I've seen several games in pgn files that were annoFritzed. I assume this is similar to annotate or hannotate with Crafty?
1 edit
Originally posted by petrovitchI cannot read the German so I do not know what Rybka can do in Arena. It can be downloaded into Fritz if you have the cd. The only thing that ChessMaster may offer over the other programs is that it has lots of computer generated opponents with their various playing styles. I think that some other programs have added chess lessons like ChessMaster has.
Off topic, but I've never used ChessMaster, Rybka, and haven't had Fritz since the MS/DOS version. What features do any of these program has that are unique? Besides their rating -- I've read a lot about Rybka.
I've seen several games in pgn files that were annoFritzed. I assume this is similar to annotate or hannotate with Crafty?
- Joined
- 21 Apr 06
- Moves
- 4211
1 edit
Quite a few engines get this position wrong too. It comes from Kasparov vs FritzX3D. The human idea in this position is to probably play f5 (with f4 ideas to generate counter play) then maybe Rf7 Nf8 Ne6 Nd8 and white has a very hard time breaking through. Even if the knight manover is debateable the move f5 isnt, best move by far. In the game Fritz played Nf6 then what can only be called a series of bad moves which allowed Kasparov to win.