Don Dailey you need to teach underpromotion to Komodo!

[hgm]

You were not specific in your statement , you should have said "therefore the experimental version of Komodo which Don Dailey is working on can NOT see it when it plays as Black"

No, when we are discussing posted output of an explicitly specified Komodo version, and I refer to that output, saying that Komodo printed it, there is no need to specify that version again. People who read with understanding would think that obvious.

In fact that there is never any need to specify it; if I say 'Komodo sees', and there exists a single Komodo version that indeed sees it, I am already right. Only people who don't understand what was said, and therefor don't realize their mistake when quoting it out of context alters the meaning, and then try to throw in some warped logic to exclude the Komodo versions for which it was true as being true Komodo versions, would fail to see that...

[pichy]

You were not specific in your statement , you should have said "therefore the experimental version of Komodo which Don Dailey is working on can NOT see it when it plays as Black"

No, when we are discussing posted output of an explicitly specified Komodo version, and I refer to that output, saying that Komodo printed it, there is no need to specify that version again. People who read with understanding would think that obvious.

In fact that there is never any need to specify it; if I say 'Komodo sees', and there exists a single Komodo version that indeed sees it, I am already right. Only people who don't understand what was said, and therefor don't realize their mistake when quoting it out of context alters the meaning, and then try to throw in some warped logic to exclude the Komodo versions for which it was true as being true Komodo versions, would fail to see that...

The point is you specifically said Komodo and if a new person who just read this for the first time will assume that the available Komodo can see it too

[Don]

Well, that is what Marco has been explaining to you the whole time: setting up the trap is a blunder, DTM-wise, because it gets you checkmated very quickly (if the opponent does not let himself be suckered into it). Komodo of course sees this, so it prefers to avoid the futile trap, and play a better move in stead.

It is Not a better move, since you never know what your opponent is thinking at that time and you have to play for a chance of your opponent going wrong, and I prefer to set up the known trap then to take a route which will lead to losing after an agonizing long time playing

I think you are nit-picking here and being a bit ridiculous. In the general case it's just bad advice to play for cheap-shots and traps. In the best case you get a win but on average you will lose more than you win playing this way.

As far as I know every chess program assumes the opponent will play the best move, unless there is a bug in the search. You try to cast that as some kind of serious weakness which it isn't, but even if you could do opponent modeling without any risk, the benefit's are marginal at best, and at the levels computers play at I would suspect that it's immeasurable. Maybe it's a good strategy for playing beginners and other weak players but not for 3000+ plus players.

[pichy]

Well, that is what Marco has been explaining to you the whole time: setting up the trap is a blunder, DTM-wise, because it gets you checkmated very quickly (if the opponent does not let himself be suckered into it). Komodo of course sees this, so it prefers to avoid the futile trap, and play a better move in stead.

It is Not a better move, since you never know what your opponent is thinking at that time and you have to play for a chance of your opponent going wrong, and I prefer to set up the known trap then to take a route which will lead to losing after an agonizing long time playing

I think you are nit-picking here and being a bit ridiculous. In the general case it's just bad advice to play for cheap-shots and traps. In the best case you get a win but on average you will lose more than you win playing this way.

As far as I know every chess program assumes the opponent will play the best move, unless there is a bug in the search. You try to cast that as some kind of serious weakness which it isn't, but even if you could do opponent modeling without any risk, the benefit's are marginal at best, and at the levels computers play at I would suspect that it's immeasurable. Maybe it's a good strategy for playing beginners and other weak players but not for 3000+ plus players.

You are correct in saying that when engine versus engine are pitted they will always play the best move and they won't fall for a trap that can be detected in the next move such as Rc4+.

PS: We might be a small percentage of players that like to play versus great engine like your instead of pitting it vs other engines, but I demand the very best and that is why I pointed out the underpromotion. Not because it is a weakeness which will NOT ad any additional Extra Elo points, but when we purchase great engines we expect the very best like Houdini. And now with the fixed in underpromotion your engine and others deserve to be in the rank of 3000+

[pichy]

Well, that is what Marco has been explaining to you the whole time: setting up the trap is a blunder, DTM-wise, because it gets you checkmated very quickly (if the opponent does not let himself be suckered into it). Komodo of course sees this, so it prefers to avoid the futile trap, and play a better move in stead.

It is Not a better move, since you never know what your opponent is thinking at that time and you have to play for a chance of your opponent going wrong, and I prefer to set up the known trap then to take a route which will lead to losing after an agonizing long time playing

I think you are nit-picking here and being a bit ridiculous. In the general case it's just bad advice to play for cheap-shots and traps. In the best case you get a win but on average you will lose more than you win playing this way.

As far as I know every chess program assumes the opponent will play the best move, unless there is a bug in the search. You try to cast that as some kind of serious weakness which it isn't, but even if you could do opponent modeling without any risk, the benefit's are marginal at best, and at the levels computers play at I would suspect that it's immeasurable. Maybe it's a good strategy for playing beginners and other weak players but not for 3000+ plus players.

You are correct in saying that when engine versus engine are pitted they will always play the best move and they won't fall for a trap that can be detected in the next move such as Rc4+.

PS: We might be a small percentage of players that like to play versus great engine like your instead of pitting it vs other engines, but I demand the very best and that is why I pointed out the underpromotion. Not because it is a weakeness which will NOT ad any additional Extra Elo points, but when we purchase great engines we expect the very best like Houdini. And now with the fixed in underpromotion your engine and others deserve to be in the rank of 3000+. And I don't ask for too much in an engine which is over 3000+, but at least your next version should be able to find the best move Bf8+ in the position underneath

[d]6R1/8/2pB3k/2P4p/5p1q/5P2/4P1K1/8 w - - 0 1

[Arpad Rusz]

Try to solve this study:

Á.Rusz, Magyar Sakkvilág, 2010, 3rd Pr. (version)
[d]8/2P3PR/8/2RB1r2/7p/8/p6r/k1K5 w - - 0 1
White wins

Here's the solution:

After 1. Bc4! Rff2! (1... Rxc5 2. g8=Q +-) 2. Rxh4! Rxh4 (2... Rc2+ 3. Kd1 Rhd2+ 4. Ke1 Rg2 5.Bf1! Rxc5 6. Bxg2 Kb2 7. Ra4 +-) 3. Bb3 the following position is reached:

[d]8/2P3P1/8/2R5/7r/1B6/p4r2/k1K5 b - - 0 3

We have three lines starting from here (A, B, C):

A. (My main line)
3...Rhh2 4. Ra5!! (4. g8=Q? Rb2 5. Qg7 Rh1+ 6. Bd1 Rhh2 7. Bb3 (7. c8=Q Rhc2+ 8. Rxc2 stalemate) 7... Rh1+ 8. Bd1 Rhh2 positional draw) 4... Rh1+ 5. Bd1 Rh4! 6. Rxa2+! Rxa2

[d]8/2P3P1/8/8/7r/8/r7/k1KB4 w - - 0 7

7. c8=R! Phoenix (7. c8=Q? Rc4+! 8. Qxc4 Rc2+ 9. Kxc2 stalemate) 7... Rh3! (7... Rb4 8. Bc2 +-) 8. g8=R! Phoenix +- (8. g8=Q? Rc3+! 9. Rxc3 Rc2+! 10. Kxc2 stalemate)

B. (The computer's main line)
3... Rb2 4. Ra5! Rh1+ (4...Rhh2 5. Rxa2+! +-) 5. Bd1 Rb1+ 6. Kc2 Rh2+ 7. Kc3 Rg2 (7... Rc1+ 8. Kb4 Rg2 9. Rc5 Rxc5 10. Kxc5 Kb1 11. c8=Q a1=Q 12. Qf5+ Kc1 (12... Kb2 13.Qe5+ Kb1 14. Qe4+ +-) 13. Qf4+ Rd2 14. Qc4+ Kxd1 15.Qf1+ +-) 8. Rc5 Rg3+ 9. Kd2 (9. Kd4? Rb4+! 10. Kd5 Rg5+ 11. Kc6 Rg6+ 12. Kd5 Rg5+ perpetual check) 9... Rb2+ 10. Bc2 Rxg7 (10...Rg2+ 11. Ke3 Rxg7 12. c8=Q +-) 11. c8=Q Rg2+ 12. Kd3 Rgxc2

[d]2Q5/8/8/2R5/8/3K4/prr5/k7 w - - 0 13

13. Rc3! (13.Rxc2? Rxc2 14. Kxc2/Qxc2 stalemate) 13... Rxc3+ (13...Rd2+ 14. Kc4 Kb1 15. Qf5+ +-) 14. Qxc3 Kb1 15. Qe1#

C. (An additional line)
3... Rh1+ 4. Bd1 Rh4 5.Rc2! (5. g8=Q? Rc4+! 6. Rxc4 Rc2+ 7. Kxc2 stalemate) 5... Rc4 6. g8=Q Rcxc2+ 7. Bxc2 Rxc2+ 8. Kd1 Rxc7 9. Qh8+ Kb1 10. Qb8+ +-

Of course the first main line was the reason for composing this study (showing a double Rook Phoenix: two Rooks are sacrificed but later they both "reborn" by minor promotion).
The second main line is also valuable (it was discovered during composition by computer analysis). Of course no engine would choose the suboptimal 13...Rxc3+(?!) instead of 13...Rd2+(!?) (which also loses), but it is perfectly right to choose it in the solution. (The main line of a win study must feature optimal moves only for white.)

[pichy]

Try to solve this study:

Á.Rusz, Magyar Sakkvilág, 2010, 3rd Pr. (version)
[d]8/2P3PR/8/2RB1r2/7p/8/p6r/k1K5 w - - 0 1
White wins

Here's the solution:

After 1. Bc4! Rff2! (1... Rxc5 2. g8=Q +-) 2. Rxh4! Rxh4 (2... Rc2+ 3. Kd1 Rhd2+ 4. Ke1 Rg2 5.Bf1! Rxc5 6. Bxg2 Kb2 7. Ra4 +-) 3. Bb3 the following position is reached:

[d]8/2P3P1/8/2R5/7r/1B6/p4r2/k1K5 b - - 0 3

We have three lines starting from here (A, B, C):

A. (My main line)
3...Rhh2 4. Ra5!! (4. g8=Q? Rb2 5. Qg7 Rh1+ 6. Bd1 Rhh2 7. Bb3 (7. c8=Q Rhc2+ 8. Rxc2 stalemate) 7... Rh1+ 8. Bd1 Rhh2 positional draw) 4... Rh1+ 5. Bd1 Rh4! 6. Rxa2+! Rxa2

[d]8/2P3P1/8/8/7r/8/r7/k1KB4 w - - 0 7

7. c8=R! Phoenix (7. c8=Q? Rc4+! 8. Qxc4 Rc2+ 9. Kxc2 stalemate) 7... Rh3! (7... Rb4 8. Bc2 +-) 8. g8=R! Phoenix +- (8. g8=Q? Rc3+! 9. Rxc3 Rc2+! 10. Kxc2 stalemate)

B. (The computer's main line)
3... Rb2 4. Ra5! Rh1+ (4...Rhh2 5. Rxa2+! +-) 5. Bd1 Rb1+ 6. Kc2 Rh2+ 7. Kc3 Rg2 (7... Rc1+ 8. Kb4 Rg2 9. Rc5 Rxc5 10. Kxc5 Kb1 11. c8=Q a1=Q 12. Qf5+ Kc1 (12... Kb2 13.Qe5+ Kb1 14. Qe4+ +-) 13. Qf4+ Rd2 14. Qc4+ Kxd1 15.Qf1+ +-) 8. Rc5 Rg3+ 9. Kd2 (9. Kd4? Rb4+! 10. Kd5 Rg5+ 11. Kc6 Rg6+ 12. Kd5 Rg5+ perpetual check) 9... Rb2+ 10. Bc2 Rxg7 (10...Rg2+ 11. Ke3 Rxg7 12. c8=Q +-) 11. c8=Q Rg2+ 12. Kd3 Rgxc2

[d]2Q5/8/8/2R5/8/3K4/prr5/k7 w - - 0 13

13. Rc3! (13.Rxc2? Rxc2 14. Kxc2/Qxc2 stalemate) 13... Rxc3+ (13...Rd2+ 14. Kc4 Kb1 15. Qf5+ +-) 14. Qxc3 Kb1 15. Qe1#

C. (An additional line)
3... Rh1+ 4. Bd1 Rh4 5.Rc2! (5. g8=Q? Rc4+! 6. Rxc4 Rc2+ 7. Kxc2 stalemate) 5... Rc4 6. g8=Q Rcxc2+ 7. Bxc2 Rxc2+ 8. Kd1 Rxc7 9. Qh8+ Kb1 10. Qb8+ +-

Of course the first main line was the reason for composing this study (showing a double Rook Phoenix: two Rooks are sacrificed but later they both "reborn" by minor promotion).
The second main line is also valuable (it was discovered during composition by computer analysis). Of course no engine would choose the suboptimal 13...Rxc3+(?!) instead of 13...Rd2+(!?) (which also loses), but it is perfectly right to choose it in the solution. (The main line of a win study must feature optimal moves only for white.)

As you know by now Computers might have reached over 3000+ mainly due to the extraordinary calculating capability, but in a position like this where the creativity and long term visualization of the final position is required nothing and I repeat nothing can beat the strongest computer ever created (human brain)

[ernest]

As you know by now Computers might have reached over 3000+ mainly due to the extraordinary calculating capability, but in a position like this where the creativity and long term visualization of the final position is required nothing and I repeat nothing can beat the strongest computer ever created (human brain)

No doubt this is one of the most profound sentences I have read in this Forum...

[Don]

As you know by now Computers might have reached over 3000+ mainly due to the extraordinary calculating capability, but in a position like this where the creativity and long term visualization of the final position is required nothing and I repeat nothing can beat the strongest computer ever created (human brain)

No doubt this is one of the most profound sentences I have read in this Forum...

If you are on this forum long enough you will periodically see someone come along who has just discovered what has been known since the earliest days of computer chess and for some reason they think nobody else has figured it out except for them. I don't wish to embarrass him, but this is getting rather old - and he doesn't have anything to say but he thinks he does.

[pichy]

As you know by now Computers might have reached over 3000+ mainly due to the extraordinary calculating capability, but in a position like this where the creativity and long term visualization of the final position is required nothing and I repeat nothing can beat the strongest computer ever created (human brain)

No doubt this is one of the most profound sentences I have read in this Forum...

If you are on this forum long enough you will periodically see someone come along who has just discovered what has been known since the earliest days of computer chess and for some reason they think nobody else has figured it out except for them. I don't wish to embarrass him, but this is getting rather old - and he doesn't have anything to say but he thinks he does.

I would be embarassed to say a statement like this and not be able to back it up.


"Computer outplay humans in endgames too. There are specific weaknesses computer have which sometimes show up, but it's not enough to prevent the large ELO superiority they have over humans."
_________________
It doesn't matter what he does, he will never amount to anything.
(Albert Einstein's teacher to his father, 1895)