Page 2 of 3
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Sun May 12, 2019 6:35 am
by Modern Times
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Sun May 12, 2019 7:53 am
by lkaufman
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
OK, so he must have chosen only positions where the king is NOT between the two rooks, figuring that then there is no great need to castle. Still, too different from normal chess for me. As for FRC, the main question is whether all 960 positions should be used, or just some subset (advocated by Kasparov), and if a subset what are the criteria for selection or elimination? The idea of choosing only from positions with king and at least one rook at home, with or without the option of FRC castling with a rook not at home, has some appeal to me; I'm not sure how many positions that gives. As long as castling on one side is possible, maybe that's enough to play by normal rules, allowing all engines to play with no special programming.
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Sun May 12, 2019 11:48 pm
by JohnWoe
lkaufman wrote: ↑Sun May 12, 2019 7:53 am
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
OK, so he must have chosen only positions where the king is NOT between the two rooks, figuring that then there is no great need to castle. Still, too different from normal chess for me. As for FRC, the main question is whether all 960 positions should be used, or just some subset (advocated by Kasparov), and if a subset what are the criteria for selection or elimination? The idea of choosing only from positions with king and at least one rook at home, with or without the option of FRC castling with a rook not at home, has some appeal to me; I'm not sure how many positions that gives. As long as castling on one side is possible, maybe that's enough to play by normal rules, allowing all engines to play with no special programming.
I consider standard chess as subset of Chess960. And shuffle chess also.
But on Winboard/XBoard there's this shuffle chess variant. Both rooks and a king are on their original squares. So 3 white squares for white bishop. 2 black squares for black bishop. etc. So 3(WB) x 2(BB) x 3(N) x 2(N) x 1(Q) = 36 positions any non-chess960 engine can( or not ? ) play. So Chess36. You can memorize so few positions but it's a good start.
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Mon May 13, 2019 4:37 am
by lkaufman
JohnWoe wrote: ↑Sun May 12, 2019 11:48 pm
lkaufman wrote: ↑Sun May 12, 2019 7:53 am
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
OK, so he must have chosen only positions where the king is NOT between the two rooks, figuring that then there is no great need to castle. Still, too different from normal chess for me. As for FRC, the main question is whether all 960 positions should be used, or just some subset (advocated by Kasparov), and if a subset what are the criteria for selection or elimination? The idea of choosing only from positions with king and at least one rook at home, with or without the option of FRC castling with a rook not at home, has some appeal to me; I'm not sure how many positions that gives. As long as castling on one side is possible, maybe that's enough to play by normal rules, allowing all engines to play with no special programming.
I consider standard chess as subset of Chess960. And shuffle chess also.
But on Winboard/XBoard there's this shuffle chess variant. Both rooks and a king are on their original squares. So 3 white squares for white bishop. 2 black squares for black bishop. etc. So 3(WB) x 2(BB) x 3(N) x 2(N) x 1(Q) = 36 positions any non-chess960 engine can( or not ? ) play. So Chess36. You can memorize so few positions but it's a good start.
But the two knights are interchangeable so it's only 18 positions, not 36. This is the chess18 I mentioned in an earlier post. I've already run some engine matches with it and noticed a low draw percentage.
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Mon May 13, 2019 10:58 am
by pohl4711
lkaufman wrote: ↑Sun May 12, 2019 7:53 am
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
OK, so he must have chosen only positions where the king is NOT between the two rooks, figuring that then there is no great need to castle. Still, too different from normal chess for me. As for FRC, the main question is whether all 960 positions should be used, or just some subset (advocated by Kasparov), and if a subset what are the criteria for selection or elimination? The idea of choosing only from positions with king and at least one rook at home, with or without the option of FRC castling with a rook not at home, has some appeal to me; I'm not sure how many positions that gives. As long as castling on one side is possible, maybe that's enough to play by normal rules, allowing all engines to play with no special programming.
Correct. I can remember, that the King was never between the rooks. But I dont have that set anymore, I built it long time ago...
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Mon May 13, 2019 11:03 am
by Modern Times
pohl4711 wrote: ↑Mon May 13, 2019 10:58 am
Correct. I can remember, that the King was never between the rooks. But I dont have that set anymore, I built it long time ago...
Well I have it if you want it
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Mon May 13, 2019 11:09 am
by pohl4711
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
How about trying my Drawkiller openings?!
https://www.sp-cc.de/drawkiller-openings.htm
They look like this: (Kings on opposite side of the board, Queens not on the same line, all non-pawns still on row 1/8. Only 5 of 16 non-pawns not on their normal position...
That gives very, very low draw rates...(nearly halved, compared to classical opening sets) take a look at the testing results on my Drawkiller-site. So, Drawkiller is much better, than FRC. And Drawkiller has much more different positions, than FRC.)
[d]knbrqbnr/ppp2pp1/4p2p/3p4/3P4/4PP2/PPP3PP/RNBQRBNK w - - 0 0
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Tue May 14, 2019 1:01 am
by carldaman
pohl4711 wrote: ↑Mon May 13, 2019 10:58 am
lkaufman wrote: ↑Sun May 12, 2019 7:53 am
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
OK, so he must have chosen only positions where the king is NOT between the two rooks, figuring that then there is no great need to castle. Still, too different from normal chess for me. As for FRC, the main question is whether all 960 positions should be used, or just some subset (advocated by Kasparov), and if a subset what are the criteria for selection or elimination? The idea of choosing only from positions with king and at least one rook at home, with or without the option of FRC castling with a rook not at home, has some appeal to me; I'm not sure how many positions that gives. As long as castling on one side is possible, maybe that's enough to play by normal rules, allowing all engines to play with no special programming.
Correct. I can remember, that the King was never between the rooks. But I dont have that set anymore, I built it long time ago...
This may be less relevant to the discussion, but I thought that FRC/chess960 required the K to be between the Rooks. Yes/no?
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Tue May 14, 2019 4:24 am
by lkaufman
carldaman wrote: ↑Tue May 14, 2019 1:01 am
pohl4711 wrote: ↑Mon May 13, 2019 10:58 am
lkaufman wrote: ↑Sun May 12, 2019 7:53 am
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
OK, so he must have chosen only positions where the king is NOT between the two rooks, figuring that then there is no great need to castle. Still, too different from normal chess for me. As for FRC, the main question is whether all 960 positions should be used, or just some subset (advocated by Kasparov), and if a subset what are the criteria for selection or elimination? The idea of choosing only from positions with king and at least one rook at home, with or without the option of FRC castling with a rook not at home, has some appeal to me; I'm not sure how many positions that gives. As long as castling on one side is possible, maybe that's enough to play by normal rules, allowing all engines to play with no special programming.
Correct. I can remember, that the King was never between the rooks. But I dont have that set anymore, I built it long time ago...
This may be less relevant to the discussion, but I thought that FRC/chess960 required the K to be between the Rooks. Yes/no?
That's the point, his set did NOT contain any chess960 positions, so king could NOT be between the rooks.
Re: Komodo 13.01 MCTS - chess960 results and site updated
Posted: Tue May 14, 2019 4:30 am
by lkaufman
pohl4711 wrote: ↑Mon May 13, 2019 11:09 am
Modern Times wrote: ↑Sun May 12, 2019 6:35 am
Yes FRC is the prime candidate.
The 50 positions was from the Rybka forum more than 10 years ago:
"Stefan Pohl has created a set of 50 shuffle chess positions, which can be used as an opening database for testing. There are NO castlings possible, and the set does NOT contain any Chess960 positions. 50 positions allow engine matches with 100 games (with switching sides per position)."
Anyway I stuck with chess960.
How about trying my Drawkiller openings?!
https://www.sp-cc.de/drawkiller-openings.htm
They look like this: (Kings on opposite side of the board, Queens not on the same line, all non-pawns still on row 1/8. Only 5 of 16 non-pawns not on their normal position...
That gives very, very low draw rates...(nearly halved, compared to classical opening sets) take a look at the testing results on my Drawkiller-site. So, Drawkiller is much better, than FRC. And Drawkiller has much more different positions, than FRC.)
[d]knbrqbnr/ppp2pp1/4p2p/3p4/3P4/4PP2/PPP3PP/RNBQRBNK w - - 0 0
The reason I don't like your Drawkiller openings (either for testing or to play myself) is that opposite-side castling is not the norm in chess. I'm sure they make for interesting games, but the emphasis on pawn-storms and direct attacks on the king makes it too different from normal chess for me. It's just not representative of standard chess, where positional play is critical. So far my tests indicate that the 18 positions from chess960 that allow normal castling on both sides are also much less drawish than normal chess, without the above objection. But, only 18 positions, too few!