Computerchess Misc (4)
Rolf Tueschen
against his computer. A few years ago I put the gamescores into a
collection of his games. In the case of the following score I was in doubt
whether my dad had White or Black. Could you find a way how to determine
the human side? In case it should be needed I could provide you with the
model of the computerprogram as well.
Here's the game.
1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Bd2 d5 5.e3 a6 6.Qa4+ Nc6 7.Be2 0-0 8.Ne4
Bxd2+ 9.Nxd2 Bd7 10.Qb3 Rb8 11.Ngf3 b5 12.cxb5 Na5 13.Qc3 c6 14.bxa6 Qb6
15.Rb1 Nc4 16.Bxc4 dxc4 17.a7 Qxa7 18.a3 c5 19.b4 cxb3 20.dxc5 Rfc8 21.Nxb3
Ne4 22.Qa1 Nxc5 23.Nxc5 Qa5+ 24.Nd2 Rxb1+ 25.Qxb1 Qxc5 26.Qd3 Bb5 27.Qb3
Qc1+ 28.Qd1 Qxa3 29.e4 Rc1 30.Nc4 Rxd1+ 31.Kxd1 Qa1+ 32.Kd2 Qxh1 33.Nd6 Bd7
34.g3 Qxh2 35.Ke3 Qh6+ 36.f4 Qg6 37.Kf3 h5 38.e5 Bc6+ 39.Kf2 Qc2+ 40.Ke3
Qg2 41.Nc8 Qxg3+ 42.Kd4 Qxf4+ 43.Kc5 Qxe5+ 44.Kxc6 Qc3+ 45.Kd7 Qd4+ 46.Ke7
Qf6+ 47.Ke8 h4 48.Ne7+ Qxe7+ 49.Kxe7 h3 50.Kd6 Kf8 51.Ke5 Ke7 52.Ke4 h2
53.Kf3 h1Q+ 54.Ke3 f5 55.Kd4 Qd5+ 56.Ke3 Qe4+ 57.Kd2 f4 58.Kc3 f3 59.Kb2
Qd3 60.Ka1 Qd2 61.Kb1 f2 62.Ka1 f1Q# 0-1
1. What colour did my father play?
2. What are the crucial positions for you to decide on?
Thanks for your help.
Rolf Tueschen
Robert Hyatt
: After my father's death I found a couple of final chessgames he had played
: against his computer. A few years ago I put the gamescores into a
: collection of his games. In the case of the following score I was in doubt
: whether my dad had White or Black. Could you find a way how to determine
: the human side? In case it should be needed I could provide you with the
: model of the computerprogram as well.
worthwhile idea, Rolf. One idea, if you have crafty, and assuming you put
the game in a file "game.dad"
annotate game.dat bw 1-999 .5 60
which will run thru the game, looking for any move where the score dropped
by a half-pawn or more. If only one side has such moves, you can probably
assume your dad played that side (assuming he wasn't a GM-class player,
of course). If both sides have blunders you might have to run it several
times and reduce that 60 (60 seconds per move may be too high and it might
find blunders that an older chess machine could have missed.)
good luck...
: Here's the game.
: 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Bd2 d5 5.e3 a6 6.Qa4+ Nc6 7.Be2 0-0 8.Ne4
: Bxd2+ 9.Nxd2 Bd7 10.Qb3 Rb8 11.Ngf3 b5 12.cxb5 Na5 13.Qc3 c6 14.bxa6 Qb6
: 15.Rb1 Nc4 16.Bxc4 dxc4 17.a7 Qxa7 18.a3 c5 19.b4 cxb3 20.dxc5 Rfc8 21.Nxb3
: Ne4 22.Qa1 Nxc5 23.Nxc5 Qa5+ 24.Nd2 Rxb1+ 25.Qxb1 Qxc5 26.Qd3 Bb5 27.Qb3
: Qc1+ 28.Qd1 Qxa3 29.e4 Rc1 30.Nc4 Rxd1+ 31.Kxd1 Qa1+ 32.Kd2 Qxh1 33.Nd6 Bd7
: 34.g3 Qxh2 35.Ke3 Qh6+ 36.f4 Qg6 37.Kf3 h5 38.e5 Bc6+ 39.Kf2 Qc2+ 40.Ke3
: Qg2 41.Nc8 Qxg3+ 42.Kd4 Qxf4+ 43.Kc5 Qxe5+ 44.Kxc6 Qc3+ 45.Kd7 Qd4+ 46.Ke7
: Qf6+ 47.Ke8 h4 48.Ne7+ Qxe7+ 49.Kxe7 h3 50.Kd6 Kf8 51.Ke5 Ke7 52.Ke4 h2
: 53.Kf3 h1Q+ 54.Ke3 f5 55.Kd4 Qd5+ 56.Ke3 Qe4+ 57.Kd2 f4 58.Kc3 f3 59.Kb2
: Qd3 60.Ka1 Qd2 61.Kb1 f2 62.Ka1 f1Q# 0-1
: 1. What colour did my father play?
: 2. What are the crucial positions for you to decide on?
: Thanks for your help.
: Rolf Tueschen
--
Robert Hyatt Computer and Information Sciences
hy...@cis.uab.edu University of Alabama at Birmingham
(205) 934-2213 115A Campbell Hall, UAB Station
(205) 934-5473 FAX Birmingham, AL 35294-1170
Rolf Tueschen
> If both sides have blunders you might have to run it several
>times and reduce that 60 (60 seconds per move may be too high and it might
>find blunders that an older chess machine could have missed.)
I think you gave me a good idea. Thanks.
But what if the actual monsters are still on zero-seconds-level too strong?
Just my idea remembering that you once made such a statement.
>good luck...
Robert Hyatt
: Robert Hyatt <hy...@crafty.cis.uab.edu> wrote:
:>good luck...
No idea... that's why you have to play with the time. However, one side
should have *far* more blunders than the other. Tht would be the only
hope. For an old computer you might also look for "computer-moves" like
Kh1-g1-h1 and so forth...
Rolf Tueschen
>No idea... that's why you have to play with the time. However, one side
>should have *far* more blunders than the other. Tht would be the only
>hope. For an old computer you might also look for "computer-moves" like
>Kh1-g1-h1 and so forth...
Here it was a program no later than 1984. Think it was an early version of
MEPHISTO. I e.g. couldn't believe that my dad should have played this a6-a7
move and then after it was taken a2-a3. I mean that is a whole tempo. So I
took him as Black. But then, he did not play a single game with the NIMZO
before. He always played his g6 KI.
I try to insist on you because I really fear that your actual remedy will
not work for that level of the time. Approximately max. 1600.
Any idea?
Robert Hyatt
: Robert Hyatt <hy...@crafty.cis.uab.edu> wrote:
:>No idea... that's why you have to play with the time. However, one side
:>should have *far* more blunders than the other. Tht would be the only
:>hope. For an old computer you might also look for "computer-moves" like
:>Kh1-g1-h1 and so forth...
: Here it was a program no later than 1984. Think it was an early version of
: MEPHISTO. I e.g. couldn't believe that my dad should have played this a6-a7
: move and then after it was taken a2-a3. I mean that is a whole tempo. So I
: took him as Black. But then, he did not play a single game with the NIMZO
: before. He always played his g6 KI.
: I try to insist on you because I really fear that your actual remedy will
: not work for that level of the time. Approximately max. 1600.
: Any idea?
None other than to look for what I would call "typical computer moves"
which were *very* common. IE go to the end of the game... did the opponent
(when losing) prolong mate by moves like Qxg7+ KxQ, Rxf7+, Kxf7??? That's
a classic computer type move that humans generally don't play...
IE classic "horizon effect" moves. If you see a piece trapped, check the moves
around that. Computers are known to toss pawns to "save" pieces (save ==
push the loss beyond the search horizon where it "doesn't exist").