Author: mike schoonover
Date: 05:45:52 12/14/02
Go up one level in this thread
2002-12-14 - A message of me to the programmers in the message
http://www.talkchess.com/forums/1/message.html?270568 (CCC, only with password)
and the message in the ChessBits forum:
http://forum.myphorum.de/read.php?f=1578&i=11017&t=11017 is a portion, which I
played recently against Fritz 8. Me charged that I won in such a way, as I
played only by the " horizon effect " of the computer. Is correct. I would like
nevertheless to take up and into the space place some aspects thereby their
programmer somewhat think! In the played portion I won quasi already almost from
the first course! It was not " Untervariante " of the Sicilian, but this
position would come with a very large probability on the board, as soon as the
program plays 1... c5, or? This version, I know already several years long, and
her thoroughly analyzed. Here with Tim crab are some such my killers: DEFENDING
HUMANITY'S to HONOR you programmers would like I now something to say: The
runner victim on f7 is not correct also if the black king goes into the center
of the board. Fact is that even then, in each of my portions, it gives for black
a refutation of the white play! If in such a way black in the correct instant
its quality returns (which is naturally easily said as done) is black better!
Nevertheless I could observe that programs in almost each version after the
runner victim on f7 and following lady chess on h5, then either Kg8 or g6 to
play, if I did not flash! That is practically so also correct, because black
protects its king and has enough material advantage, in order to win. Why
programs can such as Fritz therefore not also after the course 5. Bd2 late
likewise always Kg8 or g6 play? We summarize again: The runner victim on f7 is
not correct, also not if black after Qh5+ with the king into the center of the
board marches! In all versions the programs (Fritz at least) play " nevertheless
" soon Kg8 or g6, not if white this refined course 5. Bd2?! plays. There is a
practical difference between 5. Bd2 or another white course HOWEVER not! White
will win only then, even if the black king is later in the center of the board,
if black commits an error! White will always lose, if plays black exactly, even
if its king is in the board center! From it the following result for me results:
Black can pull his king in everyone of the versions shown (see examples) into
the center of the board without fear to have: if it plays exactly in such a way
it will win, if it plays falsely, then it will lose! Logical would therefore be:
it cannot black completely exactly (!) play then it should ALWAYS pull, thus
after g8 - or beforehand to g6 play its king after course Qh5+ in security,
according to which it would win with high security. Like this portion however
shows (here down again the notation in PGN), does not adhere the programs not to
the principles mentioned by me (they play once so, once differently - not
according to the mentioned principles!). In my eyes there programming weakness
is present, which one should not charge to me! It would be better, if the
programmers would think some over my records! Thanks. [ Event " Blitz:30 ' " ] [
Site "?" ] [ DATE " 2002.??.??" ] [ Round "?" ] [ White " Eduard, Nemeth " ] [
Black " Fritz 8 " ] [ ECO " B21 " ] [ Result " 1-0 " ] 1. e4 { 0 } c5 { 0 } 2.
d4 { 2 } cxd4 { 0 } 3. Bc4 { 3 } Nc6 { 0 } 4. a3 { 5 } Nf6 { - 0.84/11 37 } 5.
Bd2 { 4 } Nxe4 { - 2.00/12 54 } 6. Bxf7+ { 6 } Kxf7 { - 2.31/11 20 } 7. Qh5+ { 4
} Ke6 { - 1.88/12 107 } 8. Qg4+ { 3 } Kd5 { - 2.16/11 38 } 9. b4 { 94 } Ne5 { -
3.06/11 124 } 10th c4+ { 26 } dxc3 { - 2.84/13 51 } 11. Nxc3+ { 4 } Nxc3 { -
2.72/12 99 } 12. Qh3 { 30 } Nb5 { - 2.84/10 32 } 13. Bf4 { 66 } d6 { - 2.91/9 85
} 14. Rd1+ { 13 } Kc6 { - 2.44/10 33 } 15. Rc1+ { 14 } Kb6 { - 0.53/12 22 } 16.
Qe3+ { 63 } Ka6 { 0.28/12 25 } 17. a4 { 16 } b6 { 0.28/10 7 } 18. Qe4 { 31 } d5
{ 2.06/11 12 } 19th axb5+ { 11 } Kxb5 { 6.28/12 10 } 20. Qxe5 { 19 } Bg4 {
6.28/11 13 } 21. Qc3 { 18 } Qc8 { 6.37/11 3 } 22. Qd3+ { 5 } Qc4 { 6.75/13 77 }
23. Rxc4 { 2 } dxc4 { 7.47/11 24 } 24. Qd5+ { 5 } Ka4 { 7.75/11 69 } 25. Qxa8 {
7 } 1-0 Eduard Nemeth
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.