Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: more examples for search-based stupidity

Author: Thorsten Czub

Date: 09:00:58 06/12/98

Go up one level in this thread


Here is another example HOW EASY king attacks are, and how quiet they
happen. I guess humans would not OVERSEE and make these heavy mistakes.
But chess programs still make these major positional blunder moves.
Although no tactics is involved, the blunder-moves give away the game.


[Event "k6/200 60/60"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "CSystem Tal"]
[Black "Genius5"]
[Result "*"]

1. c4 c5 2. Nf3 Nc6 3. Nc3 g6 4. e3 {here CSTal was out of book} Nf6 5.
d4 cxd4 6. exd4 d5 7. cxd5 Nxd5 8.
Qb3 Nxc3{last bookmove of genius5} 9. Bc4 {+0,97} Nd5 10. Bxd5 e6 11.
Bxc6+ bxc6 12. O-O Qb6 13. Qc3 Bb4
14. Qc2
Ba6 15. Rd1 O-O 16. Ne5 {+1.25} Rfd8 17. Kh1 c5 {?? Whaaat ? c5 is
suicide. how can the bishop b4 defend the
hole on g7 ?!? Genius5 played this move with increasing evaluations and
scored +0,90} 18. a3 {cstal sees the mess
and evaluates +2,14} Ba5 19. Bg5 Rd5 20. Bf6 Qb5 {Genius is smelling
something. Much too late. Score goes up
and down. Fail low for Rc8 -1.33, Qb5 was played with -0,48. This is of
course too late.} 21.
Qc1 {cstal says +3,23} Bd8 22. Ng4 {+3,80} Qe8 23. dxc5 Rxd1+ 24. Qxd1
h5 25. Qf3 {+3,25}Rc8 26. Nh6+
{+3,75}Kh7 27.
Bxd8 Kxh6 28. Bh4 Kh7 29. b4 Bb5 30. Rd1 {3,90} *

The rest of the game is not important. White needs many many moves to
win the position. I want you to show how
easy Genius was outplayed due to the fact that simple things and patters
have been overseen.
There is NO  - really NO tactics in it. Just easy moves of
king-attacking.

In the same way Mr.Carstens killed Junior 4.6, he killed the other
Paris-champ.
Again - we don't know how many times he took back moves or played
alternatives. I don't care about this. We cannot prove that he betrayed.
Because we don't see what he hide. So we see only what the result is
(same problem with autoplayers, we don't see the games the autoplayer
did NOT play because it knows the line is a loss). We can replay the
result, but this does not give any evidence that the game was not a big
betrayel.
I hope some of you would one day understand this easy fact. And not
claim day over day that since we cannot prove betrayel, we can 100% say
there is no betray in xyz. Thats nonsense. We cannot prove ANYTHING in
this world. This is the fact. And the fact that a side has no evidence
does not show anything. It does not show that he guy/company is guilty ,
nor the opposite. But using these stuff about proving or not proving in
a discussion about morals is just ignoring the fact that we cannot prove
reality, because ANYthing in reality is IDEA.
Even today nobody knows how gravity works. There are lots of theories.
But no theorie works 100%. There is no gravity out there. And can be
proved.
World dows not work this way, and chess or computerchess is the same.
We cannot prove, what we cannot see. How do you want to prove that
Carstens faked the game by playing dozens of efforts before he got the
games ? You can't ! So how can you exclude he betrayed ? You can't. So -
how can you exclude the autoplayers cheat ? Only because you have no
evidence ? But than world is a grat betrayel since ANYTHING in the world
cannot be proved !


[Event "40/120 k6/200"]
[Site "?"]
[Date "1998.??.??"]
[Round "?"]
[White "Claus Carstens"]
[Black "Virtual2 "]
[Result "1-0"]
[ECO "D00"]
[Annotator "Czub,T"]
[PlyCount "47"]

{3072kB
} 1. d4 {0} 1... d5 {0.07/14 11} 2. e4 dxe4 3. Nc3 Nf6 4. f3 exf3 5.
Nxf3 5... Bg4 {This is the so called Teichmann-variation. Also not the
best
defense against BDG. Computers often play this defense because they like
the
bishop on g4 pinning the queen.   } 6. h3 Bxf3 7. Qxf3 c6 8. Qf2 e6 9.
Bg5 Nbd7
10. a3 10... Be7 {You see the pattern. Exactly the same structures as in
the
first game vs. Junior. Virtual2 plays the same moves, not understanding
1 % of
white's plan.  } 11. Qh4 O-O 12. Ne4 h6 13. Bd3 hxg5 14. Nxg5 g6 15. Qh6
Qe8
16. O-O Bd6 17. Rxf6 Nxf6 18. Rf1 Bf4 19. Rxf4 Qd7 20. Rh4 Qxd4+ 21.
Rxd4 Rfd8
22. Rh4 Nh5 23. Qh7+ Kf8 24. Qxf7# 1-0

Again: todays computer programs, no matter if world microcomputer
professional chess champion or
microcomputer chess champion. Also the today's machines cannot hide
their big weaknesses.
They have NO idea about positional-topics. And tactics is really not the
problem.
If all chess games are 100 % tactics (what is often told here) - where
is the tactics in those games here ?
And why were the programs unable to handle the topic ?
No - their problem is not tactics. And tactics is not the most common
CONTENT when it comes between human vs.
machine  (or cstal vs. machine ) games.

I would like to discuss about those things, instead of discussing about
FINDER positions and making silly lists
which program is faster than another finder. because we don't really
measure strength. We measure how fast their
searches are. This IS a PART of the strength, but most often not the
same as the real strength.

Please don't throw tactics and FINDING-key-moves in ONE melting-pot.
Also don't believe a "king-attack-position" that is found by
fast-searchers is really a king-attack-position.
Most often king-attacks are very quiet, and easy. They were NOT seen by
search. Gandalf and other programs have
specialized to find them without search. I think this is a good idea,
since humans (if not badly influenced by
"advisors" are still capable to attack ).




This page took 0.04 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.