Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Are these logfiles at IBM Website what Kaspy wanted?

Author: Jeremiah Penery

Date: 08:13:05 01/18/00

Go up one level in this thread


On January 18, 2000 at 09:51:41, Vincent Diepeveen wrote:

>On January 17, 2000 at 10:51:24, John Warfield wrote:
>
>>
>> Are these the long awaited Log files or printouts that Kasparov has been
>>begging for all these years? If so I wonder if they confirm his allegations of
>>cheating
>
>Yes. And after starting to analyse them i directly knew why they didn't
>want them to show.
>
>Deep Blue is a rather simple program people.
>It's evaluation is like Frans Morsch said so long extremely anti-kasparov.

If Diep was being programmed specifically to play Kasparov, how else would you
make the evaluation? :)

> Move  Kasparov           Deep-Blue
>----  ----------------   ----------------
>  1.  Nf3      (53:59)   d5       (0:19)
>  2.  g3       (0:07)    Bg4      (0:11)
>  3.  b3       (3:58)    Nd7      (2:24)
>
>Now here deep blue already wants to play e6 initially.
>Only at the last seconds it goes to the move Nd7
>
>  4.  Bb2      (0:11)    e6       (3:16)
>
>We see from ICC time division clearly that moves are
>more or less 3 minutes a move.
>Nothing wrong with that.
>
>-->  3. ..   Nd7 <-- 37/116:40
>---------------------------------------
>Guessing Bb2
> 3(4)[e5](-30)[e5](-30) -30v T=0
>Pe7e5 nf3e5P Nd7e5n bb2e5N Qd8e7 pd2d4 Pf7f6 be5c7P Qe7c7b nb1c3 Qc7c3n qd1d2
>Qc3d2q
> 3(5) -11  T=0
>Ng8f6 bf1g2 Pe7e6 ph2h3 Bg4f3n pe2f3B Bf8c5 o-o O-O
> 4(5) -11  T=0
>Ng8f6 bf1g2 Pe7e6 ph2h3 Bg4f3n pe2f3B Bf8c5 o-o O-O
> 5(5)[Ngf6](-1) -1  T=0
>Ng8f6 bf1g2 Pe7e6 ph2h3 Bg4f3n pe2f3B Bf8c5 o-o O-O
> 6(5)[Ngf6](-10)[e6](-2) -2  T=1
>Pe7e6 bf1g2 Bg4f3n pe2f3B Ng8f6 o-o Bf8d6 nb1c3
> 7(5) #[e6](28)[e6](28) 28^ T=2
>Pe7e6 bf1g2 Bg4f3n pe2f3B Ng8f6 o-o Bf8d6 nb1c3
> 7(6) #[e6](28)############################## 28  T=4
>Pe7e6 bf1g2 Bg4f3n pe2f3B Ng8f6 o-o Bf8d6 nb1c3
> 8(6) #[e6](21)############################## 21  T=10
>Pe7e6 bf1g2 Bg4f3n pe2f3B Ng8f6 o-o Bf8d6 nb1c3
> 9(6) #[e6](24)############################## 24  T=20
>Pe7e6 bf1g2 Ng8f6 ph2h3 Bg4f3n pe2f3B Bf8c5 o-o O-O nb1c3
>10(6)<ch> 'bb2'
> #[e6](22)############################## 22  T=87
>Pe7e6 bf1g2 Ng8f6 ph2h3 Bg4f3n pe2f3B Bf8c5 pd2d4 Bc5d6 o-o O-O
>11(6)[TIMEOUT] 22  T=164
>Pe7e6
>
>We see that the man behind this machine can't play chess at all. He wants
>Deep Blue's moves in uppercase. He let's his machine display opponents
>pieces in lower case.

This is another case where you type before you know what's going on.  Black
pieces are capitalized, and white pieces are lower-case, in both the source
piece and the captured piece.  See the even-numbered games where DB's pieces are
lower-case.

>I wouldn't want to mix them, unless i have problems seeing what move is
>an opponents move and what move is not. Anyway this is a problem more people
>have actually.
>
>Further it shows what it captures after a capture move, find that funny.
>Probably to show when recapture extensions get triggered.

Their move output format has nothing to do with anything...
I think Diep's move output format is silly, too, but that doesn't make me think
it is a weak program.
>
>Enfin, perhaps a matter of taste. Let's not get into details too much
>about this.

Because it has exactly zero real meaning.

>Let's look to its output and moves it makes!

Yes, let's.

>This is gonna be a bigtime continued mail, when i get home i'll continue
>with the silly moves made by deep blue and what it prints out in the logs,
>this is gonna get cool people!
>
>I am looking forward to the h6??? beginners move and missing of e5.

Which game(s) are those ones from?

I'm interested to see any analyses you make.



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.