Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Answers

Author: Robert Hyatt

Date: 16:22:13 02/16/03

Go up one level in this thread


On February 16, 2003 at 16:25:09, Amir Ban wrote:

>On February 16, 2003 at 12:10:35, Robert Hyatt wrote:
>
>>On February 16, 2003 at 07:59:54, Amir Ban wrote:
>>
>>>On February 15, 2003 at 13:06:55, Robert Hyatt wrote:
>>>
>>>>
>>>>
>>>>I disagree with the "played like a super-GM" player, however.  I doubt you
>>>>will find _any_ 2200 FIDE player that would play as badly as DJ played in
>>>>the first three games, up until move 30 or so.  Game 1 would not have been
>>>>played by any 2000 player I know, myself included.  So saying that it has
>>>>super-GM positional understanding is _way_ _way_ offbase.  Yes, it played
>>>>good moves at times.  But it also played _horrible_ moves at times.  And I
>>>>am not just talking about tactically horrible moves such as the blunders that
>>>>Kasparov dropped on the board, I am talking about moves such as taking the
>>>>g-pawn and getting exposed to a horrific attack.
>>>>
>>>
>>>I can't agree with any of this.
>>>
>>>It would be good to back the statement that Junior played the "first three
>>>games, up until move 30 or so" worse than 2200 with some concrete examples of
>>>where a 2200 player would play better. The three games lasted 27, 30 & 36 moves,
>>>so what does this mean at all ?
>>
>>
>>Take game 1.  I don't know of _anybody_ that would play like that, except
>>for some computers.  Totally lost.
>>
>
>This is handwaving and doesn't answer the question
>
>
>>Take game 2.  Every GM criticized the idea of "winning the exchange" instantly.
>>It took me (and other lowly humans) a lot longer to conclude "this looks very
>>dangerous for white, where prior to accepting we all thought white had a better
>>position.
>>
>>Take game 3.  Taking the g-pawn to open a file in your own king's face.  Did
>>you hear _any_ IM/GM player that thought that was a good move?  I didn't and
>>we had _several_ on ICC.
>>
>
>Why would they say it's a good move when it's the only move ? See more below.
>
>
>>So in the first three games, the program voluntarily self-destructed, even
>>though it did win a game due to a simple-to-see blunder by the opponent, and
>>it managed to draw game 2 where it could have lost, even though it was clearly
>>better in the opening.
>>
>>That's not to say that DJ didn't play well at times, but it did _not_ play
>>"super-GM position moves" in _those_ games.  Perhaps it did a few times.  But
>>a few times is not enough when playing a super-GM.
>>
>>>
>>>The one example you give, of 10... Nxg4 in game three, is wrong. Taking the pawn
>>>is the only move that does not lose quickly. I assume that what you prefer is
>>>what crafty would play, which I see is 10... h6. I don't know if this is
>>>apparent to a 2200 player, but 10... h6 11. g5 is hopeless for black. Crafty
>>>does not even expect 11. g5.
>>
>>No.  I suspect Crafty would do the same thing.  Notice that I didn't say that
>>I thought Crafty was better than DJ, because I _don't_.
>>
>
>You did not get my meaning:
>
>1. Nxg4 is the only move here
>2. Crafty doesn't play it but h6, a weak move
>
>Criticizing Nxg4 is more of a weak player's reflex than a serious comment.
>
>
>>
>>
>>>
>>>The picture you give of Kasparov missing won positions due to making "tactically
>>>horrible moves" against an opponent who shows tactical resilience (while playing
>>>like a positional patzer) simply did not happen in this match. Maybe you have
>>>been watching crafty on ICC, but not Deep Junior in NYC.
>>
>>
>>Sorry, but I watched _every_ game.  I didn't get to see all of the last three
>>games, but I watched games 1-3 from start to finish.  If you are happy with the
>>decisions it made, and if you are convinced it played like a super-GM in those
>>three games, that's up to you.  To my eye, and to the eyes of the GM/IM players
>>that were analyzing on ICC, that opinion was not prevalent.
>>
>>>
>>>Kasparov did not make any real blunders in this match, at least not the way I
>>>understand "blunder" as a move that he and much lesser players would in normal
>>>circumstances easily avoid. Kasparov's motives in describing his mistakes as
>>>"fingerfehlers" are obvious, since if they were so, then they somehow don't
>>>count and we have to count the games he lost as surely drawn, and those he drew
>>>as surely won, but we don't have to buy that.
>>
>>Certainly he did.  Game two comes to mind.  He had a forced draw.  He made
>>a move that led to a forced loss.  That is a blunder in _any_ book.  (Ng6+
>>was the forced draw, Rh5 was an outright blunder.)
>>
>>
>>>
>>>To describe 32. Rh5 of game 3 as a blunder is a gross misrepresentation. It
>>>misses a rather spectacular mate possibility. Not something that one sees in a
>>>blitz game (not even Eduard Nemeth).
>>
>>
>>No, but everyone saw Ng6+ instantly, and it only took a few seconds for GM
>>players to say "oh no" after Rh5.  (of course it took the comps a few milli-
>>seconds to see Rh5 was bad).
>>
>
>Did you actually look into the variation he missed ? Please play through it on a
>real chessboard, then come back here and tell me it's an "outright blunder" and
>the GM's would see it in a few seconds.

The definition of "blunder" is "a move that changes the final result of the
game."  IE from a win to a draw or loss, or from a draw to a loss.  This
_obviously_ qualifies.  Ng6 is an instant and forced draw unless black wants
to return the favor and turn it into a win for white.  So how can it _not_ be
a blunder to lose when a forced draw was there???

Not in any definition I see for the word "blunder".

>
>
>>
>>
>>>
>>>Calling 25... Qa1+ of game 2 a blunder is really stretching it. Kasparov, by his
>>>own words, worked it out to a forced win, but missed a rook check 18 ply down
>>>the road. This is not a blunder but a hard luck story. Anderssen's combination
>>>in the Evergreen Game was not as deep. Would we accept Dufrense saying "I was
>>>totally winning but blundered and allowed Rd1" ?
>>>
>>>Amir
>>
>>I'm not sure I'd call that a blunder by Kasparov.  Computers said Qa1 was an
>>instant draw.  While f4 was better for black.  Whether it would win or not is
>>unknown, and even Kasparov said that f4 also led to a draw, so that's perfectly
>>ok, in m book.
>>
>
>The computers said Qa1 was an instant draw ? Really ? You can leave a computer
>overnight and it still won't get close to resolving that position.

I wouldn't disagree, but yes, they _did_ say "draw".  And we had _several_
providing analysis.  Of course, whether it is right or wrong is another topic
and since Kasparov said that f4 was also a draw, that is why I did not say
Qa1 was a blunder.  It didn't change the game result.


>
>Ok, I believe that you followed the match, but you were looking through very
>peculiar glasses because everything you saw was simpler by several orders of
>magnitude than it actually was.
>


I was "looking through the glasses" of several GM and IM players, plus
listening to the chess.fm live commentary, plus watching the output of several
different computer programs, most commercial but a few amateurs including
Crafty...  I don't think that was "simpler by several orders of magnitude
than it actually was."  I don't try to out-guess GM players very often, but I
do listen to others do so, particularly when computer analysis backs up what
they are saying (Ng6+ vs Rh5 is but one example.)






>
>>Even playing Bxh7 rather than g3 in game 5 was ok, because he gave reasonable
>>justification (avoiding incredibly complicated tactics) for avoiding what was
>>probably a win for white (according to recently published GM analysis, although
>>the "final story" is not yet in...).
>>
>
>It will be analyzed for quite some time more I guess, but we already have from
>Kasparov at the closing ceremony and later that he believes it is sound.
>
>Amir
>
>
>>My only comment was addressed at "super-GM positional play" by DJ.  I saw very
>>good play, particularly when it was in great trouble.  But I didn't see
>>"super-GM positional play" in many cases..



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.