Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: What Makes a Chess Engine Better Vs Humans?

Author: Serge Desmarais

Date: 20:10:00 09/02/98

Go up one level in this thread


On September 02, 1998 at 12:10:15, Robert Henry Durrett wrote:

>On September 02, 1998 at 00:54:10, Serge Desmarais wrote:
>
><snip>
>
>>Note: The bug reported by Bob Hyatt that did not change anything against humans
>>but a lot against computers were in games played in ICC : mostly 5 minutes
>>blitzes (sometimes with an increment of 2-3 seconds) and maybe a few 15 minutes
>>games? Most GMs in ICC avoid other kind of games than blitz games. Even against
>>a computer or program that has a grave bug, I don't feel that with 2-3 minutes
>>on the clock and just taking less than 10 seconds to move, a GM would have had
>>time to "refute a weak play". GMs were being beaten by computers in blitz games
>>on much slower computers than the actual ones (even on 486 computers!)... And
>>again, in the Aegon tournaments, it was still a bit too fast for a GM to show
>>his full strenght (for most of them).
>>
>>Serge Desmarais
>
>(1) I agree 100% with the above paragraph except as noted below.
>(2) One should not underestimate the speed with which the top GMs find good
>moves in blitz games.  Maybe the top chess engines are even faster, at least at
>finding the tactical content of the position, but the top GMs also see the
>strategic content of the position with lightning speed too!  The only problem
>they may have is that they may not have enough time to research and analyze the
>tactical content to come up with a sound concrete plan to implement their
>strategic ideas.
>(3) For several months, now, I have been watching blitz games between the top
>GMs and top performers on ICC.  [Key in "games *R19-C-B-d-w" to see a list of
>the top performers who are playing a game, and key in "ga *T-w-d-C" to see a
>list of the top titled players who are playing a game.  Then key in "follow X"
>where X is the ICC handle of the player you wish to follow.]  In fact, I find
>these games to be so interesting that I have quit playing regular blitz games
>there on ICC entirely.  I get so much more out of playing what I call "solitaire
>blitz."  This is a form of chess where you try to guess the GM's move before he
>makes it.  I find this to be loads of fun!  But one thing becomes crystal clear
>right away:  These guys are fast!!! Really, really, really fast!!!
>(4) I am GREATLY impressed by the speed with which the top GMs find the most
>important ideas in a position. This speaks all the more for the chess engines
>which win games against those guys.



   Yes, they are thinking fast, but still their games are full of blunders which
computers find almost instantly. Did you ever looked at the games from the
Herceg Novi Blitz Tournament, played in 1970? Just to give you an idea of the
strenght of the event, I will give you the names of the players who were there
(this tournament was a kind of "World Blitz Championship" and each player had to
play 2 blitzes against each other player - one as White and one as Black) :

Fischer, Tal, Korchnoi, Petrosian, Bronstein, Hort, Matulovitch, Smyslov,
Reshevsky, Uhlmann, Ivkov, Ostojitch (this is about the order of their score)

   Fischer only lost ONE blitz game (to Korchnoi) out of 22!!! He also allowed 4
draws. So he won 17 blitz games out of the 22 played! He was a kind of monster
at blitz and wiped Tal, Petrosian and Smyslov completely, winning all his games!
(Remember that Tal --who came second-- was Blitz Champion of the World not so
long before his death, wiping ahead of Kasparov (the "regular" Champion of the
World but who could not match him at blitz!)...

   So, maybe Fischer is the greatest blitz player who ever lived, but if you
look at the games played with a strong program, you will see a lot of blunders
even in Fischer's play, as well as the other players. My main point is that GMs
playing in blitz usually play the first move they see that look threatening and
aggressive enough. (Lots of missed mates in 2 -- and even mats in ONE!).

   Also, I remember watching a video of the Intel lightning chess series of
tournaments (25 MINUTES games -- that is no blitz) and there was also a lot of
missed mates and winning moves you and me would never miss in a serious game!
There was also an interview given by Anand (or was it kasparov?), in Europe
Echecs I think, saying that 25 minutes games were exciting and fun to watch and
would help make chess popular amonst non-players or weak amateurs, but that
these had NO theoritical value AT ALL. Several other GMs, when asked about it by
the reporter, ALL said the same... They all agreed that that was mainly
"woodpushing" for the show and the fun, and NOTHING serious could come out of
it. Even with that, I remember watching Dlugy playing a blitz against Crafty in
the ICC awhile ago, an UNRATED blitz... A few GMs refuse SYSTEMATICALLY to play
blitz games against computers based on principles (Hubner, to name one), saying
it is pointless and meaningless...

   As for me, I usually play blitz games when I am not in a mood to play chess
(I mean regular chess)... One could play thousands of blitzes and never improve
his tournament results... I sometimes play blites against people rated over 2200
USCF and I win my share of games, despite that I was NEVER able to defeat them
in a real chess game!


Serge Desmarais



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.