Author: Mark Young
Date: 16:11:46 08/16/98
Go up one level in this thread
On August 15, 1998 at 21:52:31, Shaun Graham wrote:
> Below is a report from I.M. Larry Kaufman on Hiarcs 6 off the ICD page. Now
>he states that on a fast pentium Hiarcs 6 is GM strength. Why would he as an
>Expert on chess,an international master who owns and plays the program say this,
>unless he believed it to be true from his own experience? Is the author of
>Hiarcs paying him to say such things(i doubt it)? Now of course we could get
>some other I.M. to say the opposite, regardless though, Kaufmans testimony
>increases the evidence that comps are GM strength. For the simple reason that
>most I.Ms would simply like to claim that they(IMs) are stronger, that they have
>the edge(now of course i haven't taken a poll, but considerable experience leads
>me to believe that indeed this statement is true). Also, If you read the latest
>ccc opinion poll, it appears that the enlightened majority of the readers of
>this group hold the same comps are GM strength.
>
>"How strong is Hiarcs 6? Well, Hiarcs 5 was second on the last Swedish ("Ply")
>rating list, lagging behind Rebel 8 by about 25
>points, roughly the amount believed to have been lost by the bug. So even if the
>improvements in Hiarcs 6 don't amount to anything,
>it is likely to end up rated more or less on a par with Rebel 8, the current
>leader. THIS SHOULD PUT IT IN THE LOWER PART OF THE
>GRANDMASTER RANGE when running on a fast Pentium. This was more or less
>confirmed by a widely publicized match between Hiarcs
>6 and Canadian International Master Deen Hergott (FIDE rating 2390) played under
>standard tournament conditions, which was won
>by Hiarcs 6 by a 4-2 score, which equates to a performance rating of FIDE 2510.
>To obtain the Grandmaster title, a player must be
>FIDE rated at least 2500 (as well as obtaining three "norms"), so this result is
>not out of line with the Grandmaster claim made by
>Canadian computer chess writer Al Tomalty.
> Hiarcs also scored 4-2 in the just completed Aegon tourney, half a point less
>than the
>top computer scores (Kallisto, Rebel, and Chessmaster 5000 scored 4 1/2) but
>still a good showing. The games are played at a time
>limit just slightly faster than the usual 40/2, and the human opponents of the
>machines are mostly masters, IMs, and GMs."
I don't think IM Kaufman would say that micro computer programs are GM's unless
he thought it.
I think more people will come to the same conclusion over the next few months as
even faster hardware and stronger chess computer software hit the market. I
think we are already here. If the standard is 2500 elo rating or better.
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.