Author: Reynolds Takata
Date: 20:31:23 02/01/99
Go up one level in this thread
On February 01, 1999 at 20:14:33, Jeff Anderson wrote: >On February 01, 1999 at 10:35:33, Reynolds Takata wrote: > >>If we did it your way, Hiarcs might lose > >Whether or not Hiarcs wins or loses the match, would it not be a valuable win >for computer chess? Indeed it is really of some serious importance if the match is relatively fair. If you give the GM's 10 trumps to start with that the comp doesn't have, you are denying the comp a fair chance. Why is everyone so insistent on proving that Hiarcs and >other PC programs are GM strength. We're not insistent upon proving anything, i'm sure that more than a few of us feel that it has already been thouroughly and undeniably proved(this coming from a person who has played a couple of GM's in his time). However a few of us would like to see the nay sayers silenced. Not unfairly silenced though. I would like for the GM and Hiarcs to sit down and play tomorrow without any preparation just on there current talent and skill(this however wont happen). Since it wont happen shifts in fairness are occuring and they are happening in the favor of the GM, because the GM works to give himself an advantage. The Comp is inanimate and has only one way of equaling out the disadvantages to show what happens on an as close to as possible even footing. This requires animate beings to help make the contest fair. Giving hiarcs a good opening book and letting it analyze games is not unfair, because they are functions that it's programming allow it to do, and further something that the GM chooses to do for himself. , when they may not be? Just because this >specific creature, built to beat one GM, beats that GM, does not mean it is GM >strength. Indeed it does, mean that it's GM strength, because to beat a GM you have to be playing GM level chess or exceptionally lucky. The only case for your statement is to say that the GM wasn't playing GM strength chess, so his losing doesn't show anything. Even Fritz5 did great preparation I believe in the GM tournament it >one. I think it would be much for a sign of its true strength against humans if >it one went and play in a large American open tournament, where the oppoenents >are not know until a day, or perhaps hours before the round. > I think this would be a great thing, in fact in the past it has been pointed out that this should be done. However, there is a problem of people not playing as they would against another human, when they know they are playing a comp, thus the results are not neccessarily exactly what you would want IMHO. Regardless though i'm quite sure a comp would kill in such an event. >Sir Conan Arthur Doyle: It is a capital mistake to theorize before one has >data. Insensibly one begins to twist facts to suit theories, instead of theories >to suit facts. Well your point might be well taken if for the flawed idea that we have no evidence to support the theory. When you have comps having performance ratings of over 2600(Rebel against 4IM's and 2GM at Aegon), IM Kaufman saying progs are GM strength, A program holding a 2785 rated player to a draw in a 40/2 game and almost holding him to a draw in anither 40/2. A comp beating a IM in a 6 game match(40/2 on MUCH slower hardware than is available today)Countless experts in the field of actual chess play, like my self being a USCF life master. You have considerable and strong evidence to support the theory. In a court of law when searching for an expert witness in the field of actual chess play. Actual chess players like IM Kaufman and Dean Hergott will be accepted. A computer programmer will get the thumbs down(more than likely), and i'm a lawyer. R. Rakata
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.