Author: Robert Hyatt
Date: 15:08:49 01/30/99
Go up one level in this thread
On January 30, 1999 at 16:06:49, Matt Frank wrote: >>Simply show me the results, and I'll certainly change my mind. Two years >>ago I said this for game/30 too. I found out different because I watched 4 >>computers and 4 GMs play a round-robin (Crafty/Ferret were two of the >>computers) and I saw _every_ computer finish with a higher score than _any_ >>GM (ie the final standings had 4 computers followed by 4 GMs). But I haven't >>seen that at 40/2hr yet. I saw Anand just squeeze rebel around the throat until >>it died. Yes anand is right at the top of the GM list. But Ed chose to play >>him, so we have to use the few games we have. That pair of games showed that >>computers just don't play at that level by a wide margin. They may find a neat >>tactical shot, but they will get rousted positionally... > >Well that time has come! In the first 40/2hr game that Anand had with Rebel last >summewr the computer missed a forced win with 36...Qe6! When the game was >reviewed later Rebel showed the win one iteration later in its analysis. This >move demonstrated that Rebel 10 outplayed Anand in that game up to that point. >Now I know that you will say that the computer lost and that is true, but almost >no IM could have played on that level for that long with Anand. Remember he has >not lost or drawn to an IM in a tournament for over two years. Further in the >second 40/2hr game rebel had strong prospects with excellent opportunities but >was outplayed by Anand when he made a very strong sacrifice (this sort of >thing--outplayed by strong sacrifice--is very likely to occur to 90% + of all >GMs that play Anand). > >If you are so confident that micos have no chance against strong GMs what is >your opinion of my first post which states that my G3 233 MHZ computer running >with Hiarcs 7 software for MACs is at NO DISADVANTAGE with GMs the likes of >Benjamin, de Firmian, Yermolinsky and Christiansen at 40/2hrs. Why don't GMs >return my phone calls when I offer to set up such a match? > >Matt Frank _this_ was a problem when most thought computers were 2200 players. GM players are just not going to sit down and do what they do for a living, yet do it for 'nothing.' If you want to play one, you are going to have to pay one. IE I won't sit down with you and teach you how to write an operating system. I teach that every day. I won't sit down and do what I do every day, and do it again for nothing. And it isn't because I believe you might ask me a question I can't answer, it is because I don't want to spend that much time/effort for nothing. In the case of a GM, they burn a lot of time, and _if_ they lose a game, the game appears somewhere later to embarass them. There's little incentive for one to play... Kasparov wouldn't have thought about playing DB if it wasn't for that million bucks... for example... They certainly _know_ they are going to have to 'work' very hard to beat a computer. They know it isn't just 'play the 1500 player and roll him up without having to calculate at all'. I don't think they are 'scared' at all, but they are 'aware' and know it is going to be a struggle that won't be easy at all. And they just won't invest that kind of time/effort for zilch. IMHO of course, this from talking to _several_ GM's I know pretty well..
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.