Author: Jeroen van Dorp
Date: 17:25:57 04/09/02
Go up one level in this thread
>I think the surprise comes from the fact that many people overestimate what >computing power can do. > >It's good and refreshing to point out that with the right algorithms a slow Palm >can kick a PC in the ass. > >Computing power is not all. Don't expect it to cover algorithmic deficiencies. > >People take it for granted that even an average amateur program will beat them >at chess. And that a Palm is so slow that they will beat it easily. > >Obviously this is not true. > >I think it is worth remembering. I agree with you. As long as chess isn't solved, both superior insight (or "knowledge") and calculating powers are the best match; if that fails I'd rather be gifted with knowledge than with calculating powers. But remember that the win of CT Palm may be no miracle, but just a logical result of the rating methodology. A 2200 program will have globally a win/lose ration of 50% against an equally strong partner. It will have a 20/80% win/lose ratio against a 2400 partner. The 20% win (including impressive wipes) is no miracle, but the reason why we call that program a "2200-rated" program. And if CT Palm had a 50/50% win/lose ratio against Fritz or CT PC, we wouldn't say "amazing" either. We would say: "the estimated strenght of the program is wrong. It should be more close to 2670 or 2700." And then again I would say: amazing what chess knowledge does on slow hardware. You already knew that. As an absolute zero at chess programming I have a very strong (and still growing) suspicion that it's true. But *if* it's a random game it doesn't prove it right now. Mind you, programming a 2200+ chess program on a 10x15cm device that beats 98% of all chess players in the world is impressive enough for me to say "amazing" without even looking at Jorge's game(s)... J.
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.