Author: Jeremiah Penery
Date: 22:04:12 12/04/00
Go up one level in this thread
On December 04, 2000 at 14:22:00, Vincent Diepeveen wrote: >He was searching 8 ply against me over 10. Now that's on a nowadays >machine. So getting 11 to 13 for deep blue in 1997 fullwidth was real >good IMHO. I don't search 13 ply fullwidth for sure except pawn endgame, *** >yet i completely would annihilate deep blue for sure with DIEP, *** >especially seeing what it did do wrong, and not even caring for the fact >that my openingspreparement would be real bad. Sorry, Vincent, this statement has no basis in reality. It reminds me a lot of your "famous" statement that: "In analysing positions Diep is by far the strongest chessprogram in the world." (from http://www.students.cs.uu.nl/~vdiepeve/homepage/about.html) There are several reasons why these statements are pure nonsense. I'm only going to give a couple, because I don't want to spend forever writing this post. 1) Completely unproven, and unprovable. This is obviously the most important point. You just make these statements of huge hyperbole, fully believing them, when you have zero evidence that they are true. Is it even possible to prove that Diep is the "strongest chessprogram in the world"? Probably not. No matter how you choose to "prove" that statement, there will always be a million other ways to do it that will "prove" it wrong. As for the Deep Blue statement, it's pretty unprovable also. Deep Blue will never play Diep. If it did, and Diep won, it would surely not be a statistically valid sample from which to derive the conclusion. Just as the 6 games you see aren't a valid sample to say that DB is stronger than Kasparov, or that DB is weak, or whatever. 2) If Diep is so strong, why hasn't it performed better in the past? I look through tournament results, and I don't see that it's ever won a tournament. Maybe it has, but if it's "by far the strongest..." it should be winning them all, or at least coming very close. Looking through past results, it seems Deep Blue (at least its predecessors) did win a bunch of tournaments, and won a bunch of games against strong human opposition. In this way, it earned the right to play against Kasparov. If Diep were so strong, it should be able to do the same. I've watched various Diep accounts on ICC/FICS/etc., and there are all sorts of positions where they simply have no idea what's going on. I'm sure there are probably some positions where Deep Blue doesn't really know what's going on either, but I can't believe they're so frequent. If they were, Kasparov would have annihilated the machine, just like he did in the 1996 match. In game 6 of that match, the machine just had no idea what was going on - and I've seen just about every current program get into similar-looking situations. It didn't seem as if the 1997 version of Deep Blue had quite those problems. Maybe it played some non-optimal moves, but chess isn't perfect. If you seriously analyze _any_ chess game for a long period of time, using probably thousands of man and machine hours to help you, you'll find all kinds of mistakes that every single player makes. The real test is to see whether those mistakes can be capitalized upon OTB - Deep Blue seemed to have passed that test in 1997. Whether Kasparov did "FM a-look-like play" or really played as well as he could, he still lost. In this case, that seems to be all that we can truly evaluate, since we lack so much other important information.
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.