Author: Rolf Tueschen
Date: 03:18:01 10/10/02
On October 09, 2002 at 22:50:41, Robert Hyatt wrote: Preface: How I, Bob, would compete against Kramnik > >I have thought more about this, and I would make the following controversial >statement (the reasons will eliminate the controversy however): > > Crafty _would_ do better than Fritz in this match. Not that I would expect it >to win, but it _would_ do better. > >Now for the "why?" > >1. If we stick crafty in _right now_ all the weaknesses he has found in Fritz >are useless. Crafty evaluates endgames differently (and better in many >but not all cases). So all his pre-match preparations would go right down >the toilet, and the onus would be on him to find Crafty's (many) weaknesses, >OTB in games that count. Not that easy. One mistake is all he has to make >to lose a game, and probing for unknown weaknesses would not be >easy with only eight games to play, and with the program possibly "changing >personalities" between rounds. > >2. If Crafty had started this match it would have done better. Because I would >_not_ have agreed to the silly rule about "he gets a copy of the program >beforehand and no changes can be made to the program after that unless he >gets the new version far enough in advance to test it thoroughly." > I have no doubt Kramnik would still win. Perhaps he would even win >every game. But the games would _not_ be boring endgame losses, >the endgames would have inbalances that would make them quite interesting, >as would the middlegame. Theoretical Background: The DEEP BLUE 2 "Strategy" >I think _the_ problem with this match is the >pre-match preparation rule, which is nothing short of completely ridiculous. Short comment: When people like Rolf criticised DB2 team and IBM for their Incredible and insulting behaviour, Bob always answered that that would Have been part of the negotiations about the conditions for the match. IfKasparov had wanted he could have received much more favorable conditions! So it was his own fault! Criminological revelation of a major fault in CC NOW – when Kramnik changed such tricky behavior of the computer team by simply asking for the final version some weeks before the event And for the books with only minor changes during the match, Bob is crying "This is unfair! This is ridiculous! This is making no sense!" In other words, Bob is the second programmer after Ed Schröder, who has confessed that the CC side is in big trouble without such psycho tricks. If CC has to compete under normal conditions, as we know them for Centuries in human chess, it simply doesn't work as it should! CC needs unfair advantages, otherwise, based on chess alone, it doesn't function. Confession: >That is _the_ point, IMHO. The basic flaw with _all_ computer chess >programs is that they are static entities. They don't change on their own. >If you play me and you beat me using the same idea over and over, >I will adapt and it won't continue to happen very long. Not so with a >computer, and that is a _huge_ problem that most are overlooking... First comment from the criminological department: So, for 5 years now we hear that the learning feature of computer chess programs prevent the progs from playing the same old nonsense over and over again and Bob is now talking about static entities. Where is parallelism gone? Down to the toilet? Confession (continued) >Maybe. The 8-way systems are _not_ that impressive, IMHO. >They have a _huge_ memory bottleneck. >The quads solve this with 4-way interleaving. The 8-way boxes >_still_ use 4-way interleaving, which means you simply have >twice as many processors fighting over >the same memory bandwidth.... result == starvation. > Second comment from the criminological department: So either we do it the way Hsu and team did it or we can forget of making progress. Either we succeed in a) getting play conditions that hide our tweaking or b) we must have such deep parallelism that nobody could ever prove that we tweaked too much. I have a simple question to both Ed & Bob and to CC as such. Why can't you understand that this is all about CHESS? And not About tricks to bust the humanplayers psyche! All ideas or thoughts or proposals are most welcome. Rolf Tueschen
This page took 0.01 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.