Author: Sune Fischer
Date: 00:23:29 07/14/04
Go up one level in this thread
On July 14, 2004 at 03:10:36, Russell Reagan wrote: >On July 14, 2004 at 02:04:06, Sune Fischer wrote: > >>Even if Omid was the greates chess programmer in the world and had written the >>ultimate SMP engine, he would still need to get his hands on some super hardware >>to be competitive. >> >>It's just as much, if not more, a contest of who can get the fastest hardware >>than who can write the best program. >>So if you want to be a WCCC don't spend too much time on programming, better >>spend time on finding a big sponsor for your hardware! >> >>That's what makes it all a little silly, IMO. >> >>I don't really have an alternative solution, I believe we already have the SSDF >>as a sort of world ranking on single cpus. An 11 round tournament could not >>replace that anyway. >> >>-S. > >The WCCC is a contest with certain rules, and one of those rules is open >hardware. Some people realize what a significant part of the event this is and >they spend a non-trivial amount of time making arrangements to borrow top of the >line hardware. Others down play the hardware aspect of the competition and >complain because they get beat by someone who understood the contest better than >they did. Obviously. As I said it's just as much a contest to see who can get their hands on the fastest hardware. It's far better to spend two month searching for a big machine (how many pages was it Vincent had to write to get CPU time on the super computer??), than it is to try and improve the program (requires a years work to equal out twice faster hardware). Obviously not all chess programmers would find that very interesting. It's a bit like saying that in order to win the 100 meters sprint, you must first demonstrate that you can stand on your head and play the piano backwards. >The only thing that is silly is that we programmers think it is a programming >contest. I'm certainly guilty of that. The problem is to find a good balance as there are a number of criterias we would like to fullfil. 1) chess programming should the main focus, not who has the biggest sponsor 2) the chess being played should be of the highest level possible 3) must be room for radical approaches, such as the Brutus machine I do not see a way to satisfy all those requirements, a compromise is probably the best we can do. -S.
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.