Author: Dann Corbit
Date: 20:55:05 11/15/02
Go up one level in this thread
On November 15, 2002 at 23:28:38, Russell Reagan wrote: [snip] >I've always wondered how top chess programs would do against top humans at >correspondance time control. See the "Mike Valvo verses Deep Blue prototype" for an example of a human correspondence master blowing the doors off of a computer program. >One thing I have wondered in the past is whether >you would hold the match against top human correspondence chess players, or if >you would hold it with top GM's, only with longer time controls. I guess the >issue is, would Kasparov or Kramnik be better than the top two correspondence >chess players in the world at correspondance chess? Only way to find out is to try it. >As far as a CCC (correspondence computer chess) tournament, how would one run >that? You said the programs could only search at night, but how would you >enforce such a rule? Gentleman's agreement. It is a good idea to have someine besides the author run the program so that it might remain scientifically objective. >Maybe the player to move gets the opponent's move from a >3rd party (website?) and then has 12 hours (or however long) to submit it's >move, and then participants could start their turn at their lesiure (within >reason). But then the other player could have a variable time window to do extra >pondering. But that would also encourage each player to take their 12 hour turn >(or however long) promptly, to avoid giving the opponent extra pondering time. >Anyway, I like the idea, as long as these kinds of details could be worked out. Just use standard correspondence chess rules (breaks when needed, etc.)
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.