Author: Uri Blass
Date: 11:04:06 03/31/02
Go up one level in this thread
On March 31, 2002 at 13:23:16, Roy Eassa wrote: >On March 31, 2002 at 11:52:26, Vincent Lejeune wrote: > >>Great new for computer-chess clubers >> >>Like Boris Gulko some days ago, Ilya Smirin fight against Junior, Fritz, >>Shredder and Hiarcs. >>A little disapointement : the hardware is not upgraded :( >> >>http://www.kasparov.com/serve/templates/folders/show.asp?p_docID=20741&p_docLang=EN > > > >Great news! > >Also good is this: > >"Human opponent Prize Money: $1500 per win, $500 per draw, no money per loss >($2500 Guaranteed minimum prize)" > >I don't recall whether a similar compensation schedule existed for the Gulko >test. > >As I stated before, I think the reward for a win should be several times higher >than the reward for a draw, because the difficulty is also several times higher. > (The fact that a draw is traditionally scored as 1/2 of a win does not mean >that it is exactly 1/2 as difficult, especially for a human against a strong >computer.) > >In this case, 3x is used. That's OK, but I think at least 5x might be better. Your idea means that +2 =3 -3 is better than +1 =7 for the GM. I am afraid that it may encourage smirin to take unnecessary risks and not to do the best that he can. I think that 3x is o.k if the target is to encourage the GM to do his best after considering the assumption that humans hate risks. Uri
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.