Author: Mogens Larsen
Date: 06:57:56 05/14/00
Go up one level in this thread
On May 13, 2000 at 10:56:21, pavel wrote: >Hi Mogens, > Since positional.lrn comes default with crafty, and since fritz has >nothing as positional.lrn at all!!!, i dont think its unfair not even if its a >test ;). I believe any learning file that comes default with the program >shouldnt be turned off, because the program has been build to play with such >options. Would you play an opponent who doesnt learn from his/her mistakes?? >thanks >Pavel ;) Hey Pavel, In general I would agree with you, since I believe in testing a program as a entire entity. But I don't see the problem with testing on "equal" terms with any parameters allowed or disallowed as long as the conclusions drawn are careful and demonstrates the usage of thought. This is, and has been, my main objection to the tests posted in this forum. As you mention yourself, there's a lot of reasons why some testing is sensible and some testing isn't. Sincerely, Mogens
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.