Author: Robert Hyatt
Date: 14:44:29 11/07/02
Go up one level in this thread
On November 07, 2002 at 15:53:10, Gian-Carlo Pascutto wrote: >On November 07, 2002 at 15:10:47, Robert Hyatt wrote: > >>I made a major gaffe this morning. When I started the quad 400 for you, I >>totally forgot that I was testing a new version sent to me by Mike Byrne. It >>had some unresolved problems with Swap() and Quiesce() that I was playing >>with. I neglected to restore the normal executable >>so I managed to play a few games with something that was really broken badly... >>Makes me wish I had used an experimental version number so that it would have >been obvious that I was screwing >>things up really badly... > >Do you understand my point now about ICC play and experimental versions? > >I accidentally played the first 13 games with the 'Tal' personality >because of bugs in WinBoard support (I mainly test using UCI). It is >meant as a fun setting for humans and no good for computer play. It >lost badly to crafty until I noticed things were not right. > >If you compare the score of the first 13 games with the score in the >latter games you will notice a big difference. > >When you play on ICC you play with the latest you have, even if >it often turns out to have serious bugs. When playing in a >tournament, you use a rock solid version that can be more than >a month old. > >In my opinion there is no comparison between the two. > >-- >GCP I disagree. On ICC we play casual games, and serious games. If someone wants to play me a serious match, I plan on using a "serious version". Otherwise I am "just testing stuff here and there." But nothing precludes serious games and serious testing.
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.