Author: Terry McCracken
Date: 21:23:51 04/23/02
Go up one level in this thread
On April 23, 2002 at 23:59:27, Christophe Theron wrote: >On April 23, 2002 at 12:56:11, Roy Eassa wrote: > >>On April 23, 2002 at 12:50:24, Gian-Carlo Pascutto wrote: >> >>>On April 23, 2002 at 12:43:26, Roy Eassa wrote: >>> >>>>>What about the _beta_ testers? >>>>> >>>>>They report such results here all the time. >>>> >>>> >>>>I don't recall having seen ANY beta test reports regarding the strength or >>>>results of Chess Tiger 15, but I'll admit my memory is not 100%. :-) >>> >>>Maybe because they're not allowed :) >>> >>>Not reporting results != not doing testing >>> >> >> >>You're right. That's what I figured. But I find it a bit curious that >>Christophe seemed to go out of his way to imply that results against top >>competitors (e.g., Fritz 7) are NOT a factor in his decision of when to release >>each successive version of Tiger. Rather, he implied, he has his own set of >>benchmarks that tell him when it's ready. >> >>If the above is true, it means that coming out after Fritz 7 does not imply >>being stronger than Fritz 7 (which is the statement I was addressing >>originally). >> >>Software beta testers usually focus on finding bugs. It's inevitable in >>computer chess that they would also test strength/results, but I suspect that's >>not what Christophe wants them to focus on. > > > >I don't test against my competitors and what the beta testers report is >generally not about what the competitors do. > >I don't remember any occurence of a change I would have made because a beta >tester would have reported a problem against a given competitor. Maybe it >happened, but it is not frequent and not really important when compared to all >the other things I do. > >Anyway when the beta test stage starts, the program is almost freezed and I >don't make changes, just bug fixes. > >For me it is a race against the competition, but a blind one. I do not know >where my competitors are until the program is released. A few months after the >release I get some measure of where I stand, then I go back in "blind" mode >until the next release. > >Believe it or not. > > > > Christophe That's interesting; and maybe a little surprising? Have you any preference to any of the other top programmes, and ever considered testing CT against it? I remember back in the "Stone Age" when we all had chesscomputers as software was still weaker on PC's, that Richard Lang of Mephisto fame used Dan Spracklen's Mach III programme as an acid test for his Almeria programme. Terry
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.