Author: Harald Faber
Date: 05:59:57 01/28/05
Go up one level in this thread
On January 28, 2005 at 04:49:54, Stefan Meyer-Kahlen wrote: >On January 28, 2005 at 02:36:23, Harald Faber wrote: >>On January 28, 2005 at 01:49:29, Sandro Necchi wrote: >>>How much stronger is it? >>>This time I will leave the fun to find out to you!:-)) >>> >>>Ciao >>>Sandro >> >> >>How many ELOs did you say is Shredder 8 stronger than 7.04? How many ELOs can we >>see for real (SSDF)? ;) > >These questions are probably the reasons why you don't get any Elo expectations >from Sandro for Shredder 9, at least this is why you don't get any guess from >me. I will not make any guesses either. Actually I was asked some question on that issue - read my answer: -------------------------------------------------------------------------- > I am considering buying chesstiger 2004 would u say it is > stronger then chesstiger 15 and if so by how much. thanks you I'd say it is stronger, but I will never say any ELO numbers on how strong one program is or how much stronger than predecessor or other programs. -------------------------------------------------------------------------- >You know that exact Elo numbers are very hard to measure, when you get a number >from Sandro, it is a value according to his tests and experience, not a number >which is based on marketing. Sandro is one of the most honest people I know, he >would never tell you something which he doesn't believe is true. Don't get me wrong. Except Genius 3 -> 4 -> 5 -> 6 -> 7 I assume that EVERY (commercial) program improved from version to version. I only see problems in measuring. I am sure that Shredder 8 is stronger than only one point compared with Shredder 7.04. But neither can or would I say any number - that would be guessing only - nor do I have any idea in which area this number lies nor how to "really" measure it. >If your Elo guess is wrong, even years later people will remind you of this >accusing you (directly or indirectly) of not telling the truth. Years later is trolling IMO. And guesses are easily seen as promises and taken for granted, so disappointment is justified for those people. Especially guesses from the development team are taken serious... >This has happened many times in the past and is hard and annoying to deal with. That is why I don't and wouldn't do any guesses. Another point is that testers sometimes get totally different results or test under their own special circumstances which might not be relevant for "normal" play. Letting program A+1 play 20 Kings Indian positions, not scoring better than program A, so program A+1 is "disappointment"... >Shredder9 will be out very soon and you will get many results of it from >independent testers quickly. You can chose which one to trust This is what I always do. ;) >and if you like what you hear. Basically this depends on the results. ;) >Of course according to our belive Shredder has been imporved, otherwise we >wouldn't have released it. > >Stefan I would never assume nor accuse you or any other commercial programmer for releasing a version just because "time has come" (1 year is over). I have accurate impressions of all commercial programmers and no reason to doubt.
This page took 0.01 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.