Author: Uri Blass
Date: 01:54:57 03/26/04
Go up one level in this thread
On March 24, 2004 at 17:31:35, Dann Corbit wrote: >On March 24, 2004 at 16:53:08, Uri Blass wrote: >[snip] >>The difference is more important and 10-0 is clearly more telling than 19-11 > >It is stronger, but less reliable. No 10-0 is clearly more reliable than 19-11 It usually will not happen but it does not mean that it is less reliable when it happens(you may suspect that something in the conditions is wrong when you see 10-0 but if you see that no program was significantly slower in nps during the match than you can safely stop the match after 10-0 and say that the new program is better). If it is true, then a dominant change may >have been found. But the odds that it is true are not nearly so strong as the >odds of 19-11 being true. It is dependent on the question what you test. You may test some changes that you are not sure for everyone of them if they did the program better. > >Something about streaks: >http://chessprogramming.org/cccsearch/ccc.php?art_id=118530 > >Do you remember the controversial Cadaques tournament? >A big winning streak followed by an even bigger losing streak. > >If you took the first ten games, you will think it is dynamite. >If you take the last ten games, you will think it is a dud. It was only 5-0 for Junior against Fritz and not 10-0 and I am not talking about match between different programs when opening preperations may be important but about test to test if something new is better than something old when book is not one of the varaibles to test. 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.