Author: Uri Blass
Date: 15:13:45 09/27/05
Go up one level in this thread
On September 27, 2005 at 17:52:29, Dieter Buerssner wrote: >On September 27, 2005 at 17:29:54, Steve Maughan wrote: > >>I also was not aware that Fabian thinks it's a bug in the search. > >I think, the word "bug" is not really adequate here (and in many other posts of >this and similar threads). Perhaps Fabien mentioned this as bug, but I believe >he really meant to say: Not implemented as well, as I would like. > >It may not be easy, to define a bug. Say, some application needs to sort things >a lot. The application uses selection sort (a method, that is most probably well >working in chess engines, where typically only a very few dozen of moves have to >be sorted). When that application has to sort really lots of data, selection >sort might become the bottle neck. Still the application would work. But using >some different sorting algorithm (i.e. quick sort) would make the application >much faster. Would you consider the application, that uses the inferior sorting >algorithm buggy? I wouldn't - just not well tuned, not really scalable, ... I >guess, Fabien used the word bug (I never saw it from Fabien, but Uri cited this >many times) in the context above, whre thingw do not work as well, as he would >like - but things still work (i.e. the engine does not crash). > >Regards, >Dieter It seems that you did not read the last part of the readme of Fruit2.1 otherwise you could see that fabien claims that fruit2.1 has search bugs and even asked other programmers to help him to fix his bugs before 1.9.2005 Fabien promised in that part to select the most artistic bug fix after the date limit(the only condition is that it needs not to be null move pruning) and help the relevant programmer to fix one of his bugs based on his choice(excluding "it plays bad moves"). 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.