Author: Milton Zucker
Date: 14:54:14 02/20/99
Go up one level in this thread
On February 20, 1999 at 15:22:42, blass uri wrote: >I am going to play the 3/4 final with a computer and I want to know which >program is best to use. >Buying Hiarcs7 is one of the options. > >I found in Hiarcs7's homepage that they are going to do a better version for >correspondence games so I want to know if it is a good idea to use this version >for correspondence game >I understood that it is weaker at tournament games but the question is if it >improves with time relative to the originial Hiarcs(If it is weaker by 50 ELO in >tournament time control and weaker by 200 ELO at blitz then I can hope that it >will be better at correspondence time control). > > > >Here are the news from Hiarcs7's homepage: >>Deep search: A special deep search selectivity option will be available >specifically for >correspondence chess and overnight analysis. Search Selectivity 6 and 7 are >introduced with >this version. > >WARNING: Testing indicates the deep search selectivities of 6 and 7 are 50+ Elo >weaker at >tournament chess and faster time controls. However, overnight analysis appears >to be >improved, hence this feature is aimed at correspondence chess and overnight >analysis of >difficult positions requiring a deeper more selective search. > > > >Uri I just visited the web page and it doesn't look like the new version has been released yet, so probably the only person who could answer your question would be Mark Uniacke (unless there are some beta testers who have access to the program). But from my reading of the description, the newer version should not play weaker tournament chess as long as the user keeps the search selectivity set to "5". I think what Mark is doing is adding 2 extra settings of 6 and 7 which he says should only be used for correspondence chess or overnight analysis of chess positions. I find the description of these new settings interesting because I would have thought that a LESS selective search (or at least brute force searching to a greater number of ply before selective search kicks in) would give a more accurate analysis of chess positions at overnight analysis because of the extra time available and less of a chance that a good move would be overlooked. I would be interested if chess programmers would comment on this. ...Milton...
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.