Author: blass uri
Date: 05:28:57 02/29/00
Go up one level in this thread
On February 29, 2000 at 08:01:29, Thorsten Czub wrote: >On February 29, 2000 at 02:50:17, blass uri wrote: >>He believes that longer time control is always better for tal. >>You can play at 40'/40 + 40'/40 +40' but if you get bad results for tal then it >>is not an evidence not to trust thorsten's results. <snipped> >the same counts for many slow-searchers, like hiarcs/the king ... >when you play 60/60 instead of 40/120, they get 1/3 of the time, the >fast-searchers outsearch them. with the minutes, the fast-searchers run against >the search three explosion, so after 2 minutes they don't really come deeper, >but cstal comes deeper. >so playing 60/60 or even faster is better for fast-searchers, >and playing longer time controls is better for slow searchers IMO. >like for human beeings. they can also play much better against machines >on slow time controls. I do not agree about general theory about it. I do not think that fast searchers always outsearch slow searchers. I believe that slow searchers can be good at tactics because they need less plies to see the same tactical idea because they know better which lines to search. Example:Junior does not outsearch Hiarcs in blitz. I also think that there are other factors that influence chess programs. For example hiarcs selective search can never be bigger than 31 when other programs like Junior can search more than 31 plies in the relevant lines. I think this can cause hiarcs a problem in long time control games. 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.