Author: Thorsten Czub
Date: 18:02:53 01/20/04
Go up one level in this thread
On January 20, 2004 at 20:49:40, Paul Doire wrote: >You have been here a long time and I understand your point. >I also respect your tournaments, the "range" in your opinion is not linear. no - the increase of strength is linear with time within a certain range. out of this range the increase of strength is more a function of the search-strategy of different engines. so when testing out of this range you get the speed how the search tree is build, not the strength of the engines. >That is your opinion. I also agree that the engines will perform differently on >different hardware. Additionally, length of matches makes a difference, I have >proven this myself. >The leaders are so close it is hard to say whether it is linear or non-linear. >I can run the same tourney 5 times and get different results. >As I am sure you can too. that is why a bigger time control gives more security. you can rely better on the numbers. for the same reason you make many events. because you want to rely on the numbers. the best would be 40/120 and many events. but this takes a while. you test the opposite. FAST time controls on slow machines, ok - you make many events. but i doubt that this gives a fair chance. on a small distance, a car, a horse, a human beeing and a human beeing on a bike are doing a race. who is winning after 5 feet ? who will win after 20 feet ? and who is the winner after 1000 feet ? why don't you play 40/120. that would give 40/40 on 1200 machines. this seems to be a serious Time control in my eyes. >My omly point originally was that I think Hiarcs is not the strongest engine. >Refer to SSDF for additonal confirmation.
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.