Author: enrico carrisco
Date: 01:37:06 10/11/05
Go up one level in this thread
On October 11, 2005 at 04:13:25, Jonas Cohonas wrote: >>What's intersting is that the kn/s on my Athlon64 3400 is quite different >>between Fruit2.1 and Fruit2.2. >>Fruit 2.1 around 900 kn/s in middlegame >>Fruit 2.2 around 550 kn/s in middlegame. >> >>Maybe more knowledge or badly optimized compile? > >Yes it is indeed quite slower that 2.1, now is that due to the 500kb security >code perhaps? The copy protection code will have ZERO effect on the engine. That part of the code isn't even used or accessed after it's initial check on boot. As far as the the search speed changing, that is most probably due to improvements in overall engine strength (namely knowledge.) Compare Fritz 5.32's speed with Fritz 9's speed -- but which do you think is more clever? On the road from strict pre-processor to intelligent search we tend to lose some search speed, but it is well worth the trade. I didn't notice such a drastic swing in the overall kn/s as you have stated above, however. On my 2.5GHz Athlon XP (roughly equivalent to a 2.35 or 2.4GHz A64 for Fruit) it was a difference of less than 250 kn/s between revisions. Regards, -elc.
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.