Author: Charles Worthington
Date: 09:29:46 04/09/03
Go up one level in this thread
On April 09, 2003 at 11:11:49, Aaron Gordon wrote: >On April 09, 2003 at 07:52:52, Pavel Blokhine wrote: > >>On April 09, 2003 at 07:36:31, Aaron Gordon wrote: >> >>>On April 09, 2003 at 04:20:05, Pavel Blokhine wrote: >>> >>>>On April 08, 2003 at 23:33:28, Aaron Gordon wrote: >>>> >>>>>On April 08, 2003 at 12:00:58, Anthony Cozzie wrote: >>>>> >>>>>>I believe you. I just thought it was a fun prank to pull. Athlon's don't have >>>>>>hyperthreading, nor can anyone get them to run at 2.9GHZ. I do think the athlon >>>>>>is a superior processor, though. >>>>>> >>>>>>Anthony >>>>> >>>>>I've got a friend that runs 3.1GHz with freon (Prometeia). There are also people >>>>>who get 3.0-3.4GHz with Dry Ice and Liquid Nitrogen. >>>> >>>> >>>> >>>>I hope you will all forgive my ignorance on the subject, but what the hell is K7 >>>>XP? What kind of computer is that? >>> >>>K7 is AMD's 7th Generation class of CPU. It's just an Athlon, so K7 XP means >>>Athlon XP. >> >> >>Ok thanks. But how is it supposed to be faster than a dual Xeon 2.8 Ghz? > >The Athlon is faster because it does more work per clock cycle. Imagine someone >50 feet tall making 1 foot step every second. Now imagine a midget taking 10 >steps per second.. who would cover more ground? The 50 foot person of course. >The other person is taking 10 steps per second vs 1 step per second. Same thing >applies with processors. 2.8GHz doesn't mean you're going to be faster than >2.1GHz. > >An AthlonXP 2200+ (1.8GHz) is about the speed of a P4-2.8GHz in chess, sometimes >faster. An AthlonXP 2600+/266 is 2.13GHz, so I'm not sure where Charles is >coming up with his numbers. He mainly spouts garbage from Playchess, which >honestly means absolutely nothing. The games are all different, the positions >are different so there is no "controlled" test to see which is faster. > >He was stating he gets 5 million some-odd nodes/second spikes, sure, anyone can >do that. Try running it for more than 1 second though. I could fire up Goliath >Light v1.5 and get something on the order of 7-8 million node/second spikes. Big >deal. Run a real test. I've done the controlled tests, AthlonXP's are faster by >far. Plain and simple. Actually the numbers I gave him were from controlled tests. Both were fritzmark tests running fritz 8 with 64 MB hash in identical positions. Charles
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.