Author: William H Rogers
Date: 11:16:29 06/24/02
Go up one level in this thread
On June 24, 2002 at 13:13:21, Gian-Carlo Pascutto wrote: >On June 24, 2002 at 12:08:43, Robert Henry Durrett wrote: > >>If it takes about 100 times as many clock cycles to read from or write to RAM as >>would be required to read or write to a cache, then one might expect a >>hundred-fold improvement in speed. > >A chess programs speed is not solely dependent on the speed of >the RAM, hence, this conclusion is flawed. > >(But running entirely from cache would be a major speedup indeed) > >-- >GCP I concur with the above statement. Computer programs are cpu instensive, so the design of the algorithyms as far as execution speed is also a factor. While it is true that a program is only as fast as its weakest link, most programs run really fast in normal ram. The bus speed is slowly increasing in the newer machines and I am sure that will help a lot. Take a look at the palm tops, they only have ram for storage and operation. The fact that the batteries do not last when playing chess attest to the fact the they use the cps's, which causes a big drain on the batteries. But, to give credit where it may be due, if you can get it to work that way and it really seems to show a speed up, then good luck with your idea, and keep us all posted. Bill
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.