Author: Uri Blass
Date: 23:34:52 02/22/03
Go up one level in this thread
On February 22, 2003 at 20:30:53, Charles Worthington wrote: >On February 22, 2003 at 20:28:26, Charles Worthington wrote: > >>On February 22, 2003 at 17:45:50, Matt Taylor wrote: >> >>>On February 22, 2003 at 12:05:08, Sune Fischer wrote: >>> >>>>On February 22, 2003 at 11:38:04, Uri Blass wrote: >>>> >>>>>On February 22, 2003 at 11:05:42, Charles Worthington wrote: >>>>> >>>>>>On February 22, 2003 at 07:08:53, Robert Pawlak wrote: >>>>>> >>>>>>>Charles, >>>>>>> >>>>>>>I am envious, please let us know how you enjoy your new system... >>>>>>> >>>>>>>Bob >>>>>> >>>>>>It's nothing to envy, Bob. It was a necessity for my business. I have never >>>>>>spent quite_that_much on a machine for my office before. A single 3.06 would >>>>>>have done fine for me also but considering that I use it for chess I opted for >>>>>>the dual. Rich, I am not, and it certainly stings a little to pay that price for >>>>>>a motherboard that I will have to abandon when Intel changes the design on their >>>>>>cpus. :-) >>>>> >>>>>I think that for most people there is no special reason to have a dual for >>>>>chess. >>>>> >>>>>I do not see why is it so important to have the fastest hardware and for a lot >>>>>of programs Dual is not faster than single. >>>> >>>>Maybe not faster, but things run more smoothly on a dual. Having a spare >>>>processor to handle the background processes when one chip is at full load is >>>>something you can definitely feel as a user. >>>> >>>>And maybe duals would be more popular if more software supported it, goes hand >>>>in hand I think. >>>> >>>>If only C/C++ had some support for it natively, so you could split at a lower >>>>level rather than spawning large threads all the time. Fortran is great here, >>>>simple vector operations can be done in parallel. >>><snip> >>> >>>In the operating system it is done with threads or processes. Any abstraction >>>from that still returns back to it. Anything that can be done with abstraction >>>can always be done with lower-level constructs because abstraction is built upon >>>lower-level constructs. >>> >>>-Matt >> >> >>Actually matt i opted for duals because my intel 2.8 single was getting crushed >>by them. And they are required to hold 2900 + on the playchess server >>consistently. and for 3000 they are a neccessity. Don't pay much heed to the >>rating claims you hear here Matt. I have been on the playchess server for >>awhile. The singles may have a lucky run past 2900 but they don't stay there for >>long. The notion that double speed equals 20 elo is totally wrong from my >>observation. The highest dual is at 3111 and the highest single is lucky to be >>at 2900...and even at 2900 he wont be there long if he plays all challengers. > > >In fact if you place a dual 2400mp in a 100 game match with a single 2400xp with >equal ratings to start with the dual will be a whole lot higher than 20 above >the single when they are done. It may be interesting if people can post result of 100 game match to see the difference in different time control and if it is possible to prove the theory that the difference is smaller in slower time control. 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.