Author: Robert Hyatt
Date: 17:07:21 11/28/02
Go up one level in this thread
On November 27, 2002 at 10:28:01, Sune Fischer wrote: >On November 26, 2002 at 11:06:33, Robert Hyatt wrote: >> >>If you read the "underground analysis" it would seem that this is going to get >>worse. The new AMDs are supposed to have a built-in memory controller on the >>processor chip. Unfortunately, it is now known that it is an inferior >>controller compared to late Intel offerings. The question is, how is AMD going >>to respond? The answer is unknown, but if they don't, they will get their >>clock cleaned (again). > >Haven't heard that one, but the chipset design is greatly simplified making >multiprocessing very cheap and easy. > >I heard Intel is having trouble clocking up their latest beyond 3.2 GHz or >something (running too hot!), thus the delays of the Hammer may not be so >crucial to AMD. > >Let's wait and see if HT takes off, if so AMD said they might consider it, but >for now they believe the future is in real SMP. If HT doesn't take off, Intel is >surely going to get rammed (again) :) There is absolutely no chance that hyper-threading is going to "die". It has been discussed for years. There is just as much chance that the multiprogramming O/S systems will die and be replaced by the only single-program O/S I can recall, namely "DOS"... The idea is simply "right" and it will only get better. Next step is going to be two real cpus on a single chip, each cpu hyper-threaded... > >http://www.anandtech.com/showdoc.html?i=1752 > >-S.
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.