Author: Tom Kerrigan
Date: 22:38:09 07/07/03
Go up one level in this thread
On July 07, 2003 at 22:42:51, Peter Stayne wrote: >Upon further reading, my assumption is incorrect, but the findings of >single-threaded apps getting a boost still seems to be true. Hence the three >links below and choice quotes: > >http://www6.tomshardware.com/cpu/20021114/p4_306ht-12.html > >http://www.anandtech.com/cpu/showdoc.html?i=1746&p=6 > >On which I quote: > >Fundamentally we still only have one CPU and one set of execution units, so if >the OS dispatches two threads that contend for identical resources in the CPU >then HT could reduce performance. > >In the earlier versions of Hyper-Threading, there were some pretty significant >performance drops in desktop applications with it enabled. Luckily through >revision after revision of the technology and through the addition of a few new >components (flip back a few pages to see what's new) the vast majority of >applications will see a performance increase or no performance loss at all Right, the multithreaded apps will see an increase and the single threaded apps will see no performance loss. The benchmarks agree. >http://www.aceshardware.com/read.jsp?id=50000332 > >Quote: > >It is quite remarkable how almost every single threaded benchmark still got a >small performance boost from HyperThreading, between 1 and 5%. This shows that >HyperThreading has matured as it almost never decreased performance, as it did >in the first hyperthreaded Xeons. I don't see why Johan is saying this--all they tested was multitasking situations, multithreaded apps, and some games, and the only game that showed any statistically significant increase in performance, it wasn't clear if it was multithreaded or not. (My guess is, it was.) -Tom
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.