|
|
#1 (permalink) |
|
VIP Member
![]() Join Date: Aug 2004
Posts: 1,238
|
I was wondering what cases it would be advantageous to not have hyperthreading. For example, at work I use a P4 2.26 GHz w/o hyperthreading and at home I use a P4 3.4 GHz w/ hyperthreading. My work computer runs my simulations quite fast, with CPU usage going to 100%. I run the same simulations on my home computer and it actually seems slower, and the CPU usage stays right around 65%. I noticed the same thing on my other Celeron D processor as well. Is this a CPU issue or a software issue?
|
|
|
|
|
|
#2 (permalink) |
|
Administrator
![]() Join Date: Jul 2004
Location: Canada
Age: 24
Posts: 19,900
|
While there are situations where HT is not ideal the performance drop is usually not earth shattering. For general every day stuff, leave it enabled however for most games and stuff (with some exceptions for games that are coded for HT), you'll prolly want it off (essentially disable it for anything that depends on the BUS as the individual execution units will be negotiation for those resources). Even then (worst case) you wont notice something like a million fps drop or anything
__________________
ASUS P5K Premium WiFi-AP, Q6600@3.7 / ASUS P5ND, E6400@3.8 4GB OCz Platinum XTC 8500 / 4GB CorsairXMS2 6400 5x500GB Seagate 7200.10 / 2x500 Seagate 7200.10 OCz 8800GTX 768MB @ 630/800 / 2x Galaxy 8800GT SLI |
|
|
|
|
|
#3 (permalink) |
|
VIP Member
![]() Join Date: Aug 2004
Posts: 1,238
|
Yeah, I don't think I'd ever dissable it; I do way too much multitasking to do that. I guess my question is why doesn't my simulation (basically numerical modeling) use the CPU to its max? I know my encoding programs due.
|
|
|
|
|
|
#4 (permalink) |
|
Moderator
![]() Join Date: Dec 2004
Location: Canada
Age: 24
Posts: 9,786
|
hyperthreaded CPUs always show 50% load when they are under full load for a single threaded application, only when both logical cores are in use will you ever see usage over 50%
__________________
Then, when you have found the shrubbery, you must cut down the mightiest tree in the forest with a herring!! Tiax rules all! Know this and be judged. I must plug a couple comics because they are good :D: www.ctrlaltdel-online.com www.userfriendly.org |
|
|
|
|
|
#5 (permalink) | |
|
VIP Member
![]() Join Date: Aug 2004
Posts: 1,238
|
Quote:
|
|
|
|
|
|
|
#6 (permalink) | |
|
Administrator
![]() Join Date: Jul 2004
Location: Canada
Age: 24
Posts: 19,900
|
Quote:
![]()
__________________
ASUS P5K Premium WiFi-AP, Q6600@3.7 / ASUS P5ND, E6400@3.8 4GB OCz Platinum XTC 8500 / 4GB CorsairXMS2 6400 5x500GB Seagate 7200.10 / 2x500 Seagate 7200.10 OCz 8800GTX 768MB @ 630/800 / 2x Galaxy 8800GT SLI |
|
|
|
|
|
|
#7 (permalink) |
|
VIP Member
![]() Join Date: Aug 2004
Posts: 1,238
|
So if a CPU is running with hyperthreading enabled its speed won't be equal to the speed without hyperthreading enabled due to the speed using multiple processors not being additive? Or am I just using screwed up logic?
Sorry for all the questions and no actual problem. I figured I'd just directly leech some knowledge from you guys ... and this hyperthreading thing has been bugging me for some time. |
|
|
|
|
|
#8 (permalink) | |
|
Administrator
![]() Join Date: Jul 2004
Location: Canada
Age: 24
Posts: 19,900
|
Quote:
__________________
ASUS P5K Premium WiFi-AP, Q6600@3.7 / ASUS P5ND, E6400@3.8 4GB OCz Platinum XTC 8500 / 4GB CorsairXMS2 6400 5x500GB Seagate 7200.10 / 2x500 Seagate 7200.10 OCz 8800GTX 768MB @ 630/800 / 2x Galaxy 8800GT SLI |
|
|
|
|
|
|
#10 (permalink) | |
|
Administrator
![]() Join Date: Jul 2004
Location: Canada
Age: 24
Posts: 19,900
|
Quote:
![]()
__________________
ASUS P5K Premium WiFi-AP, Q6600@3.7 / ASUS P5ND, E6400@3.8 4GB OCz Platinum XTC 8500 / 4GB CorsairXMS2 6400 5x500GB Seagate 7200.10 / 2x500 Seagate 7200.10 OCz 8800GTX 768MB @ 630/800 / 2x Galaxy 8800GT SLI |
|
|
|
|
![]() |
| Thread Tools | |
| Display Modes | |
|
|