Stress Tests?

[johaness9]

Does running stress tests on your pc components in any way harm them?

 

[sirheck]

no not really.

unless there is a problem i.e.

bad part
high temps
to high an overclock
not enough juice from the psu.

 

[BaronMatrix]

Does running stress tests on your pc components in any way harm them?

Usually only if you're running out of spec (OC) or your components are not good quality. The average components will handle stress tests well as MS does lots of stress testing on all components for Windows.

It also depends on what the tests are stressing. Most components have a high MTBF (time before failure) so stress away.

Toms did several stress tests between AMD and Intel. They called them torture tests though.

 

[johaness9]

ah ok, do you happen to know what I can use to see if my Power Supply voltages are stable?

 

[BaronMatrix]

ah ok, do you happen to know what I can use to see if my Power Supply voltages are stable?

I know there is some SW somewhere. Most good PSUs don't fluctuate much. The newer "Active PFC" PSUs are even better at efficiency. If you want the best get one of these.

 

[sprite]

Does running stress tests on your pc components in any way harm them?

Usually only if you're running out of spec (OC) or your components are not good quality. The average components will handle stress tests well as MS does lots of stress testing on all components for Windows.

It also depends on what the tests are stressing. Most components have a high MTBF (time before failure) so stress away.

Toms did several stress tests between AMD and Intel. They called them torture tests though.

You forgot the MEAN in MTBF, mean time between failure.... it is a statistical measurement. If you are going to attempt to throw around jargon, please get it right.

Also, MS does not stress test software --- HW is stress tested, software is beta tested and debugged.

This is what I mean Baron, you should not give out technical advice because you have no knowledge of the technical.

Ouch.

 

[BaronMatrix]

Does running stress tests on your pc components in any way harm them?

Usually only if you're running out of spec (OC) or your components are not good quality. The average components will handle stress tests well as MS does lots of stress testing on all components for Windows.

It also depends on what the tests are stressing. Most components have a high MTBF (time before failure) so stress away.

Toms did several stress tests between AMD and Intel. They called them torture tests though.

You forgot the MEAN in MTBF, mean time between failure.... it is a statistical measurement. If you are going to attempt to throw around jargon, please get it right.

Also, MS does not stress test software --- HW is stress tested, software is beta tested and debugged.

This is what I mean Baron, you should not give out technical advice because you have no knowledge of the technical.

Stress tests that use the HDD are HW based nitwit. Because it is driver based it does indeed stress the HW. The kernel only needs to handle failures gracefully. The HW is stress tested.

I have started 1000s of machines running NTStress as a Windows HW tester so shut up.

 

[BaronMatrix]

Does running stress tests on your pc components in any way harm them?

Usually only if you're running out of spec (OC) or your components are not good quality. The average components will handle stress tests well as MS does lots of stress testing on all components for Windows.

It also depends on what the tests are stressing. Most components have a high MTBF (time before failure) so stress away.

Toms did several stress tests between AMD and Intel. They called them torture tests though.

You forgot the MEAN in MTBF, mean time between failure.... it is a statistical measurement. If you are going to attempt to throw around jargon, please get it right.

Also, MS does not stress test software --- HW is stress tested, software is beta tested and debugged.

This is what I mean Baron, you should not give out technical advice because you have no knowledge of the technical.

Ouch.


He's a stalker. He knows I worked at MS.

 

[BaronMatrix]

Becareful, Baron hardly knows how to spell PSU ....

http://www.fonerbooks.com/power.htm, decent diagnostic tree.

Also, you can test your PSU by selectively removing components and checking for stability. HDs consume between 20-40 watts, Graphics cards are getting up in the 80-150 watt range, MB/CPU together anwhere between 75 to 150 watts depending.

If you want more exotic, Fry's electronics sells PSU loads/testers, or you can find them online:
http://www.cyberguys.com/templates/searchdetail.asp?ProductID=11854

http://www.pcpower.com/products/viewproduct.php?show=VBT
http://www.trustedreviews.com/peripherals/review/2005/02/08/ATX-Power-Supply-Group-Test/p1

Of course, according to Baron -- I don't support any of my claims, so those links may not be informative.

I couldn't care less if anyone posts a link to anything. I left out the M cause I couldn't remember if it was mean or maximum.

Maybe you need a link to a life.

 

[BaronMatrix]

Were those 1000 machins 4x4s?? I guess that would make 4000 Virtual Machines, now I see your interest in 4x4 --- increase your sampling.

What was that about changing the subject when you know you're wrong?

 

[sirheck]

8O damn you guys are doggin each other like the mods are doggin
me in the community section.

 

[BaronMatrix]

8O damn you guys are doggin each other like the mods are doggin
me in the community section.

He's a stalker like jDODO. They can't stand for an AMD supporter to be right or even competent.

I think it's a personal problem.

 

[turpit]

Stress tests that use the HDD are HW based nitwit. Because it is driver based it does indeed stress the HW. The kernel only needs to handle failures gracefully. The HW is stress tested.

I have started 1000s of machines running NTStress as a Windows HW tester so shut up.

I thought MS used monkeys to push the 'on' button

 

[BaronMatrix]

8O damn you guys are doggin each other like the mods are doggin
me in the community section.

He's a stalker like jDODO. They can't stand for an AMD supporter to be right or even competent.

I think it's a personal problem.

No, not at all -- I have argued for AMD in cases where mis-info has been posted about them as well....

You, though lack any ability to see where you are wrong and admit it... and you continue to stretch out truth and mislead, misinform, and miscontrue the data to your own personal agenda.

You lack the ability to recongize your own limitations and often need to resort to nonsensical jargon to try to create the illusion you know what you are talking about. The sad thing is you actually believe yourself:

I shall remind:

http://www.apa.org/journals/features/psp7761121.pdf


I told you before, developers who work 12 hour days are lucky if they can remember their phone number. I have no doubt that a number here or there will be misplaced. Fortunately it rarely happens in my UIs and Frameworks.

Fortunately I'm in between projects so I can rest a bit.

Stalker!

 

[sailer]

Does running stress tests on your pc components in any way harm them?

Back to the original question, I don't think a stress test would harm any hardware unless it was already near the point of failure and the stress test pushed it over the edge. Normally, I would use a stress test to figure out where weak points are, or to find out if I had reached or gone beyond the limit of a machine's capability.

An example of this is when I overclock to 2925 mhz, I can run Prime95, a couple other programs and 3DMark06 without any problem what so ever, or just run Prime95 constantly for days with no problem. But when I go up to 2990 mhz, my computer seems to run things ok, but will fail on Prime95 within 20 minutes. Process of elimination has narrowed the problem part to either the video card or the motherboard. After Vista and the R600 comes out, I'll probably replace the video card and eliminate that variable when I upgrade.

Does the Prime95 failure mean that I can't run at 2990 mhz? No, but it does tell me that if the machine crashes, I should not get mad at the machine. Prime95 told me that there was a problem, so when the problem crops up, it shouldn't be unexpected. There are other stress tests that work as well, some being more specific in telling what's wrong. Use them, and don't be afraid of them.