TR Forums

I'm looking for a recommendation for a good program for stress testing a machine that will keep track of freaking everything. Free is preferred, but if I have to fork out some cash, so be it.

I have three machines, out of 25 just like them, that are refusing to stay up. This is after meticulously replacing every part, every other part, etc. I need to figure out what the deal with these things are. Any help is appreciated. Thanks!

Every part?

Cuz its sounds like you're in a world of "hurt" over there.. No offense lol. I dont envy your situation, needless to say.

Uhhhh, yea.... Evry part? PSU???

Are these "enthusiast" built or "boutique" PC s ??? Or the Bestbuy/WalMart specials...(im including Dell and HP etc etc in this category also )

EVERY part. Like, even so far as taking transferring to another friggin' case. I think the base systems were ordered and then have been tweaked a time or two with new RAM and maybe a GPU, but yeah, every part has been replaced and as I said, dual parts as well. So one time the GPU, then the GPU and RAM. Then the CPU, then the CPU and RAM. We have an image we load onto the machines and that's been ruled out too. Of course, this is all second hand information. This is MY first time working on them... so I'm hoping that my predecessors were either sloppy or missed something very obvious.

There was a warez site, FOSI, which used to have a burn-in/system stress/information overload utility, but I can't recall the name of it and the site has since shut down. I do remember I used it quite a bit in years past, but really I just need something that gives me about the same level of data.

Wow..hmmm. That sucks. You may have hit the nail on the head with the "predecessor" thing. Sounds like your in IT or something. Or at least acting in that capacity for friends/family. But that DOES NOT sound like fun.

My only advice I could give you, is NOT very good. Hopefully someone MUCH more knowledgeable than myself will chime in here and offer some help. Shoot, you yourself probably have more tech know-how in your little finger.

If i WAS in your situation however, and ALL i had was my very limited hardware/software knowledge. I'd probably be doing an obscene amount of intelligent trial and error. Keyword in intelligent. lol. Like i would literally grab one of the "good PC s" and then one of the "bad" ones. And then (very inefficiently) take a SINGLE part from the "bad" and install in the "good", and then see what happens. Literally one at a time. Ram module to ram module..psu to psu...Rinse and repeat. But i like said. There HAS to be a more efficient way of troubleshooting. I wouldnt know unfortunately. And it kinda sounds like you may have done that already. Sry sir. Good luck to ya

Edit:
Also...may also want include more details as to what not "staying up" means. I get you. Just not very specific. For instance, does the PC post...boot into windows (or whatever OS ur running)...does it sit idle at desktop screen w/o issue. Or do the issues only manifest under "load"? Or even a specific program? Im sure this info would help any other TR members who wanted put their two cents in.

Love the sig btw

It's hard to miss hardware problems with a two-pass diagnosis from an overnight Memtest x86 followed by a long burn of simultaneous Furmark, Prime95 and looping robocopy moves.
All tools are free

I can recommend various stress test programs, but they're more targeted rather than the "test and log everything" approach. I am a big fan of the free "Open Hardware Monitor" which lets you log temps, voltages, and CPU/GPU utilization which can be helpful in conjunction with other stress test programs.

Would be helpful to know more about how these cursed systems are behaving - spontaneous shutdowns, reboots, BSODs, etc?

Yeah, you've got to assume that your predecessors didn't know what they were doing, and check/test EVERYTHING again. While you're doing that though, consider your options if it turns out that everything they said was correct. If that were the case, I would start looking at the incoming power to the problem machines. What sort of UPS/surge supressors/power conditioning are they using? Are they all on the same circuit, or located in a particular part of the building? Are the outlets properly grounded and supplying the correct (and steady) voltages?