Project 8900 WORK_QUIT (404)

Moderators: Site Moderators, FAHC Science Team

mdk777
Posts: 480
Joined: Fri Dec 21, 2007 4:12 am

Re: Project 8900 WORK_QUIT (404)

Post by mdk777 »

However, in my case the 12v output as measued by GPU-Z was very unstable on the system that kept crashing.
The Corsair HX1050W PSU in my AMD system seemed to be all over the place 0, 12, 12.3, 5, 2, 0, on/off, a mess.
Yeah, again not saying you are wrong, or that the PSU was not the cause of your problem;

but just a note to others: MB reporting of voltage is highly variable. Some do it well, others report complete BS.
just saying: if your system isn't crashing, don't run out and buy a new PSU based on the MB voltage reporting alone. :mrgreen:

PS

If you are in the market for PSU.

FOLDING is unique in the continuous load placed on your system.
As many have noted: if you are going to fold on GPU, and especially multiple GPU, it pays to get the best PSU you can afford, and error on the side of over-provisioning. :mrgreen:
Transparency and Accountability, the necessary foundation of any great endeavor!
FastFlyingVirginian
Posts: 22
Joined: Sun Feb 12, 2012 4:09 pm

Re: Project 8900 WORK_QUIT (404)

Post by FastFlyingVirginian »

I hope you can forgive resurrecting the thread, but I think I finally found the solution and wanted to provide the epilogue - it was the RAM. I had been running the system with all four slots populated with 4GB each (DDR3-1600). While this setup worked on my old i5-860/P55 system, it was generating some bizarre behavior with my i5-4670K/z87 setup. It looked to me at first like a networking issue - many websites inaccessible at random times or forgetting I was signed in, file uploads either not starting at all or becoming corrupted, access to other computers on my home network being unpredictable - but after checking everything having to do with the network several times over I was no closer to solving the problem. There were no BSODs or outright crashes, just odd network-related behavior. When I got to the point I had eliminated everything in the system but the CPU or RAM, I tried running the system with only 2 sticks of RAM at a time and voila, the problems disappeared. I've since cranked out several p8900s with no issues whatsoever, and after upgrading with 2 8GB sticks and moving the old RAM to another system, everything's still stable.

- F.F.V.
bruce
Posts: 20824
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: Project 8900 WORK_QUIT (404)

Post by bruce »

Thank you for the report, FastFlyingVirginian.

FAH does not work well if a system is not stable Because FAH uses all unused resources, it often runs hotter than even some benchmarks (like prime95) and will find instabilities that may be missed otherwise. It doesn't matter whether the instabilty is due to overclocking (usually the first assumption) or due to marginal hardware (in your case, the RAM). FAH most certainly needs a stable system.
Post Reply