• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

++ ATOT official NEF thread part IV ++

Page 526 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
This will probably be my last PC build. I'll probably upgrade the graphics card (either new or with Crossfire) next year but that will be it for me unless something drastic changes in PC gaming.
 
alfa, remember -- create a character on Elendimir! :awe:

Though if you guys all wanted to play on a server, I'd consider rerolling on a different server. That would be pretty painful though.
 
Any program that heavily stresses a CPU will be far more than adequate for me because I'll never hit even close to those usage numbers. Generally, I've seen that if it will BSOD, it will do so within the first hour or so. I could run it longer but why? If I start seeing crashes, I'll know the problem.

And yes, I used the multi-threaded version.

That's not how it works. If the system encounters an instability that you are unaware of it's possible that damage/corruption could rear its ugly head weeks or months later!

You do want to heavily stress using LinX finding the memory amount that produces the highest throughput (in GFLOPS) and select the amount of time to run. I suggest 300 minutes - 5 hours as start. Forget this 20 passes = stable nonsense. Alternatively you can install BOINC client and register for a project and allow it to crunch for 1-2 weeks. If you don't see computational errors in that time period (or system crashes) it's a pretty good bet it's stable enough for most (non critical) tasks. If you were doing the latter overclocking wouldn't even be on the table of consideration. (along with non ECC ram, etc.) 😉
 
That's not how it works. If the system encounters an instability that you are unaware of it's possible that damage/corruption could rear its ugly head weeks or months later!

You do want to heavily stress using LinX finding the memory amount that produces the highest throughput (in GFLOPS) and select the amount of time to run. I suggest 300 minutes - 5 hours as start. Forget this 20 passes = stable nonsense. Alternatively you can install BOINC client and register for a project and allow it to crunch for 1-2 weeks. If you don't see computational errors in that time period (or system crashes) it's a pretty good bet it's stable enough for most (non critical) tasks. If you were doing the latter overclocking wouldn't even be on the table of consideration. (along with non ECC ram, etc.) 😉

Too much work. 😀
 
Too much work. 😀

It is however think about how much work it will be (if even possible!) to fix/restore corrupted JPG pictures, documents, etc. that could be damaged due to cpu errata. Back in 2008 I had a so called stable system that never crashed that was used to edit videos in Avid. Only until later did I find out there were strange pixels randomly and glitches in some of the audio channels at random. It was caused by overclocking and no stability checker revealed it. Encoding the videos would, however. Fortunately this wasn't critical and I caught it before too many videos were produced.

It's always nice to find out the max o/c at safe vcore and work from there. Prime95 is definitely not what I would use though. I've had I7 systems that could prime for a week that would produce an error in LinX in minutes. That's just one example.
 
It is however think about how much work it will be (if even possible!) to fix/restore corrupted JPG pictures, documents, etc. that could be damaged due to cpu errata. Back in 2008 I had a so called stable system that never crashed that was used to edit videos in Avid. Only until later did I find out there were strange pixels randomly and glitches in some of the audio channels at random. It was caused by overclocking and no stability checker revealed it. Encoding the videos would, however. Fortunately this wasn't critical and I caught it before too many videos were produced.

It's always nice to find out the max o/c at safe vcore and work from there. Prime95 is definitely not what I would use though. I've had I7 systems that could prime for a week that would produce an error in LinX in minutes. That's just one example.

I really don't want to mess with this too much -- a few years ago I would've, but now, I really don't care and if the risk is too great, I'll slow it down. What is the latest version of this LinX?
 
Back
Top