• 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.

Weird stability problems?

TheJTrain

Senior member
After everything I've gone through, I thought it might be valuable to others to document the testing I went through to finally figure out what was wrong with my new system. I assembled a new mobo, CPU, HSF, HD, and RAM (PC3200) and proceeded to get very frustrated when WinXP kept hanging at different points in the install process - sometimes it would hang during HD formatting, sometimes during file copying, sometimes before it even examined the system. At first I thought it might be that I needed to load third-party SATA drivers, but the fact that the BIOS recognized the SATA drive and that WinXP install saw it and tried to format it made me think otherwise.

With it hanging in a different place every time, I thought it might be the RAM. Ran memtest86 and got numerous failures, on both sticks together and on single sticks. Ran memtest86 on my old RAM (PC2700) and they passed with no problem, single and dual channel. WinXP installed without a hitch - bingo, bad RAM. RMAd them without much thought, but then memtest86 failed numerous times on the new pair (but only in dual-channel, they passed when run separately).

Found some info on the web that boards can be very picky on running dual-channel, with 1) which memory it will take, 2) at what timings, 3) at what speed, and 4) at what voltage. So I proceeded to run memtest86 in every configuration I could think of, and here's what happened (all of the following are dual-channel):
400MHz, 2-3-3-10T timings (RAM's rated timings), 2.6v (default) - FAIL
400MHz, 2.5-4-2-8T timings (the BIOS's "fail safe default), 2.6v - FAIL
333MHz, 2.5-4-2-8T timings, 2.6v - PASS
333MHz, 2-3-3-10T timings, 2.6v - PASS
400MHZ, 2.5-4-2-8T timings, 2.7v (default+.1v) - PASS
400MHZ, 2-3-3-10T timings, 2.7v (default+.1v) - PASS

So apparently, the PC3200 GeIL low latency RAM needed a little extra juice from the Chaintech VNF4Ultra mobo to run stably at PC3200 speeds. Interestingly enough, my old PC2700 Kingston HyperX (not sold as a packaged dual-channel pair) ran great on the same mobo at 400MHz with 2-3-3-10T timings at default voltage. My system, now running at that last configuration in the above list, has shown zero stability problems, nor have I seen any temperature issues as a result of running at over default voltage (nothing else in the system is overclocked - trying to go as silent as possible).

So long story short, when you're having stability problems running in dual-channel, try adjusting all of those factors (brand name, speed, timings, voltage) and re-running memtest86.

Jason
 
Back
Top