hgfla Posted October 5, 2007 Share Posted October 5, 2007 I've been having a devil of a time trying to OC a known good E6600 that easily ran 400x9 in a P5W DH on my Striker. Thinking it was my new dominators that was the problem (they wouldn't run in the A1 and A2 slots without memtest86+ erroring right away in test2I temporarily swapped them out with a set of XMS2s (6400C4s) from the P5W and set for DDR2 800 and the very relaxed 5-5-5-15s left over. Didn't make a difference. But when I pulled them out they were hot, I mean really really hot, much hotter than they run in the P5W DH. VDIMM set for 2.2v in both. So I put the 8500C5Ds back in but this time and set VDIMM to 1.9 and strangely enough it worked fine, then to the lowest setting, 1.85v, and it's still banging away nicely and in A1/A2 or B1/B2 or whatever. Memtest86+ passing with flying colors, orthos now 90 minutes into its thing with lots of other stuff going on in the box and no hiccups yet. Does anyone know where on the board I can stick a VOM and see what's actually going on here? I don't think the Striker's VDIMM control is anywhere near correct. Could it really be running as much as 250mV too high?? Link to comment Share on other sites More sharing options...
DerekT Posted October 5, 2007 Share Posted October 5, 2007 You may wish to visit Xtremesystems.org http://bandwidth.se/imgs/1/25/4ac42213mem_measure.jpg Eva2000 can give you quite a bit of information regarding this problem. Link to comment Share on other sites More sharing options...
hgfla Posted October 5, 2007 Author Share Posted October 5, 2007 You may wish to visit Xtremesystems.org Eva2000 can give you quite a bit of information regarding this problem. Awesome - *Thank you* . Was that in a P5N32 thread? I thought I'd found all of the 680i threads. (do you still have the thread # handy?) Link to comment Share on other sites More sharing options...
DerekT Posted October 5, 2007 Share Posted October 5, 2007 Sorry, I kept the link for reference purposes but not the thread. Link to comment Share on other sites More sharing options...
hgfla Posted October 6, 2007 Author Share Posted October 6, 2007 I'm going through the rest of Eva2000's posts now. There's a bunch of other good, xtremesystems-like threads at i4memory.com and if I find the particular thread I'll post it back here. The image was very helpful but the discrepancy didn't measure out to be anywhere near as bad as I thought, thus the wish to find the thread and the other comments. I was only about 33mV on the high side. Good news: The Qimondas (?)(whomever Corsair using in the 8500C5D version 2.2s) cheerfully run at 1.88V measured at the advertised parms and at 800Mhz 4-4-4-12 1T. I bet they'll run even tighter when fed 2.1V like they're supposed to get. This memory rocks Those seeking to measure it are probably going to find their choke also coated with clear insulating varnish right on down to the solder pad so a micro clipon doohickey probe won't work here as easily as it does say to measure Vcore at the uninsulated Vcore choke. Fortunately the drain/collector/whatever the tab is on those three pin devices off to the left of the choke in that image also has VDIMM. Link to comment Share on other sites More sharing options...
DerekT Posted October 6, 2007 Share Posted October 6, 2007 Good news: The Qimondas (?)(whomever Corsair using in the 8500C5D version 2.2s) cheerfully run at 1.88V measured at the advertised parms and at 800Mhz 4-4-4-12 1T. I bet they'll run even tighter when fed 2.1V like they're supposed to get. This memory rocks Thanks for the heads up regarding the collectors and the Qimondas. I purchased a set of them yesterday for testing purposes. I also like the idea of the 1.9V. My ASUS Commando had a 0.5 Vdimm discrepancy. Link to comment Share on other sites More sharing options...
hgfla Posted October 7, 2007 Author Share Posted October 7, 2007 Thanks for the heads up regarding the collectors and the Qimondas. I purchased a set of them yesterday for testing purposes. I also like the idea of the 1.9V. My ASUS Commando had a 0.5 Vdimm discrepancy. Enjoy them. DDR2 800 with 4-4-3-3 1T @ 2.1V, advanced parms at auto, with the Qimondas. Stable 24 hours and going Link to comment Share on other sites More sharing options...
DerekT Posted October 7, 2007 Share Posted October 7, 2007 There is another benefit. I have two systems with Commando motherboards and the issue runs with P5K as well. The issue being a loss of CMOS settings due to crash or power loss. The high end Microns will not boot. One needs to install a single vanilla 1.8v stick and reboot, enter the BIOS, set the Vdimm to 2.1v and shut down. Replace the Microns and reboot. Then the system will start again. These Qimondas do NOT have that issue. :D: They restart with a CMOS settings loss. A major plus for those who are not all that computer savvy. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.