4500mhz nice
this is my new settings for 4861mhz on 5690:
 
 

i use 1.45volt in bios with no problems.
you need really great watercooling for the cpus
 
 

What OS are you currently using? I've been having stability issues with Windows recently. I was using Windows Server 2016 and switched to 10 Pro for Workstations about a month ago. Ever since then, I blue screen like crazy with whea uncorrectable errors. I have since been switching back and forth between 7, 8.1, Server, and now on LTSB 2016, but still will blue screen even if I don't do anything, like while loading Windows. Sometimes its constant, sometimes 4-5 times a day, sometimes only once. Its not consistent with anything I do. All BIOS settings are default and I have bumped up voltage a little even on stock clocks. I think it is something to do with a Windows update and I thought it was the fall creators update, but even switching to LTSB 2015 it still crashes. I have ZERO problems under Linux. I can run at 4.5Ghz all day and do my encoding work and never crash. I only need Windows for gaming and would run it in a VM but have not been successful at all with GPU passthrough.

splinter182

d wray
What OS are you currently using? I've been having stability issues with Windows recently. I was using Windows Server 2016 and switched to 10 Pro for Workstations about a month ago. Ever since then, I blue screen like crazy with whea uncorrectable errors. I have since been switching back and forth between 7, 8.1, Server, and now on LTSB 2016, but still will blue screen even if I don't do anything, like while loading Windows. Sometimes its constant, sometimes 4-5 times a day, sometimes only once. Its not consistent with anything I do. All BIOS settings are default and I have bumped up voltage a little even on stock clocks. I think it is something to do with a Windows update and I thought it was the fall creators update, but even switching to LTSB 2015 it still crashes. I have ZERO problems under Linux. I can run at 4.5Ghz all day and do my encoding work and never crash. I only need Windows for gaming and would run it in a VM but have not been successful at all with GPU passthrough.

WHEA uncorrected errors are cpu related. with my overclock I get whea error if my cpus decide to recognize all 12 sticks of ram. if they decide to not recognize 2 sticks I'm stable in windows which happens most of the time(working on getting better ram). have you tried running stability tests on your overclock? Also make sure your windows power profile is set to high performance.

I know the whea is CPU related which is why I went back to default clocks. I can get with single 8gb stick, 3 sticks, 6 sticks or all 96gb. I've tested all with memtest and CPU with prime, Aida, and occt. It's like once it happens, it constantly does it and I do a reinstall and I'm good for a few days. Then once it happens again it's constantly doing it. That's why I was thinking it's something with a Windows update. Using the latest ISO wouldn't even install without error. It never happens under my Debian install. CPUs are on water and board has good air flow all around so temps are good.

I just blue screened now with critical process died error doing a file transfer...

Hey everyone,

I’m running X5690s, with 6x8gb, corsair dominator platinum 1866mhz, 1.5v, 9-10-9-27 Ram.

I’m currently running an overclock at 4.26mhz with 185x23.
(This seems more stable than getting to 4.25mhz with 170x25, my previous attempt)

Voltages:
Vcore = 1.375v
Vtt = 1.35v
Dimm = 1.55v
MCH Strap = 1867mhz
CPU Uncore = 4800mhz
Memory frequency I have it at 1333
Memory timmings are all on Auto.

I was getting random restarts with the stock 26 multiplier, I just couldn’t get it to be “stable,” and lower multipliers with higher base clocks seems to stop the random restarts.
Other issues I figured out was raising the dimm voltage from 1.5 to 1.55 and now all 48gb posts every single time no problems. Lower multiplier/higher base clock also seemed to help with posting full ram all the time.
(Before this I had lots and lots of issues with missing ram)

The machine does not restart or crash under no load (checking email, web browsing etc) My new issue however, is that if I put both CPUs under 100% constant load, eventually my software crashes after about 15min. (4.26mhz at 185x23)

I’m not sure if it’s memory timings (I know very little on that), if it’s voltages I’m not getting right, or other things I’m not as sure of such as MCH Strap and so on.

This is a 3D rendering and simulation machine, so it needs to be able to run under 100% load for days and sometimes weeks. (Iv done so with x5650 CPUs for the last 7 years, I just upgraded to the X5690s)

Does anyone have any ideas of what I could do to fix the crashes of full CPU load?

Ok, thanks Robbanswe.
The last pictures you posted in this thread, I presume are the setting you are using with your Samsung ram and that overclock?

So I found the ram online, but there is a small discrepancy in part numbers...
 
Yours is: M393B1K70DH0-CH9Q9
Online its: M393B1K70DH0-CH9
 
It looks like the "Q9" is not on the second one; the one that is available for purchase.
Is there is a difference to be concerned about? 
Chris
 
 

guys, can someone re-post the OC TEMPLATE?