P67A-UD4 Stability Problems

harlam357 · Jan 30, 2011

Having BSOD 124 issues with my UD4 and 2600K. I'm using two sets (4x2GB) of this RAM: http://www.newegg.com/Product/Product.aspx?Item=N82E16820231193

To preface, I was using only 2x2GB of this RAM and had a successful 14+ hour run of P95 Small FFTs (screen shot attached) at 45x100 (4.5GHz). Using 1.35v (bios) with LLC enabled. Thought I was good to go. So I popped in the other 2x2GB kit and started Folding. Then these BSOD 124 errors started happening, usually after 8~12 hours. Tried removing the second set of dimms but still got 124 errors, same time window. Have tried extra QPI up to 1.2v (bios) and that didn't help either. Not sure if the extra QPI is good for the overall life of the CPU either as this will be a 24x7 fully loaded machine.

Tried P95 Blend runs at 45x and they BSOD 124 almost immediately; so I backed down to 44x, here are my current settings. These are basically the same settings I used at 45x. Dimms are manually set 9-9-9-24 2T. This setup yielded a P95 Blend run of 6~7 hours before BSOD 124. So it's an improvement, but still not stable. Note that I've tried F6, F7b, F7f, & F7h. Currently on F6 because I don't believe I need the internal PLL overvoltage at the speeds I'm running and I'd like to be on an official bios if at all possible.

I'm really at my wits end with this rig and am even considering the possibility that there is a defect with my board or cpu. Any wisdom you can share is much appreciated.

Code:

### MB Intelligent Tweaker ###

Advanced Frequency Settings:

CPU Clock Ratio .......................................[ 44

> Advanced CPU Core Features

CPU Clock Ratio .......................................[ 44
Real Time Ratio Changes In OS .........................[ Disabled
Intel(R) Turbo Boost Tech .............................[ Disabled
Turbo Ratio (Core 1)...................................[ Auto
Turbo Ratio (Core 2)...................................[ Auto
Turbo Ratio (Core 3)...................................[ Auto
Turbo Ratio (Core 4)...................................[ Auto
Turbo Power Limit (Watts)..............................[ Auto
Turbo Current Limit (Amps).............................[ Auto
CPU Cores Enabled ............................ ........[ All
CPU Multi Threading....................................[ Enabled
CPU Enhanced Halt (C1E) ...............................[ Disabled
C3/C6 State Support ...................................[ Disabled
CPU Thermal Monitor ...................................[ Enabled (have tried Disabled too)
CPU EIST Function .....................................[ Disabled
Bi-Directional PROCHOT ................................[ Enabled (have tried Disabled too)

<

BCLK/DMI/PEG Clock Control ............................[ Enabled
BCLK/DMI/PEG Frequency Control ........................[ 1002
System Memory Multiplier (SPD) ........................[ 16.00

Advanced Memory Settings:

System Memory Multiplier (SPD) ........................[ 16.00
Performance Enhance ...................................[ Turbo
DRAM Timing Selectable (SPD) ..........................[ Quick
Channel Interleaving ..................................[ Auto
Rank Interleaving .....................................[ Auto



Advanced Voltage Control:

CPU

Load Line Calibration .................................[ Enabled
CPU Vcore .............................................[ 1.35
QPI/VTT Voltage 1.050v ................................[ Auto
System Agent Voltage 0.925v ...........................[ Auto
PCH Core 1.050v .......................................[ Auto
CPU PLL 1.800v ........................................[ Auto

DRAM

DRAM Voltage 1.500v ...................................[ Auto
DRAM Vref 0.750v ......................................[ Auto
DRAM Termination 0.750v ...............................[ Auto
Ch-A Data VRef. 0.750v ................................[ Auto
Ch-B Data VRef. 0.750v ................................[ Auto
Ch-A Address VRef. 0.750v .............................[ Auto
Ch-B Address VRef. 0.750v .............................[ Auto

Miscellaneous Settings

Isochronous Support ...................................[ Enabled (have tried Disabled too)
Virtualization Technology .............................[ Enabled (I need this)

redduc900 · Jan 30, 2011

I would give each a small bump in voltage, then one by one eliminate each as the source of the 0x124 BSODs, although I suspect either VTT or Vcore...

Code:

CPU Vcore: 1.36V
QPI/VTT Voltage 1.050v: 1.140
System Agent Voltage 0.925v: 0.935V
CPU PLL 1.800v: 1.900V
DRAM Voltage: 1.55V

harlam357 · Jan 30, 2011

I'm not going over 1.35v on the cpu core at this point. I think if the CPU can do 14 hours of P95 Small FFTs that somewhat proves CPU itself can run these speeds at this voltage; maybe just not the IMC. I've already backed the CPU down one multi from the P95 run I posted. I'll back down the multi further in lieu of adding more vcore at this point.

Tried higher QPI already... helped get me longer through a P95 run but still resulted in BSOD. Highest I've gone is 1.2v. I've read that no more should be required at the clocks I'm pushing. I'm also concerned about what is a "safe" QPI voltage for 24x7, heavily loaded use.

System Agent: bumped it up once too... didn't make a difference.

From what I've read the PLL shouldn't need any increase at these speeds. I haven't been fooling with it lately. Probably only messed with it when I was trying to find the maximum cpu clock I could boot into Windows with. I'll try this some more after my current P95 run completes, or more likely, dies.

DRAM is set to 1.5v and I'm using 1.5v dimms... it actually overvolts quite a bit already. Had no issues running memtest.

Not trying to shoot down anything you're saying here. I know there's no silver bullet when it comes to OC'ing a machine; been doing this long enough. Just wanted to get some thoughts from my OCF brethren to see what you guys think... and that it's not something obvious that I'm missing. I certainly do appreciate the response redduc.

I've also considered that this board my just not like this memory; and I don't have anything else to test with w/o robbing it from another machine that is already running.

repilce · Feb 9, 2011

Did you ever get this resolved? I would suggest trying changing the performance setting in memory options under MIT from "turbo" to "standard" if you havent already.

harlam357 · Feb 11, 2011

Nope... bought a pair of 2x4GB Jip Jaws that are on the boards approved memory list. Ran for two days under FAH load then got the same BSOD. Since I can't make this issue appear quickly I'm just running FAH on it until it BSODs... then tweaking a voltage here or there to see if it makes any difference.

So far I've tried....

- PLL up to 1.84v (actually BSOD'd quicker)

- QPI up to 1.10v and then 1.14v

I'm currently at all stock voltages except vcore @ 1.35v w/LLC and the QPI @ 1.14v.

If this doesn't work, and I expect it won't, I'll try your suggestion. I've thought about changing that to Standard but I haven't read that Turbo was causing issues for anyone; but my situation appears somewhat unique... so I've definitely give it a try.

Thanks repilce! :beer:

repilce · Feb 11, 2011

i've been having some "differences" with my setup as well actually, didnt realize the fact till after a week or so because i hadn't ran prime.. just IBT

I Think i've stumbled upon something with this setup, I went with LLC enabled, auto to everything including Vcore, i did manually set pll to 1.76 read somewhere some like lower on the <5ghz side, and kept turbo enabled, disabled real time changes.. set all tubro multis to 45 (have to set the main first) and upped the watts limit to 200 and the core current limit to 125 amps

Currently blending prime 26.5 atm, and i've noticed something that changed that so far has been a constant...variation.. that is lol..

with setting the oc with turbo enabled and power options upped, i have noticably lower vdroop, almost rock solid under load, no where dipping in difference it would before even with everything set manually.

o I'm hoping to see in about 8 hours no failures what so ever

edit:// also, load is 1.320 min 1.332 max

harlam357 · Feb 12, 2011

That's spot on what my setup shows in CPU-Z under load, 1.332v.

I haven't really experimented with the Turbo Boost stuff... since this rig is planned to run at full load all the time, I really have no use for it. But that's a nice observation on how these boards react with manual vs. auto settings.

My machine BSOD'd again... So I've moved everything back down to stock voltages except keeping vcore @ 1.35 (w/LLC) and I've set the memory performance to Standard as you suggested. Funny thing is... this has had zero effect on my FAH production. It's the same as it was with the Turbo setting. So if this setting gets me in the green then I have no issues with leaving it. Again, I feel my issue is memory related... so hopefully this will get it. :cross fingers:

If it doesn't I may give the lower PLL as shot as well... thanks for that tip!

wagex · Feb 12, 2011

yea alot of the overclocking guides say to shut down turbo mode on the ram if you are having stability issues.. but another thing in order for my board to boot past 4.4 ghz i had to update to f7b bios and have the pll over voltage enabled and now im sitting at 4.8 ghz and 1.36v

harlam357 · Feb 12, 2011

Man... what guide did you read? I saw no mention of this in any that I read. :bang head

I'm currently on F6... I can boot up into the 4.7-4.8 range but it just bombs the minute I try any stress testing. Machine is still running... that's one day. :cross fingers:

P67A-UD4 Stability Problems

harlam357

Senior Fold-a-holic

Attachments

redduc900

Inactive Moderator

harlam357

Senior Fold-a-holic

repilce

Member

harlam357

Senior Fold-a-holic

repilce

Member

harlam357

Senior Fold-a-holic

wagex

Chapstick Eating Premium Member

harlam357

Senior Fold-a-holic

Similar threads