Overclocking 3930k p9x79 getting bsod 124

[kenito]

Hi there!

I bought a used system with a 3930k and a p9x79,I have overclocked it and my problem is that I often get bsods, especially when I stream with OBS and play at the same time (then it crashes during the first 30 mins to 2 hours.

I tried clocking it moderatly (4,2, 4,5 ghz,) and I used the Asus guide to clocking it How to Overclock an Intel X79. No mather how I change my settings, it still crashes, I see in my config it runs at a 1.346v tops (with hwinfo) and settings it has 0.200+ in offsett. (I can get screenhots of the entire setup if anyone would like that)

Normally I can game reasonably and prime95, IBT runs smooth for a couple of hours without crashing (Yes I know I'm supposed to run them for 24~ hours)

I know it's not the PSU, not the RAM and not the GFX.

 

[myocardia]

Hi there!

Hi, welcome to Anandtech.

I know it's not the PSU, not the RAM and not the GFX.

How do you know this?

 

[dmoney1980]

Your voltage seems a but high for that overclock. I have a 3930k on a asus x79 deluxe board at 4.2ghz at 1.244v. I set this up in BIOS with a negative offset. Try lowering the voltage on your cpu

Here is a screenshot from the AI suite

 

[kenito]

Hi, welcome to Anandtech.



How do you know this?

Cause I've had 3 different PSU's, 2 different sets of ram and 5 different GFX cards in it.

Your voltage seems a but high for that overclock. I have a 3930k on a asus x79 deluxe board at 4.2ghz at 1.244v. I set this up in BIOS with a negative offset. Try lowering the voltage on your cpu

Here is a screenshot from the AI suite

Will it crash with bsod 124 with too much voltage?

 

[Burpo]

Check & make sure you have latest bios..

 

[kenito]

Check & make sure you have latest bios..

It is.

 

[chubbyfatazn]

BCCode: 124 0x00000124

The WHEA_UNCORRECTABLE_ERROR bug check has a value of 0x00000124. This bug
check indicates that a fatal hardware error has occurred. This bug check uses the error
data that is provided by the Windows Hardware Error Architecture (WHEA).

Just covering all bases... you've reset the clocks/voltage to stock and made sure everything was stable there? Memtested the memory?

Edit: might be worth checking out - http://www.overclock.net/t/1120291/solving-fixing-bsod-124-on-sandybridge-read-op-first

Got a minidump available?

 

[Jorge_Orwell]

Change your BIOS.
The latest is definitely NOT the greatest.
Something like that should do 4.4-5.0
without overheating and 0 crashes.

 

[kenito]

Just covering all bases... you've reset the clocks/voltage to stock and made sure everything was stable there? Memtested the memory?

Edit: might be worth checking out - http://www.overclock.net/t/1120291/solving-fixing-bsod-124-on-sandybridge-read-op-first

Got a minidump available?

Yes, I also swapped out the memory from a 4x4gb to a 2x8gb I had lying arround, same problem

http://slimbox.net/~kenito/CrashList.html

Change your BIOS.
The latest is definitely NOT the greatest.
Something like that should do 4.4-5.0
without overheating and 0 crashes.

Could I get any link that explains this certain problem?

 

[Jorge_Orwell]

Set your CPU/VTT at 1.14v
If that doesn't solve your problem,you're going to need to use the second-to-latest BIOS.
The one I had would not even boot with 3930K + latest BIOS.

 

[guskline]

What cooler are you using?

 

[kenito]

Set your CPU/VTT at 1.14v
If that doesn't solve your problem,you're going to need to use the second-to-latest BIOS.
The one I had would not even boot with 3930K + latest BIOS.

Could I get a screenshot or something of your setup?

What cooler are you using?

noctua nh-d15

 

[BonzaiDuck]

Set your CPU/VTT at 1.14v
If that doesn't solve your problem,you're going to need to use the second-to-latest BIOS.
The one I had would not even boot with 3930K + latest BIOS.

That was my first thought about it. He says he replaced a 4x4GB kit with a 2x8GB.

Stop Code 124 can occur for any number of reasons, so the careful troubleshooting is essential.

It can be a driver bug or driver conflict.

It can be a matter of mobo resources pushed too far.

It can result from RAM in conjunct with "VTT" (or VCCIO or VCCSA).

Make sure to disable all mobo "extras" like storage controllers and USB3 controllers not in use.

Do clean re-installs of all drivers -- including the gfx card(s).

And before any of that, reset the BIOS -- clear the CMOS! -- and start over from there. I would also take inventory of any proprietary software installed for the mobo. It's an ASUS board, so -- absolutely -- disable the "Turbo EVO" feature completely.

Also -- in conjunction with the ASUS software (AI Suite or whatever incarnation they call it for the p9X79) -- be absolutely sure that no more than a single hardware monitoring program is running at any given time. If so -- choose which monitor program you want to use for whatever purpose and time, and shut down every other monitor program.

I could be totally off-base with any suspicions I have, but you'll want to work through this "inventory" to nail down your problem there.

 

[kenito]

That was my first thought about it. He says he replaced a 4x4GB kit with a 2x8GB.

Stop Code 124 can occur for any number of reasons, so the careful troubleshooting is essential.

It can be a driver bug or driver conflict.

It can be a matter of mobo resources pushed too far.

It can result from RAM in conjunct with "VTT" (or VCCIO or VCCSA).

Make sure to disable all mobo "extras" like storage controllers and USB3 controllers not in use.

Do clean re-installs of all drivers -- including the gfx card(s).

And before any of that, reset the BIOS -- clear the CMOS! -- and start over from there. I would also take inventory of any proprietary software installed for the mobo. It's an ASUS board, so -- absolutely -- disable the "Turbo EVO" feature completely.

Also -- in conjunction with the ASUS software (AI Suite or whatever incarnation they call it for the p9X79) -- be absolutely sure that no more than a single hardware monitoring program is running at any given time. If so -- choose which monitor program you want to use for whatever purpose and time, and shut down every other monitor program.

I could be totally off-base with any suspicions I have, but you'll want to work through this "inventory" to nail down your problem there.

This problem only happens when it's overclocked, when I set everything back to default I can stream and game as much as I want.

 

[tweakboy]

First off. Go to bios and put everything on auto. Dont overclock it!!! Now go to desktop and do what you were doing for it to crash. If it runs smoothly then its the OC giving you the BSOD.

 

[tweakboy]

Maybe the guy you got it off was using this and has seen the BSOD but they never did anything about it. Why would he sell it .. That is a sad sidegrade

 

[TheUnk]

Do you know if the previous owner overclocked it? My 3930k was stable at 4.4 for a few years, now I have to run it under 4 or problems start, regardless of voltage.

 

[kenito]

Do you know if the previous owner overclocked it? My 3930k was stable at 4.4 for a few years, now I have to run it under 4 or problems start, regardless of voltage.

I'm pretty sure he did, and I know some parts suffered damage from unstable electricity (PSU and SSD, and maybe MB, but they got replaced)

 

[BonzaiDuck]

This problem only happens when it's overclocked, when I set everything back to default I can stream and game as much as I want.

. . . And I forgot to mention that, but it appears that you did the right thing for troubleshooting it. In my case, OC or default, I would have occasional resets or BSODs -- so occasional that it was difficult to replicate or otherwise nail down. The list of suggestions I provided -- one or the other, or in some combination -- got rid of it.

In your case, I see there are suspicions the hardware was damaged when you got it, so . . .

 

[kenito]

. . . And I forgot to mention that, but it appears that you did the right thing for troubleshooting it. In my case, OC or default, I would have occasional resets or BSODs -- so occasional that it was difficult to replicate or otherwise nail down. The list of suggestions I provided -- one or the other, or in some combination -- got rid of it.

In your case, I see there are suspicions the hardware was damaged when you got it, so . . .

If I can't get this working I have another 3930k laying around that I can try bought another used 3930k and a GTX 690 for 100$ hehe.e

 

[tweakboy]

SOmething had to be wrong for him to sell it for a cheap price. You shouldnt buy used elec stuff.. end of story.

 

[PhIlLy ChEeSe]

I have your same set up, when you over clock the memory(mine) I have to raise the VSSA also. You did not list your systems spec's, how are you cooling the CPU? I can over clock my board/CPU to 4500Ghz without touching the voltage. But I also check to maske sure my CPU isn't over or under volted in CPU-Z while running Intel Burn Test(I don't do it long).
I give .065+ to the CPU I can run 4700, think my ceiling is at 4800 with my chip as it used to hit 5400 when I had my RIVE board(it died in flame's) kinda cool too!:awe:
So I got a deal on this P9X79 in the box with a few bent pins in the Socket, seeing as my other board went up in flames I fixed this one. I have a phase change, But I don't see it doing to much to warrant putting it on this board. I swapped over my 2 Intel 520 SSD'S in RAID 0, only drivers i had to update were the chip set. My CPU is under custom water and always kept cool, here's a run I did with only 4800Ghz nothing else with summer temps in doors(better then 99%) not bad.
http://www.3dmark.com/fs/4617635

PS BSOD 124 have you looked it up? I think it's a heat issue are you monitoring your temps(if so how)? I use real temphttp://www.techpowerup.com/downloads/SysInfo/Real_Temp/

Nope its hardware or a Bad Bios, being as you have a 3XXX series CPU, I would go to far into the 4XXX as it's made for the second gen IB-E CPU'S like a 4930K for example.

 

[kenito]

Noctua nh-d14

I _might_ have made it stable now, 4.0ghz with +0.005 ofsett on vcore and vcca
VTTCPU to 1.10000 and CPU current capability to 130% in digi+ power control.
Now the temps was quite High, was gaming watchdogs maxed out at at like 59c and the vcore was at 1.321 logged with hwinfo

I have your same set up, when you over clock the memory(mine) I have to raise the VSSA also. You did not list your systems spec's, how are you cooling the CPU? I can over clock my board/CPU to 4500Ghz without touching the voltage. But I also check to maske sure my CPU isn't over or under volted in CPU-Z while running Intel Burn Test(I don't do it long).
I give .065+ to the CPU I can run 4700, think my ceiling is at 4800 with my chip as it used to hit 5400 when I had my RIVE board(it died in flame's) kinda cool too!:awe:
So I got a deal on this P9X79 in the box with a few bent pins in the Socket, seeing as my other board went up in flames I fixed this one. I have a phase change, But I don't see it doing to much to warrant putting it on this board. I swapped over my 2 Intel 520 SSD'S in RAID 0, only drivers i had to update were the chip set. My CPU is under custom water and always kept cool, here's a run I did with only 4800Ghz nothing else with summer temps in doors(better then 99%) not bad.
http://www.3dmark.com/fs/4617635

PS BSOD 124 have you looked it up? I think it's a heat issue are you monitoring your temps(if so how)? I use real temphttp://www.techpowerup.com/downloads/SysInfo/Real_Temp/

Nope its hardware or a Bad Bios, being as you have a 3XXX series CPU, I would go to far into the 4XXX as it's made for the second gen IB-E CPU'S like a 4930K for example.

 

[kenito]

I found the solution! disabling c-state 6 was the fix! And HWinfo were showing the wrong vcore and temps so now I finally got a decent and stable system at 4.2 ghz. I'll update rest of the stats soon! Vcore is down at 1.24 and temps arround 50c at full load!

*Edit I shouldnt have said that it just crashed again. I've tested with everything now, new CPU, GPU, RAM, The only think I havent tested without is DVD SSD/HDD and Motherboard.