i2500K BSOD

silverf0x · 25 Apr 2012 at 00:19

Hi Guys,

My system:

i5 2500K OC @ 4.6Ghz - voltage is set to 1.37V
Asus P8P67 Motherboard
4GB Kingston HyperX ram
1280MB GTX 570
600W Corsair PSU

Problem I am having is that when the CPU is under load, it is hitting temperatures of around 78C and then I get BSOD. This usually happens around 30 minutes into prime95, or after a little while playing games.

I bought this from another company with a guaranteed OC to 4.6GHz as I just wasn't comfortable doing it myself at the time.

It's been back to the company 3 times with this issue, and since then they have replaced the motherboard and graphics card.

What do you think is causing the problem? Let me know if you need any more info.

Thanks for the help.

MeatLoaf · 25 Apr 2012 at 00:23

Sounds to me like either the CPU or memory needs more voltage to be stable.

Is it stable at stock speeds?

silverf0x · 25 Apr 2012 at 00:35

yes was running fine at stock. Only when i enabled the oc profile did this start happening.

Justintime · 25 Apr 2012 at 06:48

Likely vcore or VTT, what BSOD? 124, 101 etc..

silverf0x · 25 Apr 2012 at 09:41

Justintime said:
Likely vcore or VTT, what BSOD? 124, 101 etc..

here is an analysis of the dump, all the errors look like this:

This was probably caused by the following module: hal.dll (hal!HalBugCheckSystem+0x1E3)
Bugcheck code: 0x124 (0x0, 0xFFFFFA8004943028, 0xBE200000, 0x5110A)

Error: WHEA_UNCORRECTABLE_ERROR
file path: C:\Windows\system32\hal.dll
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: Hardware Abstraction Layer DLL
Bug check description: 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).
This is likely to be caused by a hardware problem problem. This problem might be caused by a thermal issue.
The crash took place in a standard Microsoft module. Your system configuration may be incorrect. Possibly this problem is caused by another driver on your system which cannot be identified at this time.

Garyj881 · 25 Apr 2012 at 15:42

BSOD Codes for SandyBridge P67,Z68

0x124 = add/remove vcore or QPI/VTT voltage (usually Vcore, once it was QPI/VTT)
0x101 = add more vcore
0x50 = RAM timings/Frequency add DDR3 voltage or add QPI/VTT
0x1E = add more vcore
0x3B = add more vcore
0xD1 = add QPI/VTT voltage
0x9C = QPI/VTT most likely, but increasing vcore has helped in some instances”
0X109 = add DDR3 voltage
0x0A = add QPI/VTT voltage

Justintime · 25 Apr 2012 at 17:43

'Likely' the cpu core voltage, add 2 notches up.

silverf0x · 27 Apr 2012 at 00:06

Justintime said:
'Likely' the cpu core voltage, add 2 notches up.

Thanks for the advice guys, will get back to the company and check what they will do.

DoneADougalOnSofa · 27 Apr 2012 at 09:44

silverf0x said:
Problem I am having is that when the CPU is under load, it is hitting temperatures of around 78C and then I get BSOD.

Upgrade, or at least check your CPU cooler- this is way too hot! Anything over 70C will likely cause you problems...
I can MilkyWay flat out at 4.5gig and max temps are 57-61C.
Check the cooler is seated correctly, and you are using decent thermal paste. If you are using the Intel stock cooler, it's time to upgrade...!
I'd be doing this before upping any voltages (which will only increase the heat!).

Rusty0611 · 27 Apr 2012 at 09:47

dougal1331 said:
Upgrade, or at least check your CPU cooler- this is way too hot! Anything over 70C will likely cause you problems...

This is bad advice. You aren't going to get problems going above 70c at all. It's not ideal but to say you're going to get problems is just incorrect.

chillidog · 27 Apr 2012 at 13:43

silverf0x said:
Hi Guys,

My system:

i5 2500K OC @ 4.6Ghz - voltage is set to 1.37V
Asus P8P67 Motherboard
4GB Kingston HyperX ram
1280MB GTX 570
600W Corsair PSU

Problem I am having is that when the CPU is under load, it is hitting temperatures of around 78C and then I get BSOD. This usually happens around 30 minutes into prime95, or after a little while playing games.

I bought this from another company with a guaranteed OC to 4.6GHz as I just wasn't comfortable doing it myself at the time.

It's been back to the company 3 times with this issue, and since then they have replaced the motherboard and graphics card.

What do you think is causing the problem? Let me know if you need any more info.

Thanks for the help.

your vcore might need to be bump up a bit may be 1.38/9 to get it stable.
as to temps,for 30 mins into prime seem to be little high but this could depends on what cooler you have got.but 78 is high ,but still ok

DoneADougalOnSofa · 28 Apr 2012 at 10:36

Rusty0611 said:
This is bad advice. You aren't going to get problems going above 70c at all. It's not ideal but to say you're going to get problems is just incorrect.

Yeh, OK maybe not "problems", as such, but everything I've seen since joining these forums has recommended keeping the CPU temp below about 70C under load. I'd say this is a good rule-of-thumb target for any (SandyBridge) OC. If you disagree with that, then fair enough.
I just think that the OPs temps are a tad on the high side for the speed. Would it really hurt to have a quick look at the cooler? Hardly bad advice to check it on the off-chance, IMHO.

I try to help where possible, I'm only going on what I've learnt so far- mostly from these forums!

Like any human being I don't get everything right, but I stand by what I say in this case.
Always an option to ignore me...

http://forums.overclockers.co.uk/images/smilies/frown.gif

DoneADougalOnSofa · 28 Apr 2012 at 10:41

... I've also suffered BSOD when mine hit the high 70s in my old case...
Sorted the cooling=no BSODs.

Hmmmm...

Justintime · 28 Apr 2012 at 10:48

70C is not hot at all and more than often you'll find 'stock' configs with stock cooling hitting 80C under full load easily in OEM machines etc.. with not one detrimental effect over years of running. Keeping it cool is good, but theres absolutely no problem with hitting say 75-80C with say ibt as it would do no damage, and the cpu will not likely see that sort of load anytime soon during normal usage.

DoneADougalOnSofa · 28 Apr 2012 at 14:16

Fairy nuff.
The OP says that he gets BSODs at 78C, even after playing games for half an hour.
The BSOD message says the problem could be thermal.

True, permanent damage is highly unlikely, and for Prime/IBT etc then you'd expect high temps, but games should not tax the CPU as much.
I see no point in NOT checking the cooler- two minutes work, and free...

You can't be going on with BSODs every time you play a game, so surely ruling out any potential causes would be a good start?

Justintime · 28 Apr 2012 at 15:28

His problem is likely 4.6Ghz is just not doable with 1.37v, contrary to what a lot of people say, 4.6Ghz is not as common to do within that voltage range, many who claim stability are using outdated programs such as LINX/Prime/IBT that don't use AVX extensions etc.. and pass or just don't stress test enough or at all, for example i can run for weeks with prime 26.6 with some rigs at 4.6Ghz but BSOD within hours with 27.4 as it actually uses AVX extensions and taxes the memory controller a lot more, more vtt/vcore or reducing to 4.5 solves this.
As for temps, i test at 30C ambient for worst case scenarios and hit about 85C with both prime and LINX with no ill effect if the volts are right and having been doing this for yrs and have done with close to 150 Sandybridge based chips since release, hence me thinking its not temp related.

silverf0x · 28 Apr 2012 at 20:59

Justintime said:
His problem is likely 4.6Ghz is just not doable with 1.37v, contrary to what a lot of people say, 4.6Ghz is not as common to do within that voltage range, many who claim stability are using outdated programs such as LINX/Prime/IBT that don't use AVX extensions etc.. and pass or just don't stress test enough or at all, for example i can run for weeks with prime 26.6 with some rigs at 4.6Ghz but BSOD within hours with 27.4 as it actually uses AVX extensions and taxes the memory controller a lot more, more vtt/vcore or reducing to 4.5 solves this.
As for temps, i test at 30C ambient for worst case scenarios and hit about 85C with both prime and LINX with no ill effect if the volts are right and having been doing this for yrs and have done with close to 150 Sandybridge based chips since release, hence me thinking its not temp related.

The OC was done by a company and 'guaranteed' to work.
I ran the system fine for around 8 months on a profile that did NOT have the OC enabled. I tried the OC one or twice, but I was hitting desired performance without it so left it off.
After an unrelated motherboard failure, they set the OC profile to be active and that's when I started getting the problems.

The cooler I have is a Gelid Tranquilo, and there are fans at the back of the system and at the top, so don't think it is a case of my case environment causing a problem.

The temp seems quite high, but it is the guaranteed BSOD straight after that is causing me problems.

Justintime · 29 Apr 2012 at 03:23

Many of those 'guaranteed' clocks from various companies were just banged into the rig with relatively 'known likely to work' settings applied across the board, not individually tested, or not tested for any reasonable time to iron out bugs. Try 4.5Ghz with the same settings and test with prime.