BSOD running prime95 "Blend" torture test, error 0x00000124 (Win7 x64)

[Sniperm4n]

BSOD running prime95 "Blend" torture test, error 0x00000124 (Win7 x64)

Hi everyone,

I just finished my first computer build in 5 years last night (my previous one being an Intel Q6600 first-gen quad-core build), installed Windows 7 Ultimate x64, installed all of my core motherboard drivers via Gigabyte's website, updated my BIOS from version F3 to F7, and installed prime95, CPU-Z, and hwinfo. Please note that I installed nothing in addition to the aforementioned software (not even Windows updates). I then started a prime95 "Blend" torture test last night at around 9 PM. I periodically checked on the machine every couple hours until I went to bed around 1 AM. I then checked on it at 9 AM before I headed off to work and it seemed fine (at the 12 hour mark of the torture test). When I got home this evening, the machine was sitting at a logon prompt, and after logging in the machine told me that Windows had recovered from a serious error around 11:15 AM (the 14 hour mark of the torture test). I then checked out the Event Viewer -> System Log and found a 0x00000124 BugCheck.

Additional Hardware Info:
CPU Cooler: Noctua NH-U14S with a second added Noctua 140mm fan at the back
Case: Fractal Design Define R4 with 3 140mm Fractal Design fans
Add-in Cards: None for the sake of stress testing

Just for the heck of it, I've decided to bring Windows 7 completely up-to-date via Microsoft Update, praying that my issue can be resolved with a Hot Fix. Any help is greatly appreciated on this as I'm incredibly concerned I've managed to damage something =(. Please let me know if any additional information is needed!

Thanks,
-Snipe

 

[YoYo155]

Your CPU may be overheating:

Event[1179]:
Log Name: System
Source: Microsoft-Windows-Kernel-Power
Date: 2013-08-16T11:14:44.132
Event ID: 89
Task: N/A
Level: Information
Opcode: Info
Keyword: N/A
User: S-1-5-18
User Name: NT AUTHORITY\SYSTEM
Computer: 4770k
Description:
ACPI thermal zone ACPI\ThermalZone\TZ01 has been enumerated.
_PSV = 381K
_TC1 = 1
_TC2 = 5
_TSP = 1000ms
_AC0 = 0K
_AC1 = 0K
_AC2 = 0K
_AC3 = 0K
_AC4 = 0K
_AC5 = 0K
_AC6 = 0K
_AC7 = 0K
_AC8 = 0K
_AC9 = 0K
_CRT = 378K
_HOT = 0K
_PSL - see event data.

That was logged in the Event Viewer and correspond to the time of the bugcheck.

Download Speccy and post a screenshot.

• One while idle.
• Second while under intensive load.

Side Note:
You shouldn't use Prime95 to test a Haswell chip.
The instruction set is not updated to the newer architecture
I recommend using other tools.
Aida64 is not freeware but very much recommended.

​

 

[Sniperm4n]

Thank you for the prompt response YoYo155! Can you recommend a utility other than Prime95 that's free to stress my CPU for the second screenshot you've requested? I'll install Speccy and snap a screenshot at idle in the interim.

 

[YoYo155]

I'm positive that the Custom Builds and Overclocking section of the forums will be able to help you with stress testing Haswell .

But for now lets see how it handles temperature.

 

[Sniperm4n]

Thank you for politely pointing me to the appropriate resource, very new to this forum in case you hadn't guessed . Will review it now...

 

[Sniperm4n]

Speccy Idle Screenshot

Attached is a Speccy screenshot while at idle. Still working on figuring out the at-load screenshot. A quick note that the temp seems to jump from 33-37 (I happened to snap a screenshot at 35).

Thanks,
-Snipe

 

[YoYo155]

No rush, it's best to do the right thing.

 

[Sniperm4n]

I decided to install the 30-day trial of Aida64 Extreme 3.00.2571 Beta as it seems to be fully functional during the 30 days. I just started the stability test and the processor is running anywhere from 47-56. Would you still like to see a screenshot?

 

[YoYo155]

You can upload it if you like but I take your word for it.
In the post above, you mentioned it happened after several hours am I right?

 

[Sniperm4n]

Correct, the BugCheck happened at approx. the 14-hour mark of prime95x64 "Blend" stress testing (test was started around 9 PM last night and the BugCheck occurred at 11:14 AM this morning). I also just audited the System Log and verified that only one BugCheck has occurred. Whatcha think? I'm concerned I've damaged the proc and/or will have an unstable system.

 

[Sniperm4n]

Quick FYI that Speccy is still bouncing around between 48-57 (I haven't seen it go above 57).

 

[YoYo155]

Could you post a picture of AIDA and CPU-Z?

 

[Sniperm4n]

Certainly, is there some sort of summary screen you'd like screenshoted, or do you want a pict of AIDA in the process of stress testing, along with CPU-Z open?

 

[YoYo155]

Yes the statistics tab along side CPU-Z.

 

[Sniperm4n]

AIDA64 at Load

Excellent, thanks, screenshot attached! As a side note, I got bored and decided to install Firefox, FileZilla, and Foobar while I was doing other stuff . I also decided to leave Speccy open, and Task Manager's Performance tab open for the purposes of this screenshot.

 

[YoYo155]

A couple of things I can see.

3.3v = Low
12v = High
CPU not hitting boost clock of 3.90 .

Stop the test and post another screenshot.

 

[Sniperm4n]

AIDA64 Test Stopped

Test stopped, screenshot attached.

Thanks,
-Snipe

 

[YoYo155]

What is the make and model of your PSU?

 

[Sniperm4n]

I had a feeling that would probably be your next question . Here's some quick back story:

Due to the fact I didn't want to bring my primary Q6600 down during the build process, I decided to rip the PSU out of my "dead" HTPC (it has OS issues/a motherboard issue). Here's the make/model:

Make: Rosewill
Model: RG630-2 (Green Series)

The PSU I'll be putting into the final build is:
Make: Antec
Model: Neo HE650

 

[YoYo155]

That could be your culprit right there.

Also note that Haswell CPUs are more sensitive to PSU performance !!

• The big Haswell PSU compatibility list