Windows 7 - Kernel-Power (Event ID: 41 Task: 63 )

[BearCave54]

So yet another update.
The testing with Prime95 went fine without any errors, all the tests were passed.
Also, I suppose its worth mentioning that I ran memtest86+ and when I try to do a Probe test on the ram, two out of the three sticks never make it past 50% and memtest86% locks up completely, forcing me to restart the machine.

I've went ahead and ordered some new ram through newegg, 12gb worth of G.Skill Ripjaws that I've made certain was on the compatible / certified list for my evga x58 SLI and it should be here by the middle of next week.

 

[Jonathan_King]

Apologies for the delay...I was busy all day yesterday and am now working on my threads in order of oldest-newest.

Let us know how the RAM replacement goes; it does seem as if something is funny, given the Memtest issues.

 

[BearCave54]

Alright, I'll do that.
Thanks to everyone for all the help, and I'll post how the RAM replacement goes, good, bad or indifferent.
And if nothing else, I've learned a great deal from this about my computer, bios settings and error codes.

 

[profdlp]

... if nothing else, I've learned a great deal from this about my computer, bios settings and error codes.

When you run into someone around here who really seems to be on top of those kinds of things, you can be pretty sure that that is is exactly how they came to be such an expert. Take solace in the fact that you're well on your way.

Good luck with the replacement RAM!

 

[BearCave54]

So here's the update, which I'm sad to say is not what I was hoping for.
The replacement RAM hasn't helped the situation, which is the bad part.
The good part is that it hasn't made the situation any worse. Which I guess is a good thing. xD
At this point I'm looking pretty hard at my PSU and considering replacing it, as I'm quickly running out of options as my machine continues to shut down while I'm playing games like Rift, TF2, SC2 and the like.
To recap, that is the only time it does this. My heat temperatures are nothing too exciting, at most they run at the mid-40s on full load. And my gpu doesn't go above the same. I'm going to go get some compressed air and use the old pencil+can o' air trick to see if getting everything clear will help as well, even though I've watched the fans I have installed physically function, I know that dust can still affect airflow.

Yes, I might be grasping for straws here.

 

[Jonathan_King]

Hi Shannon,

I'm pretty confused, I'll be honest. The jcgriff2 report was supposedly generated on April 14, but many files are dated from March 12, the date you created your original report.

Please delete all jcgriff2 files on your system, save the actual .exe file. Run that again, and upload a fresh report.

 

[BearCave54]

Alright, so I went back and deleted the jgriff folder and ran the .exe again.
Did the same with the Perfmon report and here is the result. I'm hoping this will shed some light on to what the h-e-double hockey sticks is going on in this machine of mine.

 

[Jonathan_King]

Thanks a lot!

I'm still seeing that nasty error:

Event[1539]:
  Log Name: System
  Source: Microsoft-Windows-Kernel-Processor-Power
  Date: 2011-04-20T01:03:32.775
  Event ID: 35
  Task: N/A
  Level: Error
  Opcode: Info
  Keyword: N/A
  User: S-1-5-18
  User Name: NT AUTHORITY\SYSTEM
  Computer: BearCave
  Description: 
Performance power management features on processor 7 in group 0 are disabled due to a firmware problem. Check with the computer manufacturer for updated firmware.

Do you know of a way to disable specific cores of your CPU from the BIOS? It may be a redneck solution, but try disabling the 8th core. Windows is reporting it as Core 7, but remember it starts counting at 0, not 1.

The only other solution I can think of is replacing the motherboard or CPU. The motherboard would be the more logical replacement, but if you got an older CPU, it might also work.

My theory is that when gaming, that 8th core is used, causing it to overheat since power management features are not working for it. You can try to confirm this theory by running the Small FFTs and Large FFTs Prime95 tests, while watching Speedfan. See if you notice a temperature difference between the cores.

http://www.sevenforums.com/tutorials/100352-cpu-stress-test-prime95.html

 

[BearCave54]

I'm afraid I don't know how to do that, Mr. King.
If you have a tutorial or something to do so with, it would be appreciated.
Also, would that be an issue that could be associated with Speedstep in the Bios? Could Speedstep be causing the 8th core to be unstable or overheat maybe?

I've ran the stress tests before with Prime95 and it hasn't caused any real difference in temperature between the cores, but I'll run it again tonight and see if it corrects the issue.

Nevermind, I found the information through google, using msconfig and such.
Also, I used the tutorial I found, and went into the Advanced Options under the Boot tab and it had the number of processors listed as 1, with no checkmark and it was greyed out.
Does this mean that my machine has been using 1 core instead of the full 8 that it has access to?

 

[Jonathan_King]

Go ahead and try that MSCONFIG trick; I'm not sure what exactly it does (perhaps some research is in order? ), but it won't hurt, certainly. When done, try playing the game and post a new jcgriff2 report if you get more trouble.

I do know that the fact that the box was unchecked does NOT mean Windows is using only 1 core; Windows will automatically use all cores unless otherwise instructed.

 

[BearCave54]

Alright, so I'm back. Again.
So I tried the msconfig trick and it worked for about a week, then the same thing started happening again.
So I ran through verifier again, and it BS'd to an error of IRQL Not Less or Equal. Now whenever I try to run outside of Safe Mode, it blue screen's to that on a consistent basis.
Here is the jgriff dump for it. And I'm about ready to rip the motherboard and processor out, get the ram and video card out and rebuild. This is absurd.

 

[Jonathan_King]

Well I don't see any more of those dreaded firmware errors since April 20th. I don't know if that's good news or bad, though. I don't see any clues as to what the error is in this report. We're down to just swapping out parts now, I'm afraid. If I had to recommend a single part you purchase, it would be the motherboard, but if you can borrow a PSU, CPU, or RAM, that would also be good.

 

[BearCave54]

Is it possible that MSE could be causing this error and / or issues that I'm experiencing?
I've seen reports from other individuals having this issue and it seems to almost always be a third party software, network adapter, or an issue with antivirus software, is why I am asking.
I've recently replaced the RAM, so I can scratch that from the list. Unfortunately, I don't have any way of borrowing a PSU or CPU from anyone else, so I'd like to narrow it down as much as possible and not end up buying an entire new system to try and get this resolved.

 

[Jonathan_King]

The Verifier_Enabled dump points to dxgkrnl.sys, a Windows system driver. We can pretty much assume the problem is with hardware as a result.

I've never seen MSE cause a crash before, in the time I've been looking at BSODs (January 2010). I'd be very surprised if MSE were to cause a crash now.

 

[BearCave54]

If I'm not mistaken that particular file is a Direct X system file.
I'll go ahead and download the newest Dx redistributable and install it before I look at replacing hardware, see if that clears everything up. It might very well be the problem, as the issues started again after I installed a demo of another game on Steam and after a little over a week of solid stability. I'll post again and keep you updated Mr. King. Thanks again for the assistance.

Edit: I've also found some information on the evga forums stating that there is a known issue with Windows 7 SP1 and eVga Precision running during 3-D application use ( gaming ) that is causing issues with locking up. I'm thinking this might also be an issue, as I always have Precision running when I'm gaming. I'm going to test for a few days without Precision running while I'm doing said gaming on Rift, and see if the situation improves. I'll keep everyone updated.

 

[BearCave54]

I've also found some information on the evga forums stating that there is a known issue with Windows 7 SP1 and eVga Precision running during 3-D application use ( gaming ) that is causing issues with locking up. I'm thinking this might also be an issue, as I always have Precision running when I'm gaming. I'm going to test for a few days without Precision running while I'm doing said gaming on Rift, and see if the situation improves. I'll keep everyone updated.

 

[primcrew]

what games you playing that it crashes on?

 

[Jonathan_King]

Don't bother with the DirectX package; the problem is not actually being caused by DirectX. It certainly won't hurt to give the EVGA Precision removal a try though!

 

[BearCave54]

So the system freezes are happening FAR less frequently now, though they are still occurring and I think I might have found another issue.
This is the RAM I am using: Newegg.com - G.SKILL Ripjaws Series 12GB (3 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) Desktop Memory Model F3-12800CL9T-12GBRL

Now when I go into evga e-leet tuning utility, it has my timings for that RAM set to the following:
CAS Latency: 6 clocks
tRCD: 6 clocks
RAS Precharge: 6 clocks
Cycle Time: 19 clocks
tRFC: 86 clocks
Command Rate: 1T

The specifications for that RAM on the website state that it should be running at a timing of 9.
So my question is, with it running that low, could that be causing system instability as well? And if so, how do I change the timings? I've made it a point for a good long while to never mess with anything like that, due to a lack of comfort with it and knowledge. And the above timings are what it has set the ram for upon installation.

 

[Jonathan_King]

Boot to the BIOS setup, and enter Frequency/Voltage Control > Memory Feature, and change tCL to 9.

http://www.evga.com/support/manuals/files/132-BL-E758.pdf

Worth a shot, certainly.