Event 41 Kernel-Power, Bugcheck 0 Hard Hang

[lekro]

I know that there's lots of these threads about Event 41 from Kernel-Power, but I've looked everywhere and nothing seems to work.
I have Windows 7 64-bit.

Here's my system (it's unbalanced, don't tell me about it)

Intel Core i7-3930k at default clock speeds.
Kingston HyperX Predator DDR3 8GB at 1333 MHz (timings ARE set correctly)
Intel DX79SI motherboard
MSI Twin Frozr III GTX 660 2GB at 1.03 GHz
WD Caviar Black 1TB 7200RPM SATA III
Lite-On 24x DVD-RW drive (I don't think this matters that much)
Corsair TX650 PSU (650 watt)
Cooler Master Hyper 620


I have disabled the CPU C State in the BIOS, and that seemed to increase the stability by a little bit, but it's still not good enough (Random hard hang while loading a level in a game, or turning on Premiere Pro, etc. still happens)

These freezes are very unpredictable, and if I try going to ... say, Youtube, right after I reset the computer from one, a hard hang is imminent. On the other hand, I can keep my computer running for hours and hours doing something (like Prime95 and FurMark) and it will still be running smoothly.

I have checked my CPU/GPU temps and they are fine.
CPU is 35 C at idle and 60 C under max load. (Prime95)
GPU is 33 C at idle and 64 C under max load. (FurMark)

Every time a hard hang occurs:
- All sound turns off
- Computer doesn't do anything without a hard power off or a push of the reset button
- HDD light on case remains either on or off
- Computer will accept no input.
- If I'm playing a game in which the Windows mouse pointer is hidden, it appears on top of the crosshair.

There is no minidump or BSOD, and the bugcheck code in the Event log is 0.

Here's my event log data (all that I have):

- <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
- <System>
  <Provider Name="Microsoft-Windows-Kernel-Power" Guid="{331C3B3A-2005-44C2-AC5E-77220C37D6B4}" /> 
  <EventID>41</EventID> 
  <Version>2</Version> 
  <Level>1</Level> 
  <Task>63</Task> 
  <Opcode>0</Opcode> 
  <Keywords>0x8000000000000002</Keywords> 
  <TimeCreated SystemTime="2013-01-18T06:13:29.075220400Z" /> 
  <EventRecordID>13512</EventRecordID> 
  <Correlation /> 
  <Execution ProcessID="4" ThreadID="8" /> 
  <Channel>System</Channel> 
  <Computer>Kratoa</Computer> 
  <Security UserID="S-1-5-18" /> 
  </System>
- <EventData>
  <Data Name="BugcheckCode">0</Data> 
  <Data Name="BugcheckParameter1">0x0</Data> 
  <Data Name="BugcheckParameter2">0x0</Data> 
  <Data Name="BugcheckParameter3">0x0</Data> 
  <Data Name="BugcheckParameter4">0x0</Data> 
  <Data Name="SleepInProgress">false</Data> 
  <Data Name="PowerButtonTimestamp">0</Data> 
  </EventData>
  </Event>

I have a Ralink Wifi N adapter included with my motherboard (and I thought it was the problem) but I replaced it with my old Atheros A/B/G one and I still had the problem.

One strange thing about this is it tends to happen more when I use the internet. I could play SHOGUN 2 for a long time while Steam couldn't connect to its servers, but when it did, I had a hard hang within 5 minutes.
Another thing I tend to do is listening to music on Youtube and playing Minecraft Multiplayer at the same time. This used to easily hard hang my system, but after disabling CPU C State it occurs much less frequently, although it still occurs.

I HAVE seen the Microsoft KB article, but it was no help as I did everything listed and nothing improved.

I have seen some posts saying that audio drivers (or any drivers) may be a problem, but I've updated them all and still I have the problem. Others said to disable extra audio adapters, but I use all three of mine (NVIDIA HDMI audio, Analog out (Headphones, Speakers, etc), and my USB headset. I've tried this, but the problem persists.

I've also done memory testing, removed each of the modules, leaving the other in, but both times I could reproduce the problem.

Is there anything I've missed? (Tell me of anything I might have missed)

And there I am. No solution found anywhere, although this thread seemed to be useful, but didn't really solve my problem.
Anyone? Any suggestions of what I should do? (or more information you need?)


EDIT: Aha! I think I found something that might give some information actually. I went snooping around Event Viewer for a while and came up with this: (It happened at about the same time as the 41)

- <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
- <System>
  <Provider Name="Microsoft-Windows-Kernel-Power" Guid="{331C3B3A-2005-44C2-AC5E-77220C37D6B4}" /> 
  <EventID>91</EventID> 
  <Version>0</Version> 
  <Level>3</Level> 
  <Task>88</Task> 
  <Opcode>0</Opcode> 
  <Keywords>0x1000000000000020</Keywords> 
  <TimeCreated SystemTime="2013-01-15T04:32:18.720544600Z" /> 
  <EventRecordID>27</EventRecordID> 
  <Correlation /> 
  <Execution ProcessID="1280" ThreadID="3116" /> 
  <Channel>Microsoft-Windows-Kernel-Power/Thermal-Operational</Channel> 
  <Computer>Kratoa</Computer> 
  <Security /> 
  </System>
- <EventData>
  <Data Name="ProcessorId">0</Data> 
  <Data Name="ThrottleMSR">0x0</Data> 
  <Data Name="ElapsedTime">0</Data> 
  <Data Name="LogInterval">1000</Data> 
  </EventData>
  </Event>

This happens for all cores (i.e. 0, 1, 2, 3, 4, 5)
I'm not very good at deciphering this, so I don't know what to do from here. I read somewhere that I might have to bump VCCIO a bit, but I'm hesitant and don't want to melt the chip.

EDIT 2: I have disabled Intel SpeedStep,and will see if I get any changes.

 

[redduc900]

What you read as far as possibly needing to bump VCCIO voltage for stability was correct. Default VCCIO voltage (Processor I/O) is 1.050V, so start by bumping it in increments to no greater than 1.15V, checking for stability after each incremental increase w/ OCCT, LinX. Prime, etc.