Overclock.net › Forums › Intel › Intel CPUs › SYSTEM_THREAD_EXCEPTION_NOT_HANDLED BSOD yet again. DMP analysis included.
New Posts  All Forums:Forum Nav:

SYSTEM_THREAD_EXCEPTION_NOT_HANDLED BSOD yet again. DMP analysis included. - Page 3

post #21 of 31
Alright well, good news and bad news.

I can't get into the stack info from the dmp, which I'm guessing is because a) I don't have an Intel processor or b) I'm missing other files related to the dmp that I'd need to debug it.

From what I did get - I would say it has to do with your video card/s. The only way to tell is to let the BSOD keep happening. For that, do this:

Windows Key + Pause/Break -> Advanced System Properties -> Advanced -> 'Settings' under Startup and Recovery -> Uncheck 'overwrite any existing file'

Then when you BSOD again, PM me, send me the newest dump file. If all of the BSOD's are pointing to the same file causing the problem (this one was caused by Firefox), then we can zero in (hopefully) on what driver is causing it (which is most likely the case).

Best thing to do for now is to check each card by themselves (5850's) and see if they both operate the way they are supposed to. It's possible the second you got and/or setting up CF is causing the problem.
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
post #22 of 31
Sigh...


Why haven't you done the obvious thing yet?

Set the CPU to stock and see if the BSOD goes away?
This will point whether it's being caused by the overclock or not.

What's your idle voltage, anyway?

And 1.52v isn't the maximum safe voltage. I've explained this in several other threads. Max VID is 1.52v and that isn't the same as vcore. Vcore is VID after vdroop is applied, depending on load and current, and this assumes you are NOT using loadline calibration. The loadline chart (vdroop) is right on the Intel PDF documents for the processor.

So assuming a VID of 1.52v, the maximum voltage is going to be lower the higher the load is. Since there is usually over 0.1v vdroop at that range, high load will be between 1.39-1.41v after vdroop is applied.

If you're using loadline calibration, you are going out of Intel's specifications.

Anyway, back to the main topic:
You sort of tried having the tail wag the dog here. Doing 12 hours of prime and memtest was a big waste of time. Set the board to stock first and see if the BSOD goes away. If it does, then you know what the problem is, without wasting 24 hours of your electricity and time. Then, try 4.4 ghz (i believe you said you were at 4.5) at the same voltage and see if the problems reoccur. if they STILL do not reoccur, then you solved the problem MUCH faster.

Just have to think
BTW if you're using C3/C6 disable it.
Also try raising cpu pll voltage to 1.88. (after you try what I told you above, to pinpoint the problem).

@Ganja:
He(she?) could have found out much soooner if it was the video cards by setting the system to stock first.

@OP:
Can you do 1 more thing for me if you don't want to see if it occurs at stock (I STILL recommend you test BOTH at stock and at 4.4 ghz...I do know what I'm talking about; I recommend you only try this after the other things). Try 4.5 ghz and set your PCH voltage at 1.2v. Adding a second video card while overclocked has caused crashes but they have usually been video driver errors/vpu recoveries/game crashes. Several people have fixed that by upping PCH, since that controls the chipset.
Edited by Falkentyne - 5/29/11 at 4:04pm
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
Reply
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
Reply
post #23 of 31
Thread Starter 
Quote:
Originally Posted by GanjaSMK View Post
Alright well, good news and bad news.

I can't get into the stack info from the dmp, which I'm guessing is because a) I don't have an Intel processor or b) I'm missing other files related to the dmp that I'd need to debug it.

From what I did get - I would say it has to do with your video card/s. The only way to tell is to let the BSOD keep happening. For that, do this:

Windows Key + Pause/Break -> Advanced System Properties -> Advanced -> 'Settings' under Startup and Recovery -> Uncheck 'overwrite any existing file'

Then when you BSOD again, PM me, send me the newest dump file. If all of the BSOD's are pointing to the same file causing the problem (this one was caused by Firefox), then we can zero in (hopefully) on what driver is causing it (which is most likely the case).

Best thing to do for now is to check each card by themselves (5850's) and see if they both operate the way they are supposed to. It's possible the second you got and/or setting up CF is causing the problem.
Huge thank you man. Just did everything you said to do. When I BSOD I will be sure to send it to you ASAP.

I will power off my system and remove the card I recently put in, and play around with one card for the time being.

Quote:
Originally Posted by Falkentyne View Post
Sigh...


Why haven't you done the obvious thing yet?

Set the CPU to stock and see if the BSOD goes away?
This will point whether it's being caused by the overclock or not.

What's your idle voltage, anyway?

And 1.52v isn't the maximum safe voltage. I've explained this in several other threads. Max VID is 1.52v and that isn't the same as vcore. Vcore is VID after vdroop is applied, depending on load and current, and this assumes you are NOT using loadline calibration. The loadline chart (vdroop) is right on the Intel PDF documents for the processor.

So assuming a VID of 1.52v, the maximum voltage is going to be lower the higher the load is. Since there is usually over 0.1v vdroop at that range, high load will be between 1.39-1.41v after vdroop is applied.

If you're using loadline calibration, you are going out of Intel's specifications.

Anyway, back to the main topic:
You sort of tried having the tail wag the dog here. Doing 12 hours of prime and memtest was a big waste of time. Set the board to stock first and see if the BSOD goes away. If it does, then you know what the problem is, without wasting 24 hours of your electricity and time. Then, try 4.4 ghz (i believe you said you were at 4.5) at the same voltage and see if the problems reoccur. if they STILL do not reoccur, then you solved the problem MUCH faster.

Just have to think
BTW if you're using C3/C6 disable it.
Also try raising cpu pll voltage to 1.88. (after you try what I told you above, to pinpoint the problem).
Yeah, I'll do that first. As for overclocking to 4.5 here's what I did.

45x multi
1.35v core voltage
and set memory timings and voltages to advertised settings.

That's all the guide I read said to do, didn't touch anything else.

Quote:
Ai Overclock Tuner - Manual
BCLK/PEG Frequency - 100

Turbo Ratio:

By All Cores (Can Adjust In OS)
- 45

Memory Frequency - Set to AUTO or 1333MHz for now - If you are using high speed memory, please still leave this setting low while we find out what the CPU can do.

CAS Latency
- 8
RAS to CAS Delay - 9
RAS Pre Time - 8
RAS ACT Time - 24
DRAM Command Mode - 1T

CPU Voltage - Manual Mode
CPU Manual Voltage - 1.25v -
In the screen shot below, you can see that next to the CPU Voltage, there is a value of 1.224v, this is what was set on the previous restart, and once you save and exit and re-enter the BIOS, your current configuration will be shown, I decided to leave this in so you can see.
DRAM Voltage - 1.5v -
Same as last time!
P95 Blend passed fine and that was that. But then I started BSODing so the first thing that came to mind was vcore, so I bumped it all the way up to 1.35v and I'm still getting BSODs.
Edited by pjBSOD - 5/29/11 at 4:05pm
post #24 of 31
no remove old card and try new one first
Gaming Rig
(13 items)
 
  
CPUMotherboardRAMOS
4790k @ 4.8GHZ z97 sabretooth mark s 290 Win 7 64bit 
KeyboardMouse
Saitek Saitek 
  hide details  
Reply
Gaming Rig
(13 items)
 
  
CPUMotherboardRAMOS
4790k @ 4.8GHZ z97 sabretooth mark s 290 Win 7 64bit 
KeyboardMouse
Saitek Saitek 
  hide details  
Reply
post #25 of 31
Quote:
Originally Posted by E-Peen View Post

Yeah, I'll do that first. As for overclocking to 4.5 here's what I did.

45x multi
1.35v core voltage
and set memory timings and voltages to advertised settings.

That's all the guide I read said to do, didn't touch anything else.

P95 Blend passed fine and that was that. But then I started BSODing so the first thing that came to mind was vcore, so I bumped it all the way up to 1.35v and I'm still getting BSODs.

Hey, from what I have read (since I don't have a SB processor) there is a little bit more to getting the voltages and clocks set correctly. Namely the first thing is not leaving a few things set to Auto, which is what I think you've done besides the VCORE.

I have read around that setting the correct PLL voltage of some sort along with another voltage (VCORE?) can help you get things stable. Check around bud, see if you can get some SB help from the Intel pros.
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
post #26 of 31
Thread Starter 
I just don't get how I can pass 12 hours of P95 Blend, and then randomly BSOD. I'm new to Intel and obviously new to SB so this all just makes no sense to me.

Anyways, I checked the ultimate SB OC guide on here, and it says 0x3B BSOD code is add more vcore.... But I have 1.35 VCC already.. ._. I guess I'll raise it even more.

Quote:
Originally Posted by Falkentyne View Post

@OP:
Can you do 1 more thing for me if you don't want to see if it occurs at stock (I STILL recommend you test BOTH at stock and at 4.4 ghz...I do know what I'm talking about; I recommend you only try this after the other things). Try 4.5 ghz and set your PCH voltage at 1.2v. Adding a second video card while overclocked has caused crashes but they have usually been video driver errors/vpu recoveries/game crashes. Several people have fixed that by upping PCH, since that controls the chipset.
I know it'll alleviate all of my questions if I just set my cpu to stock, but I want to see if bumping the VCC to 1.55v and the PCH voltage to 1.25v will fix it before I throw everything back to stock.

Edit: Okay, I just raised the PCH voltage to 1.25v, and my VCC was only 1.325v, so I bumped it to 1.375v.
Edited by pjBSOD - 5/29/11 at 4:51pm
post #27 of 31
What is your idle voltage? (the voltage when you aren't doing anything)

You're not using offset voltages, so are C3/C6 state enabled or disabled? (They should be disabled).
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
Reply
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
Reply
post #28 of 31
Thread Starter 
Idle VCC fluctuates between 1.368v and 1.376v.

C3/C6 are enabled. What's C3/C6?
post #29 of 31
C3/C6 are completely useless. They're used with C1E/EIST, and will park some idle cores into a deep sleep (voltage cut, clock cycles turned off)...it might save you at most 5W while idle (like 14W to 10W) but it tends to cause problems because the vcore given at 1 thread/light load will sometimes be too low for stability, if the CPU is ramped up to full speed when it happens (example: 5 ghz with offsets C3/C6 disabled: 1.524v 1 thread/light load, 1.512v medium (average) load, 1.50v heavy/max load= no problem. C3/C6 enabled: coming from idle to 1 thread: 1.48v 1 thread--> BSOD (almost always 0x124) possible if you already know you need at least 1.50v for full stability.
Likewise, the data has to be cached into L3 cache when using C3/C6, so it can be retrieved later.
All of this can cause strange anomolies to happen. (if you're using bit higher vcore than you need, you may not notice anything), butI did notice that when I used C3/C6 with offsets on superPi, the cores took too long to ramp up, totally butchering my score on the first run. Disabling C3/C6=acted like it should. Plus I had a BSOD while typing idle in a game lobby with it enabled....never happened after I disabled it.

Using C3/C6 saves you maybe 10 cents a month on electricity. The critical settings if you want to downclock and downvolt while idle are C1E (hardware based, overrides all OS settings) and EIST (speedstep--relies purely on the OS power plan settings to work).
Edited by Falkentyne - 5/29/11 at 6:54pm
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
Reply
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
Reply
post #30 of 31
Thread Starter 
So in easiest terms, I should go into the BIOS and disable both C3/C6? I also read they DO cause some issues.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › SYSTEM_THREAD_EXCEPTION_NOT_HANDLED BSOD yet again. DMP analysis included.