BSOD during video rendering.

[EarthDog]

It does not mean revision 3, no. On the board, it will say "rev X".

 

[jproductions]

In the corner it says rev. 1.0, so I am guessing its the original one. Its an old board had to flash the bios to get it to recognize this chip.

 

[Johan45]

That should be fine I was just checking

 

[jproductions]

Took memtest86 about 4 hours to complete one pass, but it came back with no errors. From what I was Googling it has to be some sort of boot issue, but I got several different ways of fixing it.

I didn't mention this before, but it took me 3 days and 5 tries of installing windows in order for it to properly work on these new drives in RAID0. Every single time BOOTMGR was missing, never knew exactly what the problem was because the 5th try just worked fine.

 

[ShrimpBrime]

Can you elaborate on that a bit? I've stuffed drives WAY fuller than 15% free space available without error...

Sure!

A read on MFT http://msdn.microsoft.com/en-us/library/windows/desktop/aa365230(v=vs.85).aspx

I believe errors are most common like this over USB.

Bad drive configuration. Or maybe a drive is going bad. Cables may need a wiggle.

I agree on the high temps as well. From my experience, anything over 50c with an overclock will eventually error.

I didn't mention this before, but it took me 3 days and 5 tries of installing windows in order for it to properly work on these new drives in RAID0

So you eventually found the right driver and is set up in AHCI mode correct?

 

[jproductions]

Sure!
I agree on the high temps as well. From my experience, anything over 50c with an overclock will eventually error.

But the last 3 BSODs my temps were around 25-35max. I dont know, I am not expert, I have been looking into watercooling, just is going to be expensive because I am going to need a new case, plus the smaller block shaped coolers don't cool as well as my current setup (according to other forums and reviews with my same CPU with OC)

So you eventually found the right driver and is set up in AHCI mode correct?

No, RAID. It will not boot in AHCI. I know everyone told me to use AHCI, but again, doesnt boot and my two IT friends dont know why. So I left it in RAID.
Edit: yes I had the drivers all along, windows wouldn't recognize them on 4 different usb flash drives.

 

[RJARRRPCGP]

The error was 0xA.

That error code is typical AMD core instability and CPU instability may also cause you to get an error message about being unable to initialize or report an "internal error" and etc.

(Also, Windows 95 and Windows 98 will report an "illegal operation", too.)

It can happen any time when the CPU is real unstable.

And Prime95 requires settings to be changed:

The defaults aren't sufficient!

You must enable round off checking on one of the menus, by checking it and the same with sum checking.
Likely why Prime95 reported nothing!

Late versions of Prime95 stubbornly have those disabled by default.

 

[jproductions]

That error code is typical AMD core instability and CPU instability may also cause you to get an error message about being unable to initialize or report an "internal error" and etc.

(Also, Windows 95 and Windows 98 will report an "illegal operation", too.)

It can happen any time when the CPU is real unstable.

And Prime95 requires settings to be changed:

The defaults aren't sufficient!

You must enable round off checking on one of the menus, by checking it and the same with sum checking.
Likely why Prime95 reported nothing!

Late versions of Prime95 stubbornly have those disabled by default.

Well, prime does report errors when I dont have proper voltage. But I will do what you said and check those settings and run it again.

 

[ShrimpBrime]

Well, prime does report errors when I dont have proper voltage. But I will do what you said and check those settings and run it again.

Ahh there it is....Cpu errors. Unstable.

 

[jproductions]

Ahh there it is....Cpu errors. Unstable.

Like I said, only when I have the wrong voltage set.

While running Prime95:
Now that I have said that, lets totally contradict myself and say that I did get a fatal error, I didn't give me an error message, but I think it said expected less than 0.4, received 0.5. Worker stopped, on worker #5. This was at 4.4Ghz, and I don't know what that means, but I think it could have to do with heat, my room is hotter than normal right now and my temps got up to 65C. (this was also after I set the SUM error checking and Round off checking)

I went down to 4.2Ghz, temps maxed out at 56C, still running now going on an hour with no errors. I will update if anything changes.

Still don't believe this has anything to do with ntoskrnl.exe error.

 

[ShrimpBrime]

Like I said, only when I have the wrong voltage set.

While running Prime95:
Now that I have said that, lets totally contradict myself and say that I did get a fatal error, I didn't give me an error message, but I think it said expected less than 0.4, received 0.5. Worker stopped, on worker #5. This was at 4.4Ghz, and I don't know what that means, but I think it could have to do with heat, my room is hotter than normal right now and my temps got up to 65C. (this was also after I set the SUM error checking and Round off checking)

I went down to 4.2Ghz, temps maxed out at 56C, still running now going on an hour with no errors. I will update if anything changes.

Still don't believe this has anything to do with ntoskrnl.exe error.

65c is too hot.

ntoskrnl.exe Or IRQ not less or equal.

Under volt / over heat error. Very common AMD OC BS.

Without stable clocks (default/stock) you can't diagnose between HW and SW.

I'd clear CMOS and re-evaluate the OC.

 

[jproductions]

Well at 4.2 I got no errors after 1 hour with a max temp of 56c. I killed it so I can play a game, 10 minutes in my computer just automatically restarts(not sure temp I wasn't looking). No BSOD and not to mention I played the same game yesterday for about an hour at 4.4ghz with no issue. I don't know if it was a fluke or what.

I am getting over this, and I currently don't have any video to render for a while. I am going to put everything back to stock, if I still get BSODs then there is no way its temp or voltage. If I don't, then it has to be one or the other.

 

[redduc900]

Download BlueScreenView v1.47 to find out which driver is causing the errors...

http://www.nirsoft.net/utils/blue_screen_view.html

... and run a chkdsk /f on the root volume.

I am getting the ntoskrnl.exe error.

I ran disk check

About to burn the memtest86+ and see if that gives me any errors.

That just tells us a crash took place in the Windows kernel. What driver is shown in BSV as causing the error? And what were the results of running a chkdsk /f on the root volume?

 

[jproductions]

That just tells us a crash took place in the Windows kernel. What driver is shown in BSV as causing the error? And what were the results of running a chkdsk /f on the root volume?

It said disk is in use or something similar. Perform check after restart [y][n] so I rebooted saw the disc check start walked out of the room and when I came back it was already booted into windows. So I assumed no errors.

Blue screen viewer doesn't really give me any further information. Bug check code: 0x0000000a, caused by driver: ntoskrnl.exe, caused by address: ntoskrnl.exe+75c00. Other than that there is no real valuable information. All I can tell you when I does it I am transferring large files. One from a USB3 WD external, and 2 ~200mb files are being uploaded using FileZilla via the local drive, while I am watching YouTube or a couple times while rendering in Premiere. Once while I was watching Vimeo, but wasn't transferring any files. Just a random crash, same error.

Not much more I can tell you. (Sorry for any misspelling I am on my phone)

 

[redduc900]

Run it again, but this time watch the output while it's running the disk check.

 

[RJARRRPCGP]

Turn down the core clock by 10 Mhz. I would drop the multi.

 

[jproductions]

Sorry I was out of town, I just ran disk check and got no errors, no corrupt files or what ever it said.

I am going to leave it at the stock 3.6ghz and try to duplicate the errors to rule out the overclock, but wont be able to do that until tomorrow.

Thanks for the help everyone.

 

[jproductions]

So after a month of no real issues, well blue screen related, I finally got another blue screen. All my issues I have been trying to sort out the past month are image corruption issues in Lr4, Lr5 seems to have solved this problem, but while exporting images using Lr5 I got another blue screen. I was exporting images from my internal SSD to the a different folder in the same SSD.
Error:
Ox1A MEMORY_MANAGEMENT ntoskrnl.exe

Currently, default clock, default everything. in the past month I have run memtest86+ with no errors, checked all my SMART attributes on my SSDs and according to Mushkin support: no issues.

Suggestions on this one?