[SOLVED] Unrepeatable BSOD, fixed few problems I think, more inside - Windows 8.1 x64

Alright. As far as I can remember the only time I had this card in this machine was when it was in with the old faulty motherboard. I got a crash while either playing cod world at war or at desktop while my internet was completely uninstalled and turned off as I thought my ethernet card was the culprit. So it was hard to determine that there was another problem in addition to the faulty motherboard at the time.

This card was in my old computer for 2 and a half years with no issues at all guaranteed.

This has been the longest period with no crashes on this system. Though it worries me that a problem occurred with this card (560 TI) because of the above 2.5 years no issues with it.

Edit: It comes to my attention that this exact scenario has occurred before. Exiting VLC or fullscreen during VLC causing bluescreens that is. Not while just watching the show but always triggered with closing fullscreen or completely. To clarify this has happened with all 3 cards, the 7950, 270x, and 560 TI.

Since you told me to change to DX9 from DX11 in WoW, I haven't had a crash in it. Neither have I in league of legends, where I think i've played enough games to warrant a reasonable sample size in comparison to the 7950 and 270x crashes
 
Do you have another motherboard to install that video card in that you'd use for a 'decent' period of time to test whether or not it crashes with VLC?

Edit: It comes to my attention that this exact scenario has occurred before. Exiting VLC or fullscreen during VLC causing bluescreens that is. Not while just watching the show but always triggered with closing fullscreen or completely. To clarify this has happened with all 3 cards, the 7950, 270x, and 560 TI.
Click to expand...

Well, as far as I know, there's no full screen bug with the latest VLC. I use it every day and I don't have any issues on a multi-monitor setup.

Regards,

Patrick
 
Last edited:
Actually my girlfriend had been using the 560 TI in my old computer the entire time it hadn't been in my current one, and has had no issues whatsoever with ANY of the cards. So while I've been rotating them out of the system I currently use, they've still been in use by her with no issues.
 
Okay, so here's the rundown then:

1. You have a faulty motherboard (yes, again). It's not rare for a manufacturer to give you another broken one or the same one back but 'repaired'.

2. I suspect a possible power supply problem. Have you at any point in our troubleshooting tried a different one?

3. Really really bad video drivers. Have you tried beta, latest, few versions behind the latest, etc?

Regards,

Patrick
 
Tried all variations of drivers with all cards.

Motherboard was a different one, brand new, unused as far as i know. I checked with the store they shipped it from before they actually did ship it to confirm it was not refurbished.

Have not tried anything regarding the power supply
 
Alright, we're narrowed it down then.

Any way for you to get a PSU of the same (or close to) power of yours? If not, so long as yours is under warranty they may just accept an RMA if you mention you've exhausted all options and suspect the PSU is a culprit. It depends on the manufacturer, but if it's a Corsair PSU, I expect no problems.

Side note while we're here, there's a BIOS update for your board:

MSI USA - Computer, Laptop, Notebook, Desktop, Mainboard, Graphics and more

V 1.9 -- You're at 1.8:

Code: 
- Enhanced EIST function when use non turbo boost CPU.
- Improved USB device compatibility.
- Improved S3 function.
- Improved PS2 device compatibility when "MSI Fast Boot" is enabled.
- Improved compatibility.

It's worth a shot.

-- Be sure you clear the CMOS (remove battery or use jumper) before and after updating the BIOS. It should go like so:

1 - Clear CMOS.

2 - Update BIOS.

3 - Clear CMOS.

4 - Boot to Windows.

Regards,

Patrick
 
Will do. That's for the rapid response.

I'll maybe try and borrow that psu, rma mine regardless and see how that goes? Can't hurt that much I guess.

Edit: I don't mean to scare you when I ask this, but my board has a clear cmos button apparently at the back IO panel. Do I just turn of computer and press it?
 
I'll maybe try and borrow that psu, rma mine regardless and see how that goes? Can't hurt that much I guess.
Click to expand...

Sounds like a good plan, yeah. FWIW, most manufacturers really don't give you a hard time if you politely explain why you're RMA'ing the product. I mean, you have a good reason in your case.

my board has a clear cmos button apparently at the back IO panel. Do I just turn of computer and press it?
Click to expand...

Most high-end boards do, I wish mine did! You're right on turning it off, but we need to go a bit further, and I should have elaborated earlier. In addition to turning off the system, you need to unplug the power cord from the PSU as well to drain it. Let it sit for about a minute or so, then clear the CMOS. After ~10 seconds, plug the power cord back. Do this a second time as I mentioned after the BIOS is updated.

Regards,

Patrick
 
Alright BIOS is updated.

I'm going over to friends house on friday I learned, so I'll ask if I can borrow his psu while I rma mine.

Thanks again for the responses. I'll update when there's something to report on.
 
Hey Patrick,

After updating the BIOS, I couldn't get a hold of my friend's PSU so I haven't been able to test anything regarding my PSU.

I got a brand new error that I've not seen before. ATTEMPTED_EXECUTE_OF_NOEXECUTE_MEMORY, which leads me to believe it's the RAM or a driver that's misusing the RAM. However because I just did another quick memory test, and our previous 2 full memory tests, as well as it's functionality in my old system, I don't believe it to be faulty.

If you could take a look at it and see if it gives anything new and miraculously definite that would be greatly appreciated.

I haven't had time to look into RMAing my PSU, but I'm feeling I should asap to try and further along the problem solving process. I'll try and do that.

Thanks,

Xivinas

Edit: I was just looking at the dump file myself, and was typing in here while I got the same error again with no strenuous work on the GPU, even though it points to the GPU driver. The first crash was during watching a show on VLC player. Doesn't seem the computer has to be under any load. Jesus Christ.
 

Attachments

This just gets more and more interesting, doesn't it?

ATTEMPTED_EXECUTE_OF_NOEXECUTE_MEMORY (FC)

This indicates that an attempt was made to execute non-executable memory.

Code: 
0: kd> k
Child-SP          RetAddr           Call Site
fffff801`7ec85ad8 fffff801`7d1c0020 nt!KeBugCheckEx
fffff801`7ec85ae0 fffff801`7d1c7fe8 nt! ?? ::FNODOBFM::`string'+0x46e70
fffff801`7ec85b20 fffff801`7d065e5c nt! ?? ::FNODOBFM::`string'+0x4ee38
fffff801`7ec85b60 fffff801`7d172f2f nt!MmAccessFault+0xd9c
fffff801`7ec85ca0 ffffe001`281af000 nt!KiPageFault+0x12f [COLOR=#800080]// Hit a pagefault as the nVidia video driver tried to execute non-executable memory.[/COLOR]
fffff801`7ec85e38 fffff801`4ef68a6a 0xffffe001`281af000 [COLOR=#800080]// Non-executable memory.[/COLOR]
fffff801`7ec85e40 fffff801`4ef68ada nvlddmkm+0xf9a6a [COLOR=#800080]// nVidia video driver.[/COLOR]
fffff801`7ec85e48 fffff801`7d058613 nvlddmkm+0xf9ada [COLOR=#800080]// nVidia video driver.[/COLOR]
fffff801`7ec85e50 00000000`00000000 nt!EtwpLogKernelEvent+0x273 [COLOR=#800080]// It looks like something was using Event Tracing to log a kernel event. We won't be able to tell what it was without a kernel dump. Maybe nVidia?[/COLOR]

Code: 
BugCheck FC, {[COLOR=#ff0000]ffffe001281af000[/COLOR], 80000001000009e3, fffff8017ec85ca0, 3}

Code: 
0: kd> !pte ffffe001281af000
                                           VA ffffe001281af000
PXE at FFFFF6FB7DBEDE00    PPE at FFFFF6FB7DBC0020    [COLOR=#ff0000]PDE at FFFFF6FB78004A00[/COLOR]    PTE at FFFFF6F000940D78
contains 0000000000EE8863  contains 0000000000EE7863  contains 80000001000009E3  contains 0000000000000000
GetUlongFromAddress: unable to read from fffff8017d369104
pfn ee8       [COLOR=#0000cd]---DA--KWEV[/COLOR]  pfn ee7       [COLOR=#0000cd]---DA--KWEV[/COLOR]  pfn 100000    [COLOR=#006400]-GLDA--KWEV[/COLOR]  [COLOR=#ff8c00]LARGE PAGE[/COLOR] pfn 1001af

Here's the memory that the nVidia driver tried to execute when it hit the pagefault, we can see all the different kinds of status/protection bits.

D - Dirty, as in the page was previously written to.

A - Accessed, as in the page (or table) has been previously accessed (read).

K - This page is owned by kernel mode, not user mode.

W - Writable, as in the page is able to be written to (not just read).

E - Executable, as in the page is executable.

V - Valid, as in the page is located in physical memory.

G - Global, as in the TLB won't be flushed upon a context switch.

L - Large, as in this tells us that the page is a PDE and indicates whether or not the PDE is in the last table level.

Since we know we're dealing with a PDE:

Code: 
0: kd> dt nt!_MMPTE u.Hard
   +0x000 u      : 
      +0x000 Hard   : [COLOR=#ff0000]_MMPTE_HARDWARE[/COLOR]

Code: 
0: kd> dt _MMPTE_HARDWARE FFFFF6FB78004A00
nt!_MMPTE_HARDWARE
   +0x000 Valid            : 0y1
   +0x000 Dirty1           : 0y1
   +0x000 Owner            : 0y0
   +0x000 WriteThrough     : 0y0
   +0x000 CacheDisable     : 0y0
   +0x000 Accessed         : 0y1
   +0x000 Dirty            : 0y1
   +0x000 LargePage        : 0y1
   +0x000 Global           : 0y1
   +0x000 CopyOnWrite      : 0y0
   +0x000 Unused           : 0y0
   +0x000 Write            : 0y1
   +0x000 PageFrameNumber  : 0y000000000000000100000000000000000000 ([COLOR=#ff0000]0x100000[/COLOR])
   +0x000 reserved1        : 0y0000
   +0x000 SoftwareWsIndex  : 0y00000000000 (0)
   +0x000 NoExecute        : 0y1

Without a kernel dump it's pretty hard to say what actually went wrong here, but an nVidia driver issue is most certainly still a possibility given this bug check. I'm pretty sure we've tried just about every nVidia driver there is though, no?

Do me a favor and disable the nVidia Streaming Service in services.msc if you haven't already, and then restart the system.

Regards,

Patrick
 
Yes that is correct. As well as the card had just been running in my girlfriend's system (my old one) with the same driver.
 
Yeah, it's really hard to say. If it's working in your GF's system, as I've noted before it's either a hardware or really annoying nVidia driver issue. I still think we're dealing with a PSU issue, however it's interesting to see this bug check. If it is the PSU, it may be providing insufficient power to the GPU when it was doing page stuff, so it just seemed like a possible driver issue. Again, it's hard to say.

I edited my post to say this in case you didn't see:

Do me a favor and disable the nVidia Streaming Service in services.msc if you haven't already, and then restart the system.
Click to expand...

Regards,

Patrick
 
I've submitted an RMA request form just now. Thanks for pointing me to your edit, I didn't see it. I'll do that now.

Thanks so much for your continuous support.

Edit: My friend has a similar issue he suspects and mentioned the PSU may not be supplying enough power to the GPU during certain times. The error happens immediately after I click on something. For example what prompted the first error was opening a picture on reddit, because immediately after I clicked everything started buzzing and boom, error.
 
The pleasure is mine : )

Let me know if the RMA is accepted, and also if there's any change from disable the service.

Regards,

Patrick
 
Update:

7 days without crash, not like it's a record or anything though since the frequency of the error seems to be ~9 days. I changed the nvidia streaming service 7 days ago like you said.

I just got the replacement PSU today and it's currently installed as of this post. Will update later as well. Fingers crossed ><

Cheers,

Xivinas
 
Update:

LOL Corsair just gave me a replacement PSU with a broken fan. So my old PSU is now back in the system and I have to return both now.
 
Last edited by a moderator:

Has Sysnative Forums helped you? Please consider donating to help us support the site!

Back
Top