PDA

View Full Version : G73JH-BST7 BSOD



Shoruk3n
12-02-2011, 08:05 PM
Hey all, new G73 owner here. I just received my recertified G73JH (http://www.newegg.com/Product/Product.aspx?Item=N82E16834230021) yesterday, and after about an hour of playing League of Legends, it's starting to crash on me.

I was in the middle of a game when suddenly greenish blotches appeared on the screen, and then it BSOD'ed. After a failed reboot, I booted into safe mode and ended up uninstalling the video drivers via device manager, and it rebooted. However, the green blotches are still on the screen in both safe mode and in normal mode before reinstalling the video drivers.

When I try to reboot after installing the drivers though, it BSODs again. Right now the GPU is idling at 67C (according to HWinfo64).

What opinions might everyone have as to what the problem is? My gut is telling me I need to RMA because I feel like this is a hardware (GPU) problem.

Here's some info I think may be pertinent:

(I got this from going into the bios)

Bios version G73JH 211
VBios 012.017.000.004.0359.a33g
EC Version B12C180203

And from GPU-Z:

Bios version 012.017.000.004.035992


Not sure why GPU-Z reports a different bios version than the bios...

Any help or input would be greatly appreciated!

Edit: Forgot to mention that I bought the extended 2-year warranty, if it matters.

Edit 2: GPU-Z reports fan speed (%) as 30%, but fan speed RPM is reported as --RPM. This is while idle @ 67C. Not sure if this is important or not.

Shoruk3n
12-03-2011, 01:53 AM
Update: So I managed to update the vbios and video drivers, and now GPU-Z reports idle temp as 52C or so. Is this normal? I'm not sure what to do now.

fostert
12-03-2011, 04:30 AM
That idle (52) is still a bit warm. You say the notebook is recertified? By whom? Did they happen to open it up and clean out the dust beforehand? Did they repaste, or just have the heatsink off for a while then put it back on with no new paste? If they repasted, I'll bet its waaaay too thick (a razor thin coat thats almost transparent is best): I have never seen a good coat of paste on anything pre-pasted by a manufacturer.

Theres lots of unknowns here since a recert could be a warranty serviced notebook with a history, for example.

Firstly get a can of compressed air and blow out the vents. If dust comes out, it could be a sign that you need to open 'er up and clean it.

Shoruk3n
12-03-2011, 05:23 AM
I'm not 100% sure who it is recertified by. Tomorrow I'll try to see if I can find any dust in it, but externally it looks brand new, and the fans are whisper quiet.

All things considered though, I suppose the question I'm ultimately asking myself is should I open it up and repaste, or just RMA because I think I have a faulty unit.

dstrakele
12-03-2011, 05:55 AM
Update: So I managed to update the vbios and video drivers, and now GPU-Z reports idle temp as 52C or so. Is this normal? I'm not sure what to do now.

Did these updates resolve your BSOD issue? If not, look in the "Critical" section of the "Summary of Administrative Events" in Event Viewer for a "BugCheck" event. The reported Bug Check Code and text of this event may help point to some troubleshooting direction. Any other Errors or Warnings in the Event Logs may help as well.

Shoruk3n
12-03-2011, 06:37 AM
No such luck, still getting BSOD. Here are the details for the event viewer:

Level: Critical
Source: Kernal-Power
Event ID: 41
Task Category (63)


I'm not certain if that's what I'm looking for though. I don't see anything that is Critical Bugcheck, the only Bugchecks I see are of the category "Error" with Event ID 1001. Is that what I'm looking for?

dstrakele
12-03-2011, 06:52 AM
Choose the "System" Event Log, then click "Action-Filter Current Log" from the top menu and select the "BugCheck" source in the Filter Dialog. This will filter the System Log to display only BugCheck source events. Report the Bug Check Code and exact text of these events.

4648

Don't know how the damn cat ended up in there...

Shoruk3n
12-03-2011, 07:22 AM
Thanks for the help by the way!

Ok, here we go.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xfffffa8008f8b010, 0xfffff88002e078b8, 0x0000000000000000, 0x0000000000000002). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 120211-28407-01.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xfffffa8008f46010, 0xfffff880016078b8, 0x0000000000000000, 0x0000000000000002). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 120211-26442-01.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xfffffa8008f45010, 0xfffff88003e078b8, 0x0000000000000000, 0x0000000000000002). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 120211-30872-01.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xfffffa80091314e0, 0xfffff88003f538b8, 0x0000000000000000, 0x0000000000000002). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 120211-27674-01.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xfffffa8008b214e0, 0xfffff880042078b8, 0x0000000000000000, 0x0000000000000002). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 120211-31122-01.


Is this the correct info?

fostert
12-03-2011, 04:10 PM
I assume the above bugcheck code is the same as the BSOD Error code 0x116, which is most likely a stability issue with the GPU; see http://forums.tweaktown.com/gigabyte/43058-bsod-error-codes.html

Have you OC'ed the GPU by chance?
Alternatively it could be the memory controller (Northbridge) voltage. Dunno if the G73 has a BIOS that allows you to tweak that voltage upwards a small bit. My G74 has a crippled BIOS so I am screwed if this happens to me!

Shoruk3n
12-03-2011, 05:59 PM
Nope no OCing at all. I updated to catalyst 11.11b last night and got through 1 game of League of Legends. The GPU seems to stay pretty cool (IMO), idleing at 50C and reaching maybe 60ish under load. Just did a quick furmark burnin test and max temp was 79C, but I noticed the fan speed never changed at all, it stayed at a constant 30%. Normal?

So if I suspect the memory controller (Northbridge) is the problem, how would I go about troubleshooting it?

dstrakele
12-03-2011, 08:48 PM
As fostert posts, the Bug Check Code 0x000000116 indicates a GPU problem. You have the infamous Microsoft "Timeout Detection and Recovery" issue. Windows detects the GPU driver stops processing for more than 2 seconds and it attempts to reset the GPU. If reset fails, you get a STOP 0x116 BSOD. In most cases, the reset is successful, the display flashes and a message is displayed that the video driver has recovered.

TDR issues are difficult in that they can have many causes - bad GPU, bad video driver, bad memory, bad power supply, etc. Since your GPU failed to recover, I believe it is more likely a hardware failure.

What is the version of your video driver? You could try upgrading to a more recent version or backrevving to the stock version if you didn't experience the BSOD issues back then. You could test your memory with memtest86. You could try another power adapter if one is available. But ultimately, I think you'll need to RMA to resolve this problem.

Shoruk3n
12-04-2011, 12:56 AM
Current video driver is is 11.11b, from the sticky. I'm fully updated as far as I can tell, but still crashing. I think I will just RMA. Thanks for the help everyone.