Computer crashes with different errors every time, and no minidump -- help!

ohjustwonderfuuu · Jun 27, 2016

Summer of 2015, end of May/early June, I start crashing out of nowhere. Here's a screencap of my BlueScreenView. After trying reinstalling various drivers, I got the BSODs to stop out of nowhere one day. I'm concerned, but glad.
Fast forward to summer of 2016 and my computer starts crashing randomly once again, once every day or two, making me think this is an overheating problem due to dust buildup within the computer. I can't lookup the BSODs via BlueScreenView's convenient google search due to the fact that, despite increasing the pagefile size and confirming the settings, the minidump is never written, or at least if it is, it never saves. And then soon after, my once smoothly playable Overwatch and League games slowed down to ~25 FPS from 45 in OW, and to ~35 FPS from ~100+ in LoL. So I dust out the computer, but nothing changes, and in fact the BSODs begin to come more often.

Currently, I'm at a point where I can barely go an hour of use without getting either a BSOD or my screen freezing, with a mix of random colors on the screen. These crashes occur during anything from YouTube to simply browsing forums for help. I haven't played Overwatch or League in a bit of time because of these issues, but I've crashed many times before playing them. I've run the Windows Memory Diagnostic Tool but it says that there are no errors with my memory.

I'm at my wit's end and I don't even know what to do with the myriad problems I'm facing. I appreciate any and all help, and if there is any relevant information that I need to add to diagnose my computer, please tell me.

BSODs I've faced recently:
0x3B SYSTEM_SERVICE_EXCEPTION (sometimes attributed to atikmdag.sys, sometimes not)
0x50 PAGE_FAULT_IN_NONPAGED_AREA (Atihdw76.sys or win32k.sys or atikmdag.sys)
0x1A MEMORY_MANAGEMENT
0x7F
0x1E (win32k.sys)

Specs:
XPS 8300 from ~2011. No parts replaced.
Video Card: AMD Radeon HD 6670 (1920 x 1080 (60Hz))
Hard Drive: 1TB WDC WD10EALX-759BA1
RAM: 8GB DDR3
CPU: Intel Core i5 2310 @ 2.9GHz
Motherboard: Dell 0Y2MRG

Aragorn · Jun 27, 2016

Sounds like it could be the PSU different components getting the shunt each time (or different processes).

LukeFatwalker · Jun 28, 2016

I'd recommend that you do a re-formatting of the machine to see whether it's hardware or not. If you re-format it should be able to resolve the issue fairly quickly if it's software related. Just be sure to back up your data first with a disk imager, and if that works, start taking regular backups either with a disk imager or a snapshot tool like Rollback Rx.

If it still causes issues, it's hardware issue and is likely tied to CPU or PSU.

ohjustwonderfuuu · Jun 28, 2016

Ah, thanks for the advice all. Just before I do any reformatting/ordering a new PSU, is there any other test or possible solution that I can try?

As an update, I went 24 hours without a crash, but just now I got a 0x7F error while I was away from the computer.

LukeFatwalker · Jun 29, 2016

ohjustwonderfuuu :

You could always try re-installing the drivers. Make sure you run a backup before and after.

ohjustwonderfuuu · Jun 29, 2016

I've tried uninstalling and re-installing the drivers for my AMD Radeon HD 6670 quite a few times now. Will try again though.

Which other possible problem drivers should I be looking to reinstall?

johnbl · Jun 29, 2016

- generally your system will produce a minidump unless it is having problems with the sata subsystem.
(update the SATA driver, move the drive to another port or sata controller)
- you might consider google "how to force a memory dump using a keyboard" make the registry settings and force a memory dump while the system is working correctly. Often the memory dump will show what the problem is even before the system crashes.

- if you see error coded in parameter 1 of c0000005 you will want to remove any overclocking of the GPU, CPU and bios overclocking. Then update the BIOS and test the system memory with memtest86

- several of the bugchecks were in networking code, you would want to update the network driver. If you have a usb wireless ethernet card, you would also need to update the BIOS, CPU chipset drivers and any USB 3.0 chipset driver from your motherboard vendor.

- a bugcheck in your video cards audio driver can mean a conflict with the onboard sound driver, Be sure to update the motherboard sound driver from the motherboard vendors website.

your image, you need to expand the parameter 1 column. It contains the error code of each bugcheck. right now it is truncated. (last two characters not shown in the error code)

without the error code, I can not tell you what kind of memory the memory manager is having a problem with, (physical ram, cache management, or virtual memory (pagefile.sys)) each will have different potential causes and solutions.

ohjustwonderfuuu · Jun 30, 2016

Ah, thank you for the detailed response! Here's a better picture of BlueScreenView: http://puu.sh/pL4Uz/e644fa944b.png

Will try updating the drivers listed above. I use an ethernet cable so I'm hoping I don't have to update the BIOS?

johnbl · Jun 30, 2016

generally, a bugcheck attributed to a wired ethernet driver will not require a BIOS update.

(USB wirless might because of the BIOS support of the USB subsystem)

BIOS updates are pretty painless now, you may want to do one. It will reset your memory timings back to default.
(BIOS updated often have updated memory timings also)

You can also put your memory dump files from c:\windows\minidump directory on a cloud server like microsoft onedrive, share the files as public and I can take a quick look to see if you have any suspect driver versions installed.

make sure you run the memtest86, it will test your memory timings without having windows running.
(if it fails you will have to update the BIOS and re run the tests, which is why i just have people update the bios before running the test)

looks like you had 2 bugchecks related to video
2 related to network,
7 related to bad memory addresses used
1 or 2 coming out of the file system

bad memory timings could cause them all.
(not sure about the network bugchecks, I would have to look at the data in the memory dump)

if the memory test come out ok, just make sure you update all of the motherboard drivers.
It is pretty common for the motherboard audio driver to conflict with the graphics drivers audio support and cause the GPU drivers to bugcheck. Also, failing power to a graphics card can cause voltage fluctuations that could cause memory corruptions. (same as overheated components (check for fans that are slow or stopped working))

ohjustwonderfuuu :

ohjustwonderfuuu · Jun 30, 2016

Oh man, was going to update drivers but I can't seem to be able to operate the computer long enough before either a BSOD, or a crash with random colors popping up. Just recently I had a crash with random colors that led to a BSOD with no text. And a few times, my computer crashed, and then the power light started glowing Amber while making intervals of 4 beeps which I believe signifies RAM errors.

I'm currently looking for an external drive to make a system image on (have ~750GB on hard drive so gonna need a 1TB hard drive), however I'm not sure if I'll be able to have enough time to operate the computer before it crashes.

So, what should my next steps be? Is trying to frantically work with this computer until it crashes just further degrading the parts inside?

johnbl · Jun 30, 2016

Image result for dell motherboard beep codes
Dell beep codes
Beep Code Description
1 beep BIOS ROM corruption or failure
2 beeps Memory (RAM) not detected
3 beeps Motherboard failure
4 beeps Memory (RAM) failure

I would reset each RAM, reset the BIOS back to defaults and boot and run memtest86 to confim the RAM is working as expected (no errors)

ohjustwonderfuuu · Jul 1, 2016

I went out today, so I left Memtest on for 16 passes with no errors. Picture from my phone here: http://imgur.com/2DWtLkK

I'm glad, but also confused at how I could've done Memtest for so many passes but found no errors, yet gotten the 4 beeps signifying RAM failure.

Guess I'm back to trying to update drivers, will update if I figure anything out.

ohjustwonderfuuu · Jul 2, 2016

So my computer crashes even just on booting normally and waiting on the home screen, and still gives me the 4 beeps signifying RAM error even though 16 runs of Memtest86 have given me no errors.

johnbl · Jul 2, 2016

I would work on getting a memory dump. make sure your settings are correct, pagefile.sys is large enough
then google "how to force a windows memory dump with a keyboard" make the registry changes and try to force a memory dump while the system is working correctl.

you might also put your drive on a different SATA controller and port.

ohjustwonderfuuu · Jul 3, 2016

I can now force a blue screen via CrashOnCtrlScroll, but I'm still having trouble making the dump. I would try more, but my computer is crashing during boot (random colors/BSOD), or giving me the 4 Amber beeps error.

BadActor · Jul 3, 2016

Can you try running one stick of RAM in different slots and see if the crashes stop? Try this with both sticks.

ohjustwonderfuuu · Jul 4, 2016

After switching around the positions of the RAM sticks, I've been able to keep away the BSODs for the night. Will update if the computer crashes or blue screens again.

However, I still have the problem of low framerate in Overwatch and League. Although I'm able to start off with 50 FPS smooth, once my video card and CPU start heating up, my FPS drops to 15.

CPU (Min, Max): 37c, 64c
GPU (Min, Max): 55c, 113c

Considering I've dusted out the computer already, what should I do from here? Do I try and replace the thermal paste and see if that works?

BadActor · Jul 4, 2016

The CPU temps aren't that bad, but the GPU temps are high. Considering the crashes, the high temps, and poor game performance, your graphics card is a likely suspect. You might try removing it and running on the integrated graphics to rule out problems with the motherboard, CPU, and PSU. If you can, try the graphics card in another PC and see if the problem follows.

ohjustwonderfuuu · Jul 6, 2016

Well, after all of that work, my computer's back to normal!

What fixed my problems:
1) Switching around the positions of the RAM sticks
2) Cleaning out my GPU fan -- turns out it was so dusty that it couldn't spin.

The lack of minidump file generation is still a problem, however when the time comes I'll open up a new thread about that and fix that.

Thanks all for the help!

Computer crashes with different errors every time, and no minidump -- help!

Commendable

Distinguished

Reputable

Commendable

Reputable

Commendable

Polypheme

Commendable

Polypheme

Commendable

Polypheme

Commendable

Commendable

Polypheme

Commendable

Expert

Commendable

Expert

Commendable

Share this page