Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

Do I need a new GPU? Strange Crashes


  • Please log in to reply
4 replies to this topic

#1 0xGUI

0xGUI

  • Members
  • 8 posts
  • OFFLINE
  •  
  • Local time:01:45 AM

Posted 26 July 2017 - 02:24 PM

Operating System:

Windows 10

 

Computer Specs:

PSU: Corsair TX650 W

GPU: MSI R9 270X

CPU: AMD Phenom II x6 1055T ( OC was removed)

RAM: 4GB 1600 MHZ DDR3 TeamGroup ELite

MotherBoard: Asus M4A89GTD Pro/USB3 ( IGPU AMD ATI 4290)

Storage: WD blue 1TB 7200 rpm SSD Evo 840 120 GB

 

Description of problem:

Hello all,

 

To clarify, computer crashes/freezes at random times. Just simply browsing the web or playing a game the computer crashes.

The crashes are represented by a fixed image( freezed image of what I was doing before ) and some weird noises( not looping or continuous sound) that sound like what an old 56k modem would do.

 

When this issue began:

About a month ago. Didn't change anything in terms of hardware.

 

Recurring issue:

Yes

 

Cause/Steps to recreate the issue:

I found a way to replicate the crashes that almost always occurs.

A Firefox tab and a Chrome tab are open at the same time, playing 1080p Youtube videos side by side.

After some time it will eventually crash.

 

What I've tried so far to resolve the issue:

So I made various test that helped me realize maybe the GPU was faulty:

Fresh Windows 10 / newest AMD drivers / HDD SSD = crash

Fresh Windows 10 / old AMD drivers / HDD SDD = crash

Fresh Windows 10 / newest AMD drivers / only SSD = crash

Fresh Windows 10 / newest AMD drivers / only HDD = crash

Fresh Windows 7 / newest AMD drivers / only SSD = crash

Linux distro ( manjaro) / proprietary drivers /only SSD = crash

Linux distro ( manjaro) / IGPU -ATI AMD 4290 /SSD HDD = OK

OCCT and AIDA64 stablity test during 1h both = OK

Memtest86= 0 errors so RAM is OK.

HDD and SDD= 0 errors using crystaldisckinfo and Windows tools.

SDD firmware was updated to the last.

Temperatures are OK using HWMonitor.

Testing now:

Fresh Windows 10 / IGPU -ATI AMD 4290 /SSD HDD = for about 1h and crash as not occurred.

Did not tested PSU, but the computer didn't crash during stress tests.

 

Is there anything else I can test/do to make sure it's a GPU problem?


Edited by 0xGUI, 26 July 2017 - 02:25 PM.


BC AdBot (Login to Remove)

 


#2 hamluis

hamluis

    Moderator


  • Moderator
  • 56,287 posts
  • ONLINE
  •  
  • Gender:Male
  • Location:Killeen, TX
  • Local time:07:45 PM

Posted 26 July 2017 - 03:59 PM

BC BSODs Posting Instructions - http://www.bleepingcomputer.com/forums/t/576314/blue-screen-of-death-bsod-posting-instructions-windows-10-81-8-7-vista/ .

 

Louis


Edited by hamluis, 26 July 2017 - 03:59 PM.


#3 usasma

usasma

    Still visually handicapped (avatar is memory developed by my Dad


  • BSOD Kernel Dump Expert
  • 25,091 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Southeastern CT, USA
  • Local time:08:45 PM

Posted 26 July 2017 - 05:52 PM

In addition to hamluis' suggestion, I'd also like to see this report (after the next crash - recover from the crash, then immediately generate the report):

 

Please do the following:
- open Event Viewer (run eventvwr.msc from the "Run" dialog)
- expand the Custom Views category (left click on the > next to the words "Custom Views")
- right click on the "Administrative Events" heading
- select "Save all Events in Custom View as..."
- save the file as Admin.evtx
- zip up the file (right click on it, select "Send to", select "Compressed (zipped) folder")
- upload it with your next post (if it's too big, then upload it to a free file-hosting service and post a link here).

FYI - If we're looking for Event ID 41 errors (unexplained shutdowns), there's more info on that here:  http://support.microsoft.com/kb/2028504

While waiting for a reply, please monitor your temps with this free utility: 

SpeedFan v. 4.5.1 and later (free from here:  http://www.almico.com/sfdownload.php ) can log temperatures in a CSV file:
You may not need to log temps.  If uncertain, please ask.

To make it work (log the temps to a file) you have to do BOTH of the below:

1. Enable logging in general: Configure...Log...check "Enabled" then click on OK to save.

2. Enable logging for specific checks: Configure...Temperatures..left click on each sensor, then click on "Logged" at the bottom of the Window (for our purposes we want them all) then once you've selected "Logged" for all sensors, then click on OK to save.

The log will be located at C:\Program Files (x86)\SpeedFan

Naming: log files are named SFLogYYYYMMDD.csv, where YYYY is the year (four digits), MM is the month (2 digits, zero padded) and DD is the day (2 digits, zero padded). If a file already exists by that name, the file that already exists is renamed according to the following naming scheme: SFLogYYYYMMDD-CCCC.csv, where CCCC is a increasing number. The new file is then created with the standard file name scheme.

Notes: whenever you change the options related with logging, SpeedFan starts a new log file.

NOTE:You may want to turn logging off when we're done - as I don't know it's impact on performance or on the system.

 

Finally, as you suspect hardware, please run ALL of these hardware diagnostics:  http://www.carrona.org/hwdiag.html

Please run ALL of the tests and let us know the results - then we can move on from there.

 

Now for my discussion..........

 

First off, advances in video card testing have not kept up with advances in video card technology.

Simply put, we still use the same old tests (from years ago) to test the newest video cards.

 

What I'm seeing is video card failures that do not show up in any of the diagnostics.

As such, I'm tending to think that the only fully effective video card diagnostic is to try another video card.

But, the problem is that this can get expensive.  So I suggest other tests/analysis' first.

 

Running the old tests is helpful - if there's a problem with the part of the video card that the old tests operate on.

BUT, the old tests are not effective enough (IMO) to rule out all hardware problems.

 

So, what I'm suggesting is a combination test/analysis.  In short, if the problems persist, and everything software is fixed from the analysis - then we still have to wonder about hardware.

This especially includes removing all traces of the drivers (either with the AMD tool, or the freeware DDU tool) before downloading/installing a fresh copy of the latest, W10 compatible version of the drivers for your video card.  Then comes the hardware tests (all of those that I suggest here:  http://www.carrona.org/hwdiag.html )

 

If all the hardware tests pass, and the errors still appear to be from the video card, then it's the time to try another video card.

If you don't have one, or can't borrow one - then you're left with purchasing one.

I'd suggest purchasing it from a shop that'll let you return it for your money back if it's not needed.


My browser caused a flood of traffic, sio my IP address was banned. Hope to fix it soon. Will get back to posting as soon as Im able.

- John  (my website: http://www.carrona.org/ )**If you need a more detailed explanation, please ask for it. I have the Knack. **  If I haven't replied in 48 hours, please send me a message. My eye problems have recently increased and I'm having difficult reading posts. (23 Nov 2017)FYI - I am completely blind in the right eye and ~30% blind in the left eye.<p>If the eye problems get worse suddenly, I may not be able to respond.If that's the case and help is needed, please PM a staff member for assistance.

#4 0xGUI

0xGUI
  • Topic Starter

  • Members
  • 8 posts
  • OFFLINE
  •  
  • Local time:01:45 AM

Posted 26 July 2017 - 06:39 PM

 

Now for my discussion..........

 

First off, advances in video card testing have not kept up with advances in video card technology.

Simply put, we still use the same old tests (from years ago) to test the newest video cards.

 

What I'm seeing is video card failures that do not show up in any of the diagnostics.

As such, I'm tending to think that the only fully effective video card diagnostic is to try another video card.

But, the problem is that this can get expensive.  So I suggest other tests/analysis' first.

 

Running the old tests is helpful - if there's a problem with the part of the video card that the old tests operate on.

BUT, the old tests are not effective enough (IMO) to rule out all hardware problems.

 

So, what I'm suggesting is a combination test/analysis.  In short, if the problems persist, and everything software is fixed from the analysis - then we still have to wonder about hardware.

This especially includes removing all traces of the drivers (either with the AMD tool, or the freeware DDU tool) before downloading/installing a fresh copy of the latest, W10 compatible version of the drivers for your video card.  Then comes the hardware tests (all of those that I suggest here:  http://www.carrona.org/hwdiag.html )

 

If all the hardware tests pass, and the errors still appear to be from the video card, then it's the time to try another video card.

If you don't have one, or can't borrow one - then you're left with purchasing one.

I'd suggest purchasing it from a shop that'll let you return it for your money back if it's not needed.

 

 

The only video card that I can use besides the  R9 270x is the on-board one. Using the on-board GPU, with a Linux Distro and Windows 10, I have not experienced any kind of crash when using  the steps to recreate the issue.

Tomorrow I will post the results of the various test that were proposed here.


Edited by 0xGUI, 26 July 2017 - 06:39 PM.


#5 0xGUI

0xGUI
  • Topic Starter

  • Members
  • 8 posts
  • OFFLINE
  •  
  • Local time:01:45 AM

Posted 11 August 2017 - 03:45 PM

UPDATE:

 

I'm here just updating this topic for in case someone has the same problem.

After all, the GPU was fine and the culprit was the motherboard. It got to a point where after the crashes, a BIOS screen appeared saying that Overclocking Failed, although I did not have any type of OC present. From here I started to doubt the motherboard.

Another situation that occurred was when I was in BIOS setup screen the screen just freezes and I couldn't do anything.

At last the motherboard finally gave up and didn't want to POST.

 

Bought a new AM3+ MB and it seems that the problem as been fixed for the past couple days.

 

Cheers!






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users