PC Graphics

[spectralfx]

Hello,

I am currently trying to fix some problems that have been plaguing my gaming desktop. I have been experiencing multiple BSOD related to many different stop codes. However all of them pointed to hardware issues. Until recently I couldn't attribute it to overheating issues however a recent discovery left me quite alarmed.

The one thing these BSOD had in common is the fact that they pretty much all occurred during gaming sessions or sometimes, rather exceptionally, during video streaming over my web browser.

After much troubleshooting, including clean reinstalling my Graphic card drivers (atm running 20.1.2) using DDU in safe mode and manually reinstalling the drivers for my motherboard and many, many other troubleshooting steps available out there...

I appear to have collected some relevant data using 3Dmark that exposes some rather unusual behavior from my hardware. It appears that  for some reason, my GPU goes unstable - or - unresponsive due to a spike in both frequency and temperature. I have never witness such symptoms before.

This predicament leaves me quite perplexed. It appears that under stress, my computer attempts to crank up the frequency of the GPU drastically until the whole thing becomes unstable. Thankfully, the tool I used didn't cause my computer to BSOD like it would typically do in games and I was able to collect the data from the attachment.

I have the report from 3Dmark saved in a safe location and can submit it if required along with any other log file, but for now I am just looking for information about the behavior itself.

What could possibly cause this phenomenon and how would I go about ensuring this undesirable behavior stops?

[spectralfx]

Here is the report on a crash I had recently, directly related to AMD Radeon software.

I suspect its related to my problems:

Source
Radeon Software: Host Application

Summary
Stopped working

Date
‎2020-‎01-‎21 3:06 PM

Status
Report sent

Description
Faulting Application Path: C:\Program Files\AMD\CNext\CNext\Radeonsoftware.exe

Problem signature
Problem Event Name: APPCRASH
Application Name: Radeonsoftware.exe
Application Version: 10.1.2.1788
Application Timestamp: 5e1cdfe0
Fault Module Name: Radeonsoftware.exe
Fault Module Version: 10.1.2.1788
Fault Module Timestamp: 5e1cdfe0
Exception Code: c0000005
Exception Offset: 0000000000405657
OS Version: 10.0.18362.2.0.0.256.48
Locale ID: 4105
Additional Information 1: 0f1e
Additional Information 2: 0f1e8efbe4eda39817605f485fdb728a
Additional Information 3: a12e
Additional Information 4: a12eb06be8146e7b2c39db3f449b67c1

Extra information about the problem
Bucket ID: dc1b6c30e402a1ab21f90a1c57cca997 (1295077484694120855)

[spectralfx]

Just did a Clean install of the newest version, after reinstalling windows altogether,

and still experiencing this issue.

Please, someone out there at AMD, do something to assist me.

Source
Radeon Software: Host Application

Summary
Stopped working

Date
‎2020-‎01-‎24 10:18 PM

Status
Report sent

Description
Faulting Application Path: C:\Program Files\AMD\CNext\CNext\Radeonsoftware.exe

Problem signature
Problem Event Name: APPCRASH
Application Name: Radeonsoftware.exe
Application Version: 10.1.2.1788
Application Timestamp: 5e221339
Fault Module Name: Radeonsoftware.exe
Fault Module Version: 10.1.2.1788
Fault Module Timestamp: 5e221339
Exception Code: c0000005
Exception Offset: 0000000000405407
OS Version: 10.0.18363.2.0.0.256.48
Locale ID: 4105
Additional Information 1: 0f1e
Additional Information 2: 0f1e8efbe4eda39817605f485fdb728a
Additional Information 3: e8be
Additional Information 4: e8be9d812a8f09acb40b4093b1c3d836

Extra information about the problem
Bucket ID: 043065ce9651e416e90a6a1938a7afb7 (1804371257272545207)

[hardcoregames_]

hardware specifics?

[spectralfx]

Operating System: Windows 10 Pro 64-bit (10.0, Build 18363) (18362.19h1_release.190318-1202)

Processor: AMD Ryzen 5 3600 6-Core Processor               (12 CPUs), ~3.6GHz

Memory: 16384MB RAM (G.SKILL Ripjaws V Series 16GB (2 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Desktop Memory Model F4-3200C16D-16GVGB)

Graphic card: Sapphire PULSE Radeon RX 580 8GB GDDR5 PCI-E Dual HDMI / DVI-D / Dual DP OC w/ Backplate (UEFI), 100411P8GOCL

PSU: Rosewill PHOTON Series 650W Full Modular Gaming Power Supply, 80 PLUS Gold Certified, Single +12V Rail, Intel 4th Gen CPU Ready, SLI & Crossfire Ready - Photon-650

Motherboard: ASUS PRIME X570-P ATX Motherboard - Socket AM4 - AMD X570 Chipset - Supports DDR4-4400(OC) - 2x PCIe 4.0 x16 - 3x PCIe 4.0 x1 - CrossFireX Ready - 2x M.2 Socket3 - HDMI - USB 3.2 Gen2 Type-A - Gigabit LAN

And obviously some SSDs and HDD, just let me know if you need more information.

I ran Tests on the RAM (memtest and such) for 8 passes, not issue on that end. No system data corruption was found upon running an SFC /scannow either.

[hardcoregames_]

I refer you to a long running thread on the R5 3600 processor.

Not sure why but the november 2019 1.0.0.4B BIOS update fixed most of the problems but not all of the issues I have uncovered. BIOS 1405 is the latest one so far for your motherboard. Not sure if a new one will surface anytime soon.

The problems are not memory related, it has more to do with drivers and 20.1.3 is the latest offering

if necessary DDU can be used to clean up driver messes

[spectralfx]

I already run 20.1.3, as stated in some previous post.

I already used DDU to update my drivers.

Any fix on this issue or at least acknowledgement on AMD's part?

[spectralfx]

The issues are persisting and are preventing me from using my system by constantly causing it to crash!

The steps provided have done nothing to stop this behavior!

Please provide further assistance on this matter. This is a serious issue caused by the Radeon driver.

Source
Windows

Summary
Shut down unexpectedly

Date
‎2020-‎01-‎26 2:59 AM

Status
Not reported

Problem signature
Problem Event Name: BlueScreen
Code: a
Parameter 1: 188
Parameter 2: ff
Parameter 3: c3
Parameter 4: fffff8073bdd2e3b
OS version: 10_0_18363
Service Pack: 0_0
Product: 256_1
OS Version: 10.0.18363.2.0.0.256.48
Locale ID: 4105

Source
Radeon Software: Host Application

Summary
Stopped working

Date
‎2020-‎01-‎26 2:59 AM

Status
Report sent

Description
Faulting Application Path: C:\Program Files\AMD\CNext\CNext\Radeonsoftware.exe

Problem signature
Problem Event Name: APPCRASH
Application Name: Radeonsoftware.exe
Application Version: 10.1.2.1788
Application Timestamp: 5e221339
Fault Module Name: Radeonsoftware.exe
Fault Module Version: 10.1.2.1788
Fault Module Timestamp: 5e221339
Exception Code: c0000005
Exception Offset: 0000000000405407
OS Version: 10.0.18363.2.0.0.256.48
Locale ID: 4105
Additional Information 1: 0f1e
Additional Information 2: 0f1e8efbe4eda39817605f485fdb728a
Additional Information 3: e8be
Additional Information 4: e8be9d812a8f09acb40b4093b1c3d836

Extra information about the problem
Bucket ID: 043065ce9651e416e90a6a1938a7afb7 (1804371257272545207)

[hardcoregames_]

I guess all you can do it wait for the next driver like the rest of us

[spectralfx · Accepted solution]

I'll just request a refund on this card that is still under warranty and get a card from Nvidia then.

The way this matter has been handled after three updates ... not even a "known issue" statment from them...

AMD never provide decent drivers... its always a mess...

They can't even update it without having the code to collapse on itself (hence us having to use DDU all the time)

That is totally unaceptable, a real clown show.

[hardcoregames_]

I keep DDU on hand with a driver archive so when I need to change video cards it is easier to do with fewer headaches

[spectralfx]

The need of having to do that in the first place is a problem, not a solution.

[tgodtaylor]

Did this helped?
Can you install the Drivers with the new Card now?

[masterseby]

Hey idk if it will work for you but, try and disable Zero RPM. After I disabled this my gpu usage got stable and temperature as well.

[spectralfx]

No, this does not help with my issue.

In essence, what I am facing here is some Access_violation errors.

Its most likely a programming issue as I've tested every single parts of my system now, even RMA the Graphic card and still have this issue going.

I tested my ram, my Drives, did stress tests on the processors ... everything I can possibly clear with tests passes.

Ran a bunch of commands to restore windows, used DDU more times than I can count ... still get the damn issue...

Heck even my Bios for the Mobo is up to date!

I sincerely do not know what else I could possibly do other than spending money on some completely different hardware from a competitor...

you know...

in order to get drivers that work somewhat better with Windows ...