General Discussions

[TeachR]

So I've been having some issues with my graphics card lately and am hoping someone can help. The card keeps giving me unregular black screens and sudden system restarts. I've tried everything from a bios update to only installing te drivers and not the adrenalin software. Now I tried the undervolting and overclocking as a last resort to get the system working stable but it still crashes. Weirdly the crashes don't occur during stress testing or benchmarking the graphics card. There seems to be no obvious connection between high GPU usage and the crashes. At one point I thought my PSU isn't delivering enough power but then one would expect the black screens at high loads right?

Does anyone have any tips for a near desperate team red player?

My setup is as follows:

MSI B450-A pro max motherboard

Ryzen 5 3600

Corsair Cmk16Gx4M2Z3600C18 Vengeance Lpx 16Gb (2X8Gb) Ddr4 3600Mhz C18 Xmp 2.0 High Performance memory,

Kingston A2000 1TB flash drive 

be quiet! Straight Power 11 550W.

Running a windows 10 64-bit fully updated and the 20.12.1 driver/adrenalin version. Any help would be much appreciated.

[RPX100]

Please have a look at Windows Reliability Monitor and Event Log.
These are built in tools to monitor software, driver and operating system crashes.

Post all error messages and error/event codes here.

And also go ahead and start to google these codes yourself.
Often there can be simple solutions to these problems, or at least other people who solved it.

---

The current recommended driver for RX 5700 XT is version 20.11.2
You are using the optional 20.12.1, lets say "beta" driver (at least for navi 5000 series).

A fresh install of the driver might solve your problem:

1) Download DisplayDriverUninstaller from guru3d (link)
-> its an ZIP archive and you need to extract it before you can use it

2) run it as administrator in Windows Safe Mode (restart required)
-> it is important to use it in Safe Mode so it can clean all remaining files!

3) within DDU go to settings and block the Windows Updates (checkbox)
-> remember to undo this step when you are done with installing the new driver
-> if you don't want to use that option, you can also unplug your LAN cable instead

4) confirm the settings you made and then use DDU to uninstall the current driver
-> restart your system after this and install the recommended driver for your GPU (20.11.2)
-> restart system again

5) open DDU as administrator again and uncheck the block Windows Updates checkbox
-> you don't want your windows updates blocked permanently

[TeachR]

This was the latest error code from the reliability monitor

Source
Radeon Software: Host Application

Overview
Stopped working

Date
‎20-‎1-‎2021 15:53

Status
Verzonden rapport

Discription
Path to application with error: C:\Program Files\AMD\CNext\CNext\Radeonsoftware.exe

Problem with signature
Name of event: APPCRASH
Name of application: Radeonsoftware.exe
Version: 10.1.2.1829
Timestamp of application: 5fca253b
Name of errormodule: StackHash_bec9
Version of errormodule: 10.0.19041.662
Timestamp of errormodule: 27bfa5f0
Exceptioncode: c0000374
Exceptionmagin: PCH_B7_FROM_ntdll+0x000000000009CB04
OS version: 10.0.19042.2.0.0.256.48
Country-id: 1043
Aanvullende informatie 1: bec9
Aanvullende informatie 2: bec91acc4ae4a44497fbf0ca7c68191d
Aanvullende informatie 3: 726b
Aanvullende informatie 4: 726bc5d239ede16fcdd3bab29c1f1ab4

Extra informatie over het probleem
Bucket-id: 9b2efe2c1b8cfe8692820deeb1e9196b (1333643758493309291)

Bron
Windows

Overzicht
Hardware error

Datum
‎20-‎1-‎2021 15:53

Status
Niet gerapporteerd

Beschrijving
A problem with your hardware caused Windows to stop working correctly.

Probleem met handtekening
Naam van probleemgebeurtenis: LiveKernelEvent
Code: 141
Parameter 1: ffffe189d49a3050
Parameter 2: fffff8077d51e590
Parameter 3: 0
Parameter 4: 41e8
OS version: 10_0_19042
Service Pack: 0_0
Product: 256_1
OS Version: 10.0.19042.2.0.0.256.48
Locale ID: 8192

So it seems the AMD software is crashing. I did the DDU uninstall (again) and am now running the 20.11.2 drivers. Also did the factory reset when installing the drivers. 

Looking at this data it doesn't seem likely that this is a power failure of can that still be the case?

[RomatebitegeL]

When checking the recommended PSU for your graphics card, it seems to be 700w and not 550w, as you currently use. The AMD graphics card are known for having huge spikes of voltage occurring from time to time during load and perhaps even during normal usage and it may therefore be stable during times, and then an influx of increase voltage occurs suddenly, and if your PSU is not strong enough, the system may crash or reboot.

As for the common black screen issue, this is mainly fixed by reinstalling your graphics drivers with a factory reset and deletion of saved precepts. You do a factory reset in the beginning of the driver installation. This fix has always worked for my own black screen issues, and I use a Vega 56.

As for my self, I specifically bought a 1000w PSU for my Sapphire Vega 56 graphics card in order to comply with the 700w PSU recommendation, and in order to future proof my self.

If you buy a stronger PSU, at least a 700w, your sudden restarts of your system should be avoided.

[RPX100]

---

@RomatebitegeL wrote:

When checking the recommended PSU for your graphics card, it seems to be 700w and not 550w.

https://www.techpowerup.com/gpu-specs/powercolor-red-devil-rx-5700-xt-oc.b7249
• TDP: 225W 
• suggested PSU: 550W

---

@RomatebitegeL wrote:

The AMD graphics card are known for having huge spikes of voltage occurring from time to time during load and perhaps even during normal usage and it may therefore be stable during times, and then an influx of increase voltage occurs suddenly, and if your PSU is not strong enough, the system may crash or reboot.

While this could indeed be one possible reason, I don't think it should be the first thing to recommend another PSU.
There are enough driver related issues that can cause this. Even other system instabilities could be the reason.

[RPX100]

@TeachR LiveKernelEvent 141 does indeed look like an driver issue, rather than power problems.

This error could for example also be caused by external devices connected to your PC.
Like USB devices that need their own driver.
They might cause this because of bad drivers or incompatibility between driver versions.

Check Windows Device Manager for any issues or errors / unknown devices.
This might not be the root cause of the problem, but better make sure it isn't.

Also run Windows System File Checker:
https://www.partitionwizard.com/partitionmagic/system-file-checker-windows-10.html 

[TeachR]

I ran the checker and it fixxed a number of files, now hoping the problems will stop. All the connected devices have the drivers installed.

[TeachR]

Ok so I haven’t had the time to test the new settings but meanwhile I got a response on an email I sent to the manufacturer. They stated that the GPU requires at least a 700W PSU to operate correctly. This is making me wonder is the 550W PSU I currently have might be insufficient to handle powerspikes. Does anyone have any thoughts on this? I did some forum searching and apparently a shortage of power can cause a system to restart. 

For the sake of checking if this might be the issue I ordered a 700W PSU to see if this fixes the problem. After contacting a friend of mine who is a computer specialist and him telling me that random system reboots are usually power or heat related I thought I would give it a go.

Will let you guys know if this solves the problem as it might be a solution for someone else too.

[TeachR]

Ok so I promised an update and here it is. At the moment it looks like my black screen and reboot issues are gone. To achieve this I did a couple of things:

1. Upgrade my PSU to a 700W unit. This immediately fixed the rebooting of the system. But it did not fix the fact that the radeon adrenalin software was timing out while playing games (especially No Mans Sky). So I went ahead and searched for possible solutions for this problem. That led to the second step.

2. I undervolted my GPU and went to tweak the VRAM clock settings. At the moment I use a VRAM clock speed of 1770 which is stable. My GPU frequency is at 1999 with 1116mV. That however did not fix the problem in No Man's Sky with the time out of the adrenalin software. I fixed that by setting a manual fan curve that keeps the GPU at 60 degrees under load. The system does make some noise while gaming now but I really don't mind that.

For people who have the same issue I'm now running the following settings in the adrenalin software. You can copy this code and put it in an XML file to load it into the adrenalin software.

<GPU DevID="731F" RevID="C1">
    <PPW Value="3"/>
    <FEATURE ID="100" Enabled="0">
        <STATES>
            <STATE ID="0" Enabled="False" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="101" Enabled="7">
        <STATES>
            <STATE ID="0" Enabled="True" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="102" Enabled="0">
        <STATES>
            <STATE ID="0" Enabled="True" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="4" Enabled="True">
        <STATES>
            <STATE ID="0" Enabled="False" Value="800"/>
            <STATE ID="1" Enabled="False" Value="1399"/>
            <STATE ID="2" Enabled="False" Value="1999"/>
            <STATE ID="3" Enabled="False" Value="800"/>
            <STATE ID="4" Enabled="False" Value="1999"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="12" Enabled="False">
        <STATES>
            <STATE ID="0" Enabled="False" Value="697"/>
            <STATE ID="1" Enabled="False" Value="830"/>
            <STATE ID="2" Enabled="False" Value="1116"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="5" Enabled="True">
        <STATES>
            <STATE ID="0" Enabled="False" Value="1770"/>
            <STATE ID="1" Enabled="False" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="9" Enabled="False">
        <STATES>
            <STATE ID="0" Enabled="False" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="8" Enabled="False">
        <STATES>
            <STATE ID="0" Enabled="True" Value="0"/>
            <STATE ID="1" Enabled="True" Value="0"/>
            <STATE ID="2" Enabled="True" Value="50"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="19" Enabled="False">
        <STATES>
            <STATE ID="0" Enabled="True" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="20" Enabled="False">
        <STATES>
            <STATE ID="0" Enabled="True" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="21" Enabled="False">
        <STATES>
            <STATE ID="0" Enabled="True" Value="0"/>
        </STATES>
    </FEATURE>
    <FEATURE ID="22" Enabled="True">
        <STATES>
            <STATE ID="0" Enabled="False" Value="39"/>
            <STATE ID="1" Enabled="False" Value="30"/>
            <STATE ID="2" Enabled="False" Value="51"/>
            <STATE ID="3" Enabled="False" Value="38"/>
            <STATE ID="4" Enabled="False" Value="64"/>
            <STATE ID="5" Enabled="False" Value="50"/>
            <STATE ID="6" Enabled="False" Value="80"/>
            <STATE ID="7" Enabled="False" Value="63"/>
            <STATE ID="8" Enabled="False" Value="93"/>
            <STATE ID="9" Enabled="False" Value="83"/>
        </STATES>
    </FEATURE>
</GPU>

I hope this will help anyone who is experiencing the same problems

[TeachR]

Ok so I thought I could share my journey to a functional pc. In spite of my last post the system was not stable with the settings posted above. At the time of writing this post my system does seem stable. The GPU has been stable for a while now so this might help someone else. So for the fix for the GPU:

After months of trying I decided to reinstall windows. When I took off my case lids to reseat the GPU I noticed that the bios switch was set to the OC bios. I switched this to the silent bios and reinstalled windows. After this I installed the latest AMD driver, this was the 21.3.1 version I believe. Ever since the GPU has been stable so I blame the OC bios and think this is not stable. 

After fixing this my system kept crashing because of my headset with built in sound card. I own the HyperX 2 and this keeps on screwing with my system. A firmware update did not do the trick but after removing the USB sound card this was working fine. 

With all of this out of the way I thought this would be it. But then my system started shutting down due to cache hierarchy errors. After doing some research into this I read that doing a bios update could fix it and it did. Now I’m hoping it will stay stable for a while. 

These were my 2 cents. Just trying to help people who might have the same problems. 

[shoesfx]

have you found a solution? i'm having random black screen crashes now and then and mostly on new game releases.

[ThunderBeaver]

I was having a similar issue with my Power Color Demon RX 6900 XT.

I was running an EVGA 850 watt PSU (Recommended). Then I upgraded to a Thermaltake ToughPower 1200 watt and the problem went away.

I would suggest a minimum of a 750 watt PSU but I would recommend a 900 watt PSU so you have plenty of overclocking headroom. Plus PSU's are one of the few PC components that didn't skyrocket in price. Checkout Newegg.com they're great for comparison shopping PC parts.

[RPX100]

@shoesfx hello there!

RX 5000 series crash types and reasons

Let me know if that helped you. Cheers!