cancel
Showing results for 
Search instead for 
Did you mean: 

Graphics Cards

admaster99
Journeyman III

Is this signs of a faulty gpu?

I bought a Vega 56 a couple weeks ago I notice after playing Far Cry 5 for about 20-30 minutes I usually get what happens in the images attached. I'm wondering if this is signs of a faulty gpu? Or maybe it's just a graphical bug with the game. I haven't noticed it on any other games but I haven't really played anything else so it's hard to say for sure. Any help is greatly appreciated thanks.

Specs:

Desktop

PowerColor AMD Radeon Vega 56 RED Dragon HBM2 8GB (AXRX Vega 56 8GBHBM2-2D2HD/OC)

Windows 10 64bit

Radeon Software Version 18.9.2

Display Samsung 55" NU8000 w/ Single HDMI 2.0 Connection running at 1080p/1440p 120 hz w/ freesync

Asus Z170-A ATX DDR4 Motherboard

i7 6700k CPU /w factory clocks

750 Watt PSU Coolermaster

G.SKILL Ripjaws V Series 16GB (2 x 8GB) 288-Pin DDR4 SDRAM DDR4 3000 (PC4 24000)Model F4-3000C15D-16GVR

0 Likes
8 Replies

If will help immensely by other Users to know your computer specs as per AMD Forum rules: INFORMATION REQUIRED WHEN POSTING A QUESTION

If these artifacts only happens during one game and nowhere else it could be a problem with the game itself or the AMD Driver with that particular  game.

Is your game fully updated to the latest version?

Try installing a Previous AMD Driver and see if you get the same results.

Do you have video artifacts whenever the GPU is under heavy loads?

Is your PSU strong enough for the new Vega54 GPU Card?

You can always test your GPU Card by using a free diagnostic program called OCCT (Download​). There are others but I personally like this one because it checks the GPU under stress and for errors plus it also checks the PSU which no other diagnostic program does.

The game is currently up to date. I also have a 750 watt PSU which I believe should be adequate. This is the only time I have notice any kind of graphical glitches/artifacts but I haven't really played anything else for extended amounts of time.

I almost want to say that it only happens when I have RTSS Overlay active but I have a hard time believing that RTSS can cause this kind of graphical glitches but I'm gonna try and play several more hours this weekend and see if it happens again. I have an extra RX 580 I've been meaning to sell that I might swap in and test for several hours to see if it happens again.

I'm gonna also try the OCCT to see what kind of results I can produce.

0 Likes

Your PSU is plenty for the Vega54.

RTSS may be interfering with the AMD Driver indirectly or with the game itself. I would, just for troubleshooting purposes, uninstall it and see if the problem occurs again.

Many people have problems with AMD Driver when 3rd party software similar to MSI Afterburner is installed. I know that you can now download RTSS as a stand alone program without having to install MSI Afterburner.

There is no harm in using OCCT to test your GPU card. It will show Temperatures and PSU Voltages during the test and other useful data. OCCT creates a folder called OCCT in "Documents" folder filled with graphical data charts from the test, whether the test finished or crashed.

Make sure your PSU voltages 3.3vdc, 5.0vdc, and 12 vdc are within tolerances of 5% or less during the testing. I have learned that some motherboard sensors may not be compatible with OCCT and you may need to use a different monitoring software to get PSU voltages and other data.

AIDA64 is another good diagnostic program to stress test the GPU.

I would first try and see if you can find out if the GPU artifacts is something that occurs all the time or under GPU stress times or if it just occurs in one particular game or in all games.

If it happens all the time in all games or when the GPU is under stress than it could be a Driver or GPU card issue. 

admaster99
Journeyman III

Update

So I couldn't get OCCPT to work correctly so I proceed to use Furmark. I ran a stress test for 2 hours with no visual artifacts and a maximum temperature of 75 degrees I forgot to check the voltages. So I figured that was good and I removed MSI Afterburner and RTSS also from my PC. I haven't had that specific  problem reoccur. But just now I was playing Doom for about 1 hour then my display(Samsung 55" NU8000) went all crazy so I took a picture with my phone. I Alt+Tab to the desktop to make sure it wasn't just the game. Even after doing a hard reset the image was still like this until I unhooked the hdmi cord for a couple of minutes.

20181019_184542copy.jpg

So as of right now I'm not sure if it was the display it's self that did this or the GPU and I'm kind of freaking out because I paid like $1,000 for the display. Please help anybody thanks.

0 Likes

Samsung can remotely troubleshoot your television set as long as you have an internet connection to the television. They can update firmware, configure your television for the best viewing for the whatever use you using it for and troubleshoot any issues.

Here is the Manual for your Television set from Samsung with an image showing how to connect Samsung Support Technician remotely: http://downloadcenter.samsung.com/content/UM/201809/20180927112738059/ENG_US_KM2ATSCN-3.0.2.pdf

I appreciate the reply but does this mean you do not believe this to be a GPU related issue?

0 Likes

No, it could still be the GPU or Power issue to the GPU or even a driver issue. I just mentioned the previous comment since you said you weren't sure if it was the Television or GPU at fault.

The Previous comment will eliminate your Samsung television from the equation if Samsung Remotely troubleshoots and properly configures your Television set to work with a PC attached. Plus they will make sure the Set is up to date with the latest firmware.

The easiest way to eliminate the TV as a culprit is by connecting another Monitor and see if the same thing occurs with a different monitor attached.

What problems were you having running OCCT?

Which test were you trying to run with OCCT?

EDIT: some users are having issues with Freesync enabled. Disable Freesync and see if the problem changes or stops.

Update

I have updated my Display's firmware to the latest version. I have not currently tried the remote support with Samsung due to me needing to obtain a PIN. With that being said I did play Doom for a solid 2 hours last night with no issues. I do agree that using another display would be a good way to troubleshoot the issue, sadly it's hard to reproduce the problem because the issue is so intermittent. If I could pinpoint an exact time or place in a game when it happens, then double check with a second display that would be awesome. But so far it just seems completely random when I get graphical corruption. I also find this to be the same with freesync it's hard to test it directly when you don't know when the issue is going to occur.

After spending hours researching the Vega 56 I've come to realize that it requires quite a bit of power from the PSU. Which kind of makes me wonder if mine is adequate... GX 750W - 750 Watt 80 PLUS Bronze Certified Active PFC Power Supply | Cooler Master ....... it's fairly old at about 7+ years since I bought it brand new.

In your first response you told me to make sure the voltage was in a certain range with OCCT. How exactly is this done?

OCCT wont start in directx11 it only works in directx9 is this ok? And how do I check my voltages?

Under "GPU: 3D" tab I was running "Infinite" test type at 1920x1080 resolution. That is how I got it to work, I ran it for 1 hour last night but was not sure what I was looking for exactly. I did not see any artifacts and the temperature was comparable to that of furmark.

I really appreciate all your input thanks.

0 Likes