PC Drivers & Software

[alexx855]

Hello, im having this weird problem since i brought my new GPU a few months ago, my whole build is new, the GPU was the last  purchase, previously i had a R9 290x in this same build, all works perfectly!

The crash is sometimes a black screen on all monitors and fans turns 100% speed, nothing to do here just force shutdown, sometime i got a blue screen with the message, THREAD STUCK IN DEVICE DRIVER.

All this happens randomly, between 1 hours to 4 maximum,  while gaming and casually browse on my second monitor.

i have tried all the possibilities, clean windows install, a lot of drivers versions 18.x.x / 19.x.x, diferente pci-e socket, diferente PSU, two diferents power rails, undervolt, power limit +50%, GPU and motherboard BIOS update, basically all known tricks. 

My systyem:

• AMD Graphics Card
  • Vega 56 Gigabyte OC version gigabyte.com/Graphics-Card/GV-RXVEGA56GAMING-OC-8GD
• Operating System
  • Desktop Windows 10 64bit
• Driver version installed
  • 19.11
•  Display Devices
  • LG IPS FULLHD (1920x1080@59Hz Active DP to HDMI adapter
  • 23MP55 (1080x1920@59Hz) HDMI
  • HISENSE (1920x1080@59Hz) HDMI
• Motherboard + Bios Revision
  • Z370 AORUS Gaming 5 + Bios F10
• CPU/APU
  • Intel Core i7 8700k
• Power Supply Unit  Make, Model & Wattage
  • Seasonic 750W Gold 
• RAM
  • 16GB Gskyll 4133hz

EDIT UPDATE:

The only solutions that works is the driver 17.11.1

https://www.amd.com/en/support/kb/release-notes/rn-rad-win-17-11-1  

whql-win10-64bit-radeon-software-crimson-relive-17.11.1-nov10.exe - Google Drive 

EDIT UPDATE 2020 

I have been testing the new bios, for a week i think, no crashes.

UPDATE THE BIOS USING THIS VGA Bios Collection: Gigabyte RX Vega 56 8 GB | TechPowerUp  (or search for your manufacturer if its not Gigabyte )

Im using the default settings, with the lastest driver version,  turned off zero rpm,  changed the fan speed and power limit -50%

EDIT UPDATE MID 2020

Radeon pro drivers seems to fix all problems

https://www.amd.com/en/support/professional-graphics/radeon-pro/radeon-pro-wx-x200-series/radeon-pro... 

[munchabunch]

By undervolting you might limit your overclocking capabilities, yes. But these cards LOVE to be undervolted. You can look it up  

Mine is running fine at 1552 core clock with 1000mV and temps below 60c. That's still quite a gain over the reference Vega 56's 1474 boost core clock. And I'd rather use optimised drivers instead of the default one.

But again, it is definitely not ideal so this fix might not be for everyone.

[neophyte]

I can agree with this method to undervolt and drop clocks slightly, my last post I posted too soon without giving it time. So far what I've done since then is set the fan "zero rpm" to off (just before going into a game) and lowered the last 3 states and clocks a bit, and power limit to 0. I use vsync on all my games, in CSGO for example without vsync on the frames go above 300 and clocks are running 1400 - 1500 and fans are crazy, which to me is way overboard for this game. So at 60 fps the clocks get set much lower, and before I switched fan to "zero rpm" off with vsync on, the fans didn't even switch on at all, and as one can imagine playing a full match of around 40 - 50 mins without any cooling certain parts of the card will get very hot. I noticed this too playing pubg, only every now and then fans will come on briefly once 55 degrees C was reached then switch off again, and all of this is automatic settings that wattman sets. (Powersave balanced and turbo)

I updated to 19.4.2 drivers on the 15th, and have had no freezes yet. So far so good with this method. It's a shlep to have to go through all this trial and error to find stability, but when this card is performing its seriously beast. 

[heartless_ca]

I tried to undervolt and it lasts a bit longer but still crashes. I'm RMAing my card. I'm sick of it. Popped in my 1060 and I'll see if i BSOD or crash. Will report back when I get a new card.

[system2000]

maybe gigabyte cheaped out on the thermal pads below the backplate, maybe changing them (if the card is out of warranty) to high quality ones could fix the problem

[yeungcsm]

I'm having exactly the same issue with MSI Vega 56 airboost, trying to install 17.7.2, the installer said installed, but when looking at the device manager, it seems not installed, is it installed differently?

Considering install latest driver without installing wattman, but I will lots chill and freesync right?

[vendett3]

This is a longshot, but could you check the directx diagnostic tool? You can access it by presing windows key+r and then entering the code dxdiag on the console. Then, under the "screen" tab, you should be able to see what driver version you have install. It's a huge "if" since the device manager ought to show you the driver... Hope it helps!

[yeungcsm]

I will see if my latest attempt fix it, installed latest driver without AMD setting, as i heard the wattman is the problem. 

Does 17.7.2 comes with any AMD setting, because I'm sure after installation, no other software etc was installed

[yeungcsm]

Just had another black screen with fan on max, checked logger, temperature was around 75 degree in gpu. Thats with the latest driver without wattman. How did you install the 17.7.2,just confirmed with dxdiag, it's still showing windows basic

[vendett3]

Here you go:AMD Radeon Software Crimson ReLive 17.7.2 driver download 
Guru3d are the same guys that made display driver uninstaller.  Or you could try the official webpage: Radeon Software Crimson ReLive Edition 17.7.2 Release Notes | AMD  The download links are right at the bottom of the page.

[yeungcsm]

This was the one i downloaded, after install, it said successfully installed, but doesn't say restart pc which normally ask you that. not sure what was missing

[mojoe_90]

Did you right click and install as administrator ?  Silly I know but not doing this will cause the driver to not install correctly.   

[yeungcsm]

Yes I have, however if I wish to install again, it does say currently driver installed 17.30.xx. The reason i believe its not installed because of the poor performance and device manager still say Ms basic display driver. Rx vega was not showing at all

[mojoe_90]

If you can RMA the card if you have the same fault.     Mine was repaired under an RMA and it's been perfect since.     Still heats up my pc case by an extra 20 degrees but it hasn't missed a beat since being repaired. 

[yeungcsm]

Did they gave you a new card? Started rma online but need to call them

[mojoe_90]

No, they repaired the card.   Took about 3 weeks.

[yeungcsm]

Just filled in RMA, do you know what exactly they did to repair the card? as these crashes seems all random, previously i done like 6 stress test in a role, all OK, kind of worry they would not be able to replicate the issue

[neophyte]

Update # Card starting hanging on me so bad, got fed up and returned it. They tested it and confirmed it is faulty, replacing with a brand new one. When I go to collect it I'll speak to the techs who tested the card and find out exactly what's going on there, hopefully they can shed some light on this matter

[system2000]

i remember someone having issues with vega (not the same as the ones mentioned here) when his cpu was overclocked, at stock no issues, if you have done any cpu/memory overclocking try loading and saving default bios setting and give it a try

[jaibo]

Add me to the list, except mine is an MSI Vega 56 Air Boost not a Gigabyte card.

It's either fans ramp up 100% and display loses signal OR the BSOD with Thread Driver stuck in Loop. Both resulting in the need for a hard reboot.

I've read elsewhere (sorry, tried to find source but I've googled soooo much my history is a mess) that the thermal paste used on these cards is basically crap, but as somebody already pointed out in this thread, replacing the TIM would invalidate the warranty.

This has happened to me about 9-10 times since I bought the card some 3 months ago, including last night. I had a look to check the GPU power cables because I couldn't remember if I'd daisy-chained them, but I have 2 separate power cables from my 1050W PSU direct to card. 

I must admit I was surprised and concerned when I first installed my MSI card to see just ONE, TINY fan on the card. Especially since the 290X this replaced had 3!

The general consensus here then is that this IS a driver issue? No point in RMA'ing from what I've read here, but if this doesn't get sorted soon I'll be asking for an alternative card or some form of compensation...

Also, WHY is the thread marked "assumed answered"? Please, can a moderator change this... Or better still can we get some input from an AMD rep? 

[nixsar]

The only consensus I see on this thread is that for us who were lucky enough to solve the issue it was solved by RMA. Drivers are only a workaround as it is extremely sub-optimal to use very old drivers on GPUs bought specifically for higher end graphics. There is a lot of talk about overheating of various parts of these cards, but I see nobody providing any proof of that. I have tested my card myself and in a repair shop before RMA and we couldn't find any heating issues, thermal paste was changed as well. Since not all similar cards suffer from this issue it seems like a production problem with a certain series of components. We will probably never find out more unless somebody involved with making or repairing these speaks out.

[neophyte]

Hi guys, just so happens I got my brand new replacement card today along with the new posts.

I did ask for the technicalities of what was wrong, the only report on their system was it failed stress tests. No report on heat issues nothing which disappointed me as I'd love to have brought that news here if there was any. Looking at this new card its identical to the other one but the board is definitely thicker, it shows hynix memory now in GPUZ not samsung. Ran some 3dmark benchmarks and clock did not drop below 1550mhz and fluctuated slightly, where the other one was all over the place right down to 1330 at some points. Max temp I got was 76c at the VDDC VR and GPU core 64c running with Turbo mode selected in wattmann. I should've recorded the temps of old one feel stupid now that I didn't but anyway. The bios date on this card is 02 Jan 2018 whether that has something to do with it maybe like not one of the 1sts to have made their way out there.   

So to conclude this, thicker board + different memory? Like nixsar said we'll never know unless someone in the know how speaks out

[jaibo]

Hi, thanks for replying. However, I'm a bit surprised at your conclusion that the only fix is an RMA, when I've read a few replies here that state their returned cards (whether fixed or replaced) STILL had the same issue. 

I thought that was why many here were saying this was a software issue, especially since older drivers (installed by Windows) didn't have this problem.

Please understand, I'm not saying you're wrong, just that there are mixed messages to take from this entire thread!

Is RMA my best bet at sorting this then? If so, REALLY regretting having sold my older card already!

[nixsar]

Hi, I didn't mean that RMA is the only solution since, as you stated, some people apparently got replacements that also didn't work. I meant that the people who did solve their problem did so through RMA. My GPU also worked with the old drivers, but this is really not a solution to this problem, only a workaround. I have heard of no new drivers working with such GPUs. I could be wrong and maybe the people who wrote about it before solved their problem and don't follow this thread anymore.

[vendett3]

Sup! Although I'm not one of the first, I did try the drivers (and several combinations like old 2017 drivers and the new ones withouth Wattman) with no avail. I'm still waiting for my card to come back from RMA (a month, could you beleive it? I guess I should say thanks just for the fact that the card actually got accepted...) and only then I'll be able to give some more follow up and see if it the problem got solved.

[jdrobinson314]

The latest 19.5.1 drivers will gain a radeon pro skin if you have another radeon pro card installed in the system as well. There's that loss of several adrenaline variety features though.

This is a weird corner case, and I doubt that it will have any meaningful advantage with regard to the driver issue aspect of this discussion. 

[heartless_ca]

I sent my card in for an RMA awaiting on the return. It was accepted based on the issue type and the troubleshooting steps I took. Will update when I get it back. In the mean time I popped in my old gpu and have been running flawless for ~3weeks.

[yeungcsm]

I had same issue, but I changed to another graphic card still have crashes, not sure if underpowered psu also contributed to my issues

[jaibo]

Any more news anyone?

I just had the black screen/signal loss/ramped fans again...

I've just put a ticket in with MSI.

[neophyte]

I can confirm mine is completely sorted. Been running over a week not 1 single hiccup anywhere with new replacement, I thought I'd give it some time before reporting in

[jesusfreak5631]

I've been following the thread for the past two weeks. My Sapphire Pulse Vega 56 has had the same reboot issue since I purchased it during their promotion with Division 2 and World War Z. I've tried everything I could think of including upgrading to a 750W PSU, using separate PSU cables, switching PCIE ports, switching iGPU from "Auto" to "Off" in BIOS, re-seating GPU, using DDU uninstaller, reinstalling my entire game library, back-cycling drivers, Windows default drivers, using a different display, and finally a fresh Windows install with no avail. I even tested an older GPU and my Vega 11 graphics, both of which worked flawless (Of course). 

Here's my specs if anyone's interested. I'm convinced this is a hardware problem. 

MB: MSI B450 Tomahawk

CPU: Ryzen 5 2400g

CPU Cooler: Cooler Master Master Liquid ML120

GPU: Sapphire Pulse Vega 56

RAM: 2x8gb DDR4 Hyper Fury X 3466mhz

OS: Windows 10 64 bit

PSU: Corsair CX750M

Case: NZXT Tempest Evo

[yeungcsm]

mine MSI returned Friday, 2 crashes already, nothing was resolved

[jaibo]

This is what I'm afraid of, being weeks without a GPU and then the problem not even being resolved. I can't even check the status of my ticket, as the MSI site keeps throwing up an error when I try to view the ticket...

[system2000]

try disabling hdcp in radeon settings, display - specs - override - disable hdcp support, see if that stops the crashing

[jaibo]

Isn't that just for HDMI or DVI though? I connect to my monitor using DisplayPort only, (occasionally use HDMI to hook up to Smart TV)

[jdrobinson314]

Displayport will handle HDCP the same as HDMI.

[yeungcsm]

I will try that tonight when i back home.

Seeing so many people still have problem, can amdmatt‌ look into this?

[neophyte]

I wonder if there's a process of elimination for this problem. Like, what bios date does your guys cards have and memory type? Looking on Techpowerup it seems that the first bios's released for this card was 2017-07-30 (I may be wrong) mine shows 2018/01/02 (using AIDA64) and Hynix memory (other one Samsung) so this bios is just over 5 months older. 

I remember that other card locking up even at idle so throws a heat issue out, Nixar said he tested his at a repair shop and also showed no real signs of overheating anywhere. It's been exactly 3 weeks with this card and zero issues, I'm leaning towards either corrupt bios which conflicts with drivers - or possible memory type? 

[vendett3]

So, guys, quick and final update with my case. Last week I recived a message from my vendor saying that they recived a greenlight from Gigabyte for just straight replacing my GPU with another Vega. The thing was they didnt (and still dont) have any replacement on stock, so they gave me a RTX 2060 (here, at the local market, those gpu are more or less at the same price). Since I was sincerely pissed and tired, I accepted the tradeoff. I couldnt get any answers or clues of what happened, so my case will only serve now as an anecdote. I honestly hope every and single one of you get a response/solution and that my case had helped with something to someone. 

[yeungcsm]

Just tried the HDCP setting, still freeze, totally have enough of this

[jesusfreak5631]

Quick update. I gambled with a replacement Sapphire Pulse Vega 56, but I'm having the same issues unfortunately. I also tried this GPU with another PC, with the same problem occurring. I've been told by Sapphire support to continue with their warranty process. I'm relieved to see that my new PC build is not the problem. Best of luck to everyone else stuck in this mess.