cancel
Showing results for 
Search instead for 
Did you mean: 

PC Graphics

ASRock Phantom 6800 - Causing Hard System Crashes w/Mobo VGA Error Light

All - wondering if I should RMA my new ASRock Phantom 6800, or if other users having the same problems.

TL:DR - At random times during gaming, system crashes hard, requiring a power down.  Motherboard error light shows VGA and won't post.   Does not happen with Doom Eternal (Vulkan API).  Seems to happen often with Destiny 2 & Apex Legends.   Freesync disabled seems to increase stability.                                               

System:

Windows 10.0.19042 Build 19042

Asus X570 Tuf Gaming Plus (Wifi).     

32 GB of TridentZ Neo. (temporarily running 16 GB of Corsair Vengeance)  Ran at XMP and default JEDEC speeds.

Ryzen 3800X

BIOS version 3001.  AGESA Version "ComboV2_1180". 

Running OS and games from M.2 Samsung 970 Pro. 

(Radeon version 20.12.1)

Primary Monitor on Displayport 1440p/144hz: LG 32GK650F-B.  Secondary monitor is 1080p/60 Asus on HDMI.

Troubleshooting:

I've been troubleshooting this card for the last ~4 weeks. I've put in a new PSU thinking my 3 month old Seasonic had a problem.  It did not.  Formerly running a Seasonic 750 Watt Focus Gold, now running a Seasonic Focus Gold 850 watt. 

Fresh Windows Install on brand new SSD.

Formerly tried multiple DDU passes, but have since removed the old hard drive and Windows so that's not relevant. 

Swapped back to my old 16GB Corsair RAM in case I had a RAM problem that appeared to be GPU related.  RAM does not seem to be  factor at all. 

Crash description:

When crashes occur, they seem totally random.  It can happen in first few minutes of the game, or I can play for ~ hour or and then it will crash.  Upon crashing, both screens go black, Mobo goes through error checking startup sequence, and the light for "VGA" stays lit. Must power down system after crash. 

Some stability was achieved in Apex Legends by disabling Freesync on my monitor and in Radeon settings. 

In 3DMark, Timespy will complete about 80% of the time.  

GPU Performance metrics show cool running temps (62° GPU/ 72° Junction is absolute highest ever seen).  Boost and base behavior is exactly in line with AMD claims. 

There are Zero artifacts or other indication the GPU is having any trouble.  It works perfectly until the crash occurs.

Stable games:

MS Flight Sim, Teardown, and Doom Eternal run great.

At this point, I'm pretty sure the GPU has a hardware problem. However, given the stability in Doom Eternal, I'm wondering if perhaps this could be driver related.  I've filled out numerous AMD bug reports to provide as much detail as possible. 

Curious if anybody else has experienced this, or may have some additional troubleshooting steps I can try. 

 

0 Likes
8 Replies

A few additional notes:

-Was formerly running an RX 480.   XMP RAM, manual timing adjustments, and a mild Clock Tuner for Ryzen OC Profile.  

Stability was perfect.  

All of this removed now. 

In the course of troubleshooting, I've replaced PSU, PSU cables, and also added extra 4-Pin CPU power.  Figured I might as well run the cable when I swapped PSU's.

At times, when the "hard system crash occurs", the GPU fans will spike to 100% and get stuck there until power down.  I'd say this happens 1/10 crashes. 

I always monitor my GPU temps, and they never get anywhere close to concern.  I once saw junction temp reach 72.  

 

0 Likes
mackbolan777
Forerunner

Ok, no longer the same thread. Running 2 monitors of different resolutions/refresh rates might be an issue. Try using one and see if anything changes. Like keep the better 1440p connected and disconnect the secondary for a test. I never ran 2 unlike monitors before, always both the same size, refresh rate and resolution.

"It worked before you broke it!"
0 Likes

Well... testing plans got dashed tonight.  Stock alerts for Newegg got me distracted and I somehow managed to score a Gigabyte Windforce 6800.  Haha... still planning to do more testing on the Asrock, so I'll post info when I know more. 

 

Side note - I never installed any ASRock utilities for tuning/LED control, etc. 

0 Likes

That's the exact card I have. So in theory my settings should copy right over and this card screams. Just return the AsRock and be done. 

"It worked before you broke it!"
0 Likes

Interesting follow up note here - after the most recent patch, Forza Horizon 4 ran just fine - with Freesync, two monitors, etc.  Formerly, it had caused crashing at random times after loading in to game -usually within 5 minutes.

 

However, Cities: Skylines will induce a hard crash instantly when Freesync is on.  100% reliably too. Load the game, even a blank city, and it will crash within ~20 seconds, requiring a full system power down and VGA Error light on motherboard comes on. 

With Freesync off - I was able to load Cities: Skylines and it seemed fine.  Granted, I didn't play very long. 

0 Likes

I have the same issue, it just freezes. Everything seemed fine for two days, but after the latest driver update PC just freezes. With RX 5700 XT eveything was stable, so I don't if it's a new GPU or something else. I will use it more, try to rollback drivers if I will see any more crashes and if that will not help, I will put my RX 5700 XT and will see if it's happening too.

0 Likes

Try the 20.12.1 driver. I'm using that with zero problems.

"It worked before you broke it!"
0 Likes
ArmGadge1
Adept II

Hey, it seems we have some common problem, ASRock card, black screen and hard crash.

Could you please try scrolling twitter timeline like this?


FYI my thread is here:
https://community.amd.com/t5/graphics/rx-6800-will-crash-when-playing-rust-with-browsing-twitter-on/...

How about if Max. GPU clock set to the game clock manually?

I'll try Apex later.

0 Likes