Hello,
I had originally had the R9 Fury X Crossfire working like a dream for a few months. I being a seasoned Lighting Designer use computers to visualize complex lightshows and video content at and need an extremely high framerate. My computer began to fall apart the day was getting an error with one of the R9 Fury Xs. I decided to RMA the malfunctioning GPU in hopes to get my crossfire system back to normal.
After two weeks of waiting for the replacement device to arrive I finally get an Email saying that an R9 Fury X could not be manufactured at this time, and I would have to wait until a new shipment arrived. So another two weeks (over christmas) pass and I open a second Email saying that Sapphire would like to replace my R9 Fury X with a Watercooled Vega 64. Ecstatically I say YES, quickly sell off the 2nd R9 Fury X, and later I receive my new GPU.
Not even a day into using the brand new GPU my monitors suddenly turned to a solid color and sound would disintegrate; requiring a hard restart to get back online. I attached a picture of exactly what happens, given the screens are different hues sometimes. Naturally I assumed it was a faulty GPU considering I did not have this problem with my R9 Fury X, so I decided to RMA it once again. After two weeks I get a package with the second Vega 64, install it, and in under 24 hours of use the monitors glitch out again. Please note that during my 2 week Vega 64 hiatus I used my friends extra Nvidia 1050 and it had ZERO problems whatsoever.
Unfortunately Sapphire was not so willing to RMA it a 3rd time, claiming it was a problem with my system that was causing the crash. So naturally I took them up on their challenge, and decided to start buying brand new gear.
I purchased brand new,
ROG ZENITH EXTREME EATX TR4 Motherboard
AMD - Threadripper 1950X 3.4GHz 16-Core Processor
Corsair - H100i v2 70.7 CFM Liquid CPU Cooler
Thermaltake - The Tower 900 ATX Full Tower Case
G.SKILL TridentZ RGB Series 32GB (4 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Desktop Memory Model F4-3200C16Q-32GTZR
With much dissatisfaction I must suggest to you that I continue to have the exact same problem. Consequently, the only thing I have not purchased is a new Power Supply, I first need to test my friends PSU before I make that purchase. I've probably spent around $2,000 trying to fix this problem, and I have had to endure an insufferable amount of time and money lost due to this issue. It is unacceptable that while I am drafting stages in AutoCAD or Programming lightshows I must endure being weary of a random crash every 2-4 hours.
I have done a Clean Install of the drivers, and have updated them to version 18.3.4.
I used the Display Driver Uninstaller application recommended by AMD.
Additionally Windows 10 is up to date /w Version March 13, 2018—KB4088776 (OS Build 16299.309)
The RAM is of the DDR4 3200 Qualified Vendors List (QVL) for ROG Zenith Extreme.
Details: The crash can happen at any time and on any program.
Screens turn black and require a hard restart.
No Error Messages or Crash Dumps.
I used a single Nvidia 1050 on the same system for 3 days and not a single crash occurred.
I give speeches on Lighting Design in Vegas every year at Lighting Design International and I would hate to have to point my students and coworkers over to Nvidia. I was a huge fan of AMD until this started happening but as they say "Decisions make themselves eventually"... and there’s a push I'm feeling far away from AMD.
Has anyone ever had a problem similar to this? Or had any experience like this with Sapphire?
I hope one day they will come to their senses and realize they gave me two faulty GPUs.
-Kris
Sad to hear that. Try not to let it get you down. If you need any help on how to recover data from SSD you might be able to get some help from this forum. 
