rx480 Crashes/Black Screen (4 months, not a solution!)



First of all:

System Information


Time of this report: 2/8/2017, 14:17:44

Machine name: DESKTOP-TJFB6VV

Machine Id: {0C5C80DC-8BF0-4159-B4A0-D780D0430216}

Operating System: Windows 10 Home 64-bit (10.0, Build 14393) (14393.rs1_release.161220-1747)

Language: Italian (Regional Setting: Italian)

System Manufacturer: Gigabyte Technology Co., Ltd.

System Model: GA-970A-UD3

BIOS: Award Modular BIOS v6.00PG

Processor: AMD Phenom(tm) II X4 965 Processor (4 CPUs), ~3.4GHz

Memory: 8192MB RAM

Available OS Memory: 8190MB RAM

Page File: 4495MB used, 5614MB available

Windows Dir: C:\Windows

DirectX Version: DirectX 12

DX Setup Parameters: Not found

User DPI Setting: Using System DPI

System DPI Setting: 96 DPI (100 percent)

DWM DPI Scaling: Disabled

Miracast: Available, with HDCP

Microsoft Graphics Hybrid: Not Supported

DxDiag Version: 10.00.14393.0000 64bit Unicode

PSU Corsair GS700


Now my problem, briefly (apologizes for my english, i am not mothertongue):


I bought the new rx480 in september 2016, mounted it in my desktop and i observed a very weird thing: my monitor looses the video signal (grey screen with the flashing Digital/Analog message), audio goes on normally (sometimes i get a RRRRRRRRRRRRRRRRRRRR sound), the pc is anyway stuck (impossible to give commands, it doesnt turn off either using commands or the "turn on" key, need to reboot using the reset key, no other way).


This crash happens randomly, maybe after 2 days, but once it happens the first time, it happen more and more frequently (even after 5 minutes). I noticed the graphic card activity in wattman is really inconsistent (it CONSTANTLY goes from 0 to 100 all the time). Sometimes (but not every time) after rebooting i get the message (translated from italian) "the radeon has been reset because of a power failure" (more or less ^^).


After many fixes, now I do NOT find in windows registry any "event" at the moment of the crash or close (the only thing i get is the power-kernel error because of the forced reset and reboot)


I tried everything i red on the net, ie (by hearth):

- format C:, fresh reinstall new win10

- DDU, reinstall latest Crimson driver

- Scannow, registry cleaner and all other commands to check integrity are ok, nothing relevant detected

- TDR Delay

- everything i got from AMD support forums

- upgrade/downgrade to every Crimson i could find (from 17.1 down to 16.6)

- Bios updated

- no overlock

- test temperatures: nothing goes over 74C, never

- unplug and re-plug RAM and graphic card (reinstalling deivers)

- use wattman to increase or decrease power and voltage

- use wattman to use fixed voltages (many different tries) for different stages

- use wattman to reduce frequency

- lowered all graphic in the only game i use (blackdesert online) - i honestly do not want to blame the game as it runs (on lowest settings) on my laptop that has integrated video card

- after the first crash, once it also happened in safe mode, many times when booting or when surfing the net

- unplug CD reader, secondary HD, printer and everything could draw power

- using a UPS for both desktop and monitor



After a couple of months I brought the desktop to a technician, but he could not fix anything after many tries (crashes are so randomly and they often do appear on fresh installation/driver update only after a couple of days, so it is really hard to find the "cause"). The PC is still at the technical assistance.


Now, after 4 months of problems, sincerly, i would like to ask for a refund but the vendor (an italian online shop) told me they could need 60 days to give an answer after i send them the card and this really pisses me off.


Is there something i can still try? (i cannot personally test the single hardware components on others PC, i have not another desktop available)

Is it possible its a faulty card?

Any fast way i can check if it has any problems?