Hi all – hoping you might be able to advise me on whether I’ve broken one of my graphics cards through stupidity! Hope this is the right forum to put this is.
System: i7 3770k, ROG Maximus V Formula, 2 x ROG Matrix 7970 Platinum, Windows 10 Pro.
Getting a bit old now, but it’s been performing fine up until I drained, cleaned and refilled my watercooling loop the other day. Since then, I’ve been unable to get the secondary graphics card working, but I’m not sure whether I’ve damaged it or whether something else is amiss.
Symptoms:
1) When I install catalyst/‘radeon crimson’ from fresh, the computer jams solid when it goes to a black screen for the second time during installation of the graphics driver.
2) If I boot the computer with both cards enabled and drivers installed for both, Windows jams on a black screen between the first Windows logo and the log-on screen.
3) If I boot the computer with both cards enabled but with drivers only installed on the primary card, I get the stupid Windows 10 unsmiley face BSOD about 30 seconds after logging on, with the error ‘thread stuck in device driver’. Presumably this happens when Windows tries to install the driver onto the secondary card.
4) If I boot with only one card enabled (having disabled the other through safe mode), everything is fine. Similarly, if I try to install catalyst with the secondary card disabled (by removing the auxiliary 8-pin PCI-E leads), everything is fine.
5) GPU-Z can see the second card if it has the driver installed but is disabled in device manager. Several bits are missing though (e.g. the amount of memory), but I don’t know if that’s normal for GPU-Z looking at a disabled card. GPU-Z doesn’t work in safe-mode, which is the only time the card can be both installed and enabled without the whole thing locking up.
I’ve tried:
1) Numerous driver installs (using the Display Driver Uninstaller utility in safe mode to clean up first) using both the catalyst installer and device manager.
2) Several different drivers (including both the most recent driver and an ancient one that came on disk with the cards)
3) Reseating the card, replacing the crossfire bridge, and/or swapping the 8-pin PCI-E power leads around.
4) A clean install of Windows 10.
The only things I can think of that I haven’t tried are a different motherboard/different cards, or physically swapping the cards around (they are currently the other way round than they were before I removed them for maintenance). I don’t want to try those unless I have to, though, because I really don’t want to re-drain, re-fill and re-leak test the whole thing if I’ve fried the card.
As far as I know I treated both cards carefully while doing the cooling maintenance – nothing leaked on them either before or after, and they were stored safely while I cleaned the rest. I’ve been doing watercooling for years and this is the first time something like this has happened. That said, I’m running out of ideas of what could be wrong, other than that I’ve somehow damaged it. Any ideas/thoughts would be most appreciated! If any crash dumps, logs etc. would be useful, I'll post them.