I'm running a Sapphire Tri-X OC R9 290 (the older reference PCB version). I haven't read the thread as thoroughly as I might have but the display corruption looks exactly like it would if the memory was overclocked too far. So I'm assuming it's something to do with idle memory clocks or voltage.
Peter, do you have a Hawaii card to test? It seems like most users who have reported a similar problem in this thread are running 290 or 290X, both single and multi-gpu. Just from this page alone, KaRLiToS has 4x 290X, OafBoaster has 1x 290, gregr507 has 1x 290X, and I believe digitaldurandal has 2x 290.
Hi 96Firebird and Spanners,
Thank-you for the feedback!
That is some really good insight. From the looks of this thread, it does look like it is happening mostly to end users running R9 290 and R9 290x since I was not able to replicate the issue with R9 285 cards and R7 260x cards.
I'm going to have my technical support team (customer service team) review their RMAs and customer service logs to see what they can find for the past 6 to 8 months since OP said this started happening with the June 15.7 drivers.
Most of our customer base are graphic designers, engineers, medical, IT, military and the financial industry with plenty of end users that use multi-monitor setups of our 2560x1440 monitors, so we should have some data for for that time frame.
I did had some Hawaii cards on loan for the first half of last year but returned them. I'm going to try to get one to run some more tests.