• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Ongoing 290x issues - need advice

Associate
Joined
26 Oct 2011
Posts
180
Location
Ireland
Let's just get straight to it - I've had a string of issues with my 290x, and I'm passed the point of directing the blame at the card and *have* to consider other possibilities.

Model in question: asus dcuii oc

1st 290x: dec 2013, chain black screens, couldn't run any games. Replaced.
2nd 290x: jan 2014 - feb 2015. Occasional black screens, but tolerable. Fan bearings died after a year of intermittent use (ongoing health issues - that's another story). Replaced after fan death.
3rd 290x: march 2015. Dead within an hour of installation. Chain black screens, display corruption, eventually wouldn't even make it to desktop.
4th 290x: received just this Friday. Encountered 2-3 black screens so far. Odd screen artifacts occurring intermittently. Example: www.youtube.com/watch?v=lh6Q_oCtSRM (ew ff14 I KNOW, I have some RL mates playing, said I'd tag along when I'm not in a hospital bed, heh). Happens every 15-30 seconds, but no signs of crashing while ingame... yet. You'll see the flicker @ 6-8 seconds.

Rest of my system:

4930k (tested both stock and various levels of OC)
32gb ddr3 (2133mhz) (have run memtests, have tried at lower speeds)
asus x79-deluxe mobo
xonar essence stx sound card
corsair hx1050 psu (yes it's quite old by now, but it's always been reliable, voltages have always been consistent)
win 7

Final note: My system presents NO issues when using my mate's backup card (270x).


Thoughts? Am I just that unlucky, or am I missing something blindingly obvious (it's a distinct possibility, I'm so loaded up on medications atm that I won't even handle my pc's internals, had to leave it to someone else)
 
To add: GPU temps are well within tolerable range (~70C under load). Hardware settings remain @ factory defaults.
 
Let's just get straight to it - I've had a string of issues with my 290x, and I'm passed the point of directing the blame at the card and *have* to consider other possibilities.

Model in question: asus dcuii oc

1st 290x: dec 2013, chain black screens, couldn't run any games. Replaced.
2nd 290x: jan 2014 - feb 2015. Occasional black screens, but tolerable. Fan bearings died after a year of intermittent use (ongoing health issues - that's another story). Replaced after fan death.
3rd 290x: march 2015. Dead within an hour of installation. Chain black screens, display corruption, eventually wouldn't even make it to desktop.
4th 290x: received just this Friday. Encountered 2-3 black screens so far. Odd screen artifacts occurring intermittently. Example: www.youtube.com/watch?v=lh6Q_oCtSRM (ew ff14 I KNOW, I have some RL mates playing, said I'd tag along when I'm not in a hospital bed, heh). Happens every 15-30 seconds, but no signs of crashing while ingame... yet. You'll see the flicker @ 6-8 seconds.

Rest of my system:

4930k (tested both stock and various levels of OC)
32gb ddr3 (2133mhz) (have run memtests, have tried at lower speeds)
asus x79-deluxe mobo
xonar essence stx sound card
corsair hx1050 psu (yes it's quite old by now, but it's always been reliable, voltages have always been consistent)
win 7

Final note: My system presents NO issues when using my mate's backup card (270x).


Thoughts? Am I just that unlucky, or am I missing something blindingly obvious (it's a distinct possibility, I'm so loaded up on medications atm that I won't even handle my pc's internals, had to leave it to someone else)

That truly is a bizarre state of affairs, to have such bad luck with so many 290x's. Even tho you say your mates 270 exhibits none of these issues, I still am inclined to say for so many 290x's to fail something else must be the issue.

How long did you use the 270 for?

Each time the 290x went back as faulty, no doubt it was tested. What was the diagnosis?
 
The 270x saw the same intermittent use over the last 6 weeks or so while the 290x('s) were being replaced, maybe ~25-30hrs a week? If I'm not in hospital, I'm at my pc currently.

As for the diagnosis, none was ever given to me - I posted the issue, they gave an rma, I sent it back, it went to asus, and asus issued a replacement. So I can assume they either found an issue, or they didn't bother to test it and just replaced.

I'm currently arranging to have this 290x tested in my mate's machine in place of his 970, we'll see how that goes. But that will take a few days sadly.
 
Might be something funny going on at the RMA centre, I wouldn't put any faith in the process. Best thing you can do is refund it and buy a different brand, the Asus 290/Xs are pieces of **** anyway. I assume you did the obvious thing of a fresh Windows install.
 
Yep, only replaced my ssd with a larger one recently, so it was a clean install.

As for refunds, I believe after 16 months I'm well beyond the point of a retailers refund and I'm more reliant on the manufacturers warranty. I've tried getting in touch with Asus support directly, but after a form mail that recommended a driver from Nov '13 (because old and drivers do oh so well with newer games!), a second that recommended a gpu bios update to an older bios, to a 3rd form mail that was a copy/paste of the 1st, I simply gave up there.
 
Very odd issue. Have you tried your card in another PCI slot and see what happens? I know you said the 270 worked but maybe something with the 290x/PCI combo...?
 
Damned phone auto correct! Anyway...

Using the second pcie slot yielded the same result.

As I said above, my next test will be trying it in a completely different machine, though it will take a few days to organise.
 
290s are quite sensitive it seems. Maybe check your ram (Memtest), and maybe see if you can borrow a newer PSU.

That being said I've seen similar figures myself, out of the six 290s ive owned or built systems for others for, Five had major issues. Black screening on four of them and artifacting one the 5th.
 
So, it took a few days to get back in touch with him, no 3g signal in this hospital and all that.

He says: In games, particularly high mem usage games, there's some display corruption/flickering. But no crashes. According to him, the biggest concern he has is that if he goes web browsing after a cold boot instead of straight into a game, there's about a 4/5 chance that it'll black screen and lock up, but once in-game, it's stable, albeit with the display issues.

So that's where we stand now.
 
So, it took a few days to get back in touch with him, no 3g signal in this hospital and all that.

He says: In games, particularly high mem usage games, there's some display corruption/flickering. But no crashes. According to him, the biggest concern he has is that if he goes web browsing after a cold boot instead of straight into a game, there's about a 4/5 chance that it'll black screen and lock up, but once in-game, it's stable, albeit with the display issues.

So that's where we stand now.
I have looked up the ASUS R9 290X DC II OC, and comparing to 290x that has a stock memory clock of 1250MHz, it has a factory overclock of 1350MHz.

Those symptoms you mentioned (black screens, flicker then lock up) are classic case of the unstable memory clock on the 290/290x. Sounds to me that Asus has used Elpida memory and applied the +100MHz overclock on the memory clock without properly testing for stability.

If you are not too fussed about the factory overclock on the memory and just want a stable system, changing the memory clock would most likely solve these issues right away. The 100MHz on the memory clock would most likely make as little as 1fps in games anyway.

If you still got problem with the memory clock running at 1250MHz, then I'm afraid you'd probably have to rma again.
 
Have you tried upping the voltage; also are the drivers on each one of these card been clean installed??

Drivers are always clean installed, in fact it was a clean windows install with the most recent card.

As for voltage, it's not something I've been too keen on upping with the 290x's, they're far from the most stable cards I've encountered :P

I have looked up the ASUS R9 290X DC II OC, and comparing to 290x that has a stock memory clock of 1250MHz, it has a factory overclock of 1350MHz.

Those symptoms you mentioned (black screens, flicker then lock up) are classic case of the unstable memory clock on the 290/290x. Sounds to me that Asus has used Elpida memory and applied the +100MHz overclock on the memory clock without properly testing for stability.

If you are not too fussed about the factory overclock on the memory and just want a stable system, changing the memory clock would most likely solve these issues right away. The 100MHz on the memory clock would most likely make as little as 1fps in games anyway.

If you still got problem with the memory clock running at 1250MHz, then I'm afraid you'd probably have to rma again.

This has been my train of thought on it aswell, the lad I have running it in his system normally uses the MSI 8GB edition of the 290x and has had no issues at all.

That said, black screening *only* occurs after booting, never once it's in-game. Specifically, it occurs when going from the card's idle clocks (300/150) to its standard clocks (factory 1050/1350). Once at normal clocks it's (relatively) fine.

So the question becomes will changing the memclock from 1350 -> 1250 fix the black screening? It may fix the on-screen issues at least I suppose.
 
I have looked up the ASUS R9 290X DC II OC, and comparing to 290x that has a stock memory clock of 1250MHz, it has a factory overclock of 1350MHz.

Those symptoms you mentioned (black screens, flicker then lock up) are classic case of the unstable memory clock on the 290/290x. Sounds to me that Asus has used Elpida memory and applied the +100MHz overclock on the memory clock without properly testing for stability.

If you are not too fussed about the factory overclock on the memory and just want a stable system, changing the memory clock would most likely solve these issues right away. The 100MHz on the memory clock would most likely make as little as 1fps in games anyway.

If you still got problem with the memory clock running at 1250MHz, then I'm afraid you'd probably have to rma again.

Spot on.
 
Asus has to be one of the most frustrating companies I've ever experienced. I have Maximus IV Gene Z motherboard and it's the first Asus motherboard I've owned and to sum it up in one word would be 'fantastic', the UEFI/BIOS is excellent the best I've ever used. However I repeatedly read threads and stories like this where either their quality control is called into question (lots of complaints on the ROG forums regarding the SWIFT) and if it's not quality control it's there customer service and RMA experiences.
 
Just to note: I'm fine with dropping the memory clock from factory settings down to stock 290x settings.

That said - Things should function as advertised/sold :P

Once I'm through this next round of hospital visits I'm tossing this card out the window anyway - any recommendations on a replacement (or replacements - there's not a budget per se)?
 
Just to note: I'm fine with dropping the memory clock from factory settings down to stock 290x settings.

That said - Things should function as advertised/sold :P

Once I'm through this next round of hospital visits I'm tossing this card out the window anyway - any recommendations on a replacement (or replacements - there's not a budget per se)?
lol 100% agree with you that thing should work as it is advertised/sold, but on the flip side...if I can just trade off losing may be 1-2 fps at most in games in exchange for complete stability, I would gladly do it, and save pulling more of my hair out :p

I'd probably draw the line at if it can remain stable at AMD's stock settings or not...if it can't, then I would certainly throw that card right back at the manufacturer on RMA :D
 
Won't hurt to test; and shows that either memory is too high at that voltage....:) if you want to run memory higher most likely needs more voltages......dropping it down 100 will easily test it :)

Don't you just love our hobby :D
 
Back
Top Bottom