R9 280X High Temps...

[PowerPie5000]

My new Asus R9 280X DirectCU II TOP has been fine the last couple of days, but tonight the temps have been getting quite high and a few games are artifacting! The highest previous temp for the GPU was 73C (it's usually around 69C), but tonight it's been running at around 82C and a few games are glitching/artifacting.

Absolutely nothing has been changed since installing the card a few days ago and i'm using the latest AMD drivers. The case also has good airflow and no dust. Anyone have any ideas what could be happening?

I really don't want to RMA another 280X GPU!

[kalniel]

You shouldn't get artefacts due to the temperature - the card should be throttling well before it can cause any damage/instability, but you can always lower the power tune limit yourself in CCC to make the throttling kick in earlier to test. I would suspect drivers or some software incompatibility elsewhere, unless asus have stuck a silly OC on there.

[PowerPie5000]

You shouldn't get artefacts due to the temperature - the card should be throttling well before it can cause any damage/instability, but you can always lower the power tune limit yourself in CCC to make the throttling kick in earlier to test. I would suspect drivers or some software incompatibility elsewhere, unless asus have stuck a silly OC on there.

It's factory oc'd to 1070MHz with 6400MHz memory (the memory chips are exposed with no heatsinks!). All the reviews of this card show load temps under 70C. I think i'll remove the card and inspect it at some point as i remember the 7950/7970 DirectCU models had an issue with loose coolers! I don't want a GPU that's going to be throttling itself all the time either... That makes the factory OC useless.

[PowerPie5000]

Asus are complete tools. I've now realised my R9 280X is suffering from the same cooler fault they had with their 7950/7970 DirectCU II cards! Why can't they learn from their past mistakes?

[kalniel]

Asus are complete tools. I've now realised my R9 280X is suffering from the same cooler fault they had with their 7950/7970 DirectCU II cards! Why can't they learn from their past mistakes?

Presume it's the same cooler given it's basically the same card underneath?

[sqwerty]

Nope they used the two slot GTX770 one here I believe.

[PowerPie5000]

The cooler seems firmly fitted despite the standoffs looking a bit uneven (not sure how flat it is against the GPU core though?). Anyway, i'm getting anything up to 83C when gaming and it hit 86C during a 15 minute Furmark test.... I personally think these are pretty poor results for a card that claims to be 20% cooler than reference. Kinda wish i'd gone for a 4GB GTX 770 instead now .

At least the glitching/artifacting seems to be under control (for now). I reseated the card, wiped all drivers and reinstalled them, removed GPU Tweak, cleared system registry and even reset the motherboards bios for good measure.

Nope they used the two slot GTX770 one here I believe.

It's the same cooler found on the Asus GTX 780 (2 different fans)... Minus the backplate.

[YoloSwag]

Thermal paste is fairly cheap.

But in any case, you should run the utilities that ASUS packaged with that card. Install one of those programs that show you current temperature, fan speed, core and memory clocks, etc.
Try to record what's going on when you run those games of yours. Check the fan speed. If it's not even budging when under load then you might want to override it a bit.

I had the same problem with my EVGA GTX 660 SC, it only has that blower type stock cooler. I noticed temps hitting 78C when I think it shouldn't. I was actually short on system fans though
but what I did was, did a software control for the fan, I moved the fan curve a bit near lower temps so it would be a bit heat sensitive.

The results were remarkable, I got about 5C lower on idle and about 5C-8C lower under load which resulted in a more consistent in-game performance.