The card is auto overvolting for some reason...., this started to happen few weeks ago, never had the issue before.
The vddc is at 0.95v at idle, and 1.1625v at load.
I noticeed that working at 2d, the fan noise of the card was loud, so I went to check the temps... at 157mhz is when its idle, the temps were around 67 celsius....67 celsius clocked at 157mhz????, and so I checked the VDDC voltages, when I saw the 1,65v at VDDC I thought to myself "WTF".
Is there any good reason why I should leave it be, or should I be concerned and RMA it?
(no overclock was done, nor tried any overclocking tool either, everything is at stock and always have been, I only overclocked the cpu, witch I can't see a reason why should be involved at auto overvolting)
Yes, twice already, from 10.2 to 10.3 to 10.4 catalyst, done properly ofc. No improvement, ofc, it happens when time pass, is not instant. I noticed that if i keep rebooting the pc after some time, the VDDC is withing 1.1625v, so maybe it's a memory usage issue? dunno :S
Yesterday I opened a ticked in sapphire website(the card is sapphire), and yet I haven't been answered.
While everything looks good, no artifacts whatsoever, and the performance fps wise is perfect, I'm considering RMAing it, because the card is not supposed to overvolt right?
It definitely should not, and assuming it is (and it is not just a faulty sensor, however since the temp goes up I'm inclined to believe it) getting that high, it could significantly reduce the life of the card. Not to mention it is wasting a ton of power (on the order of 50-100W perhaps). Definitely try to RMA the card, as this sounds like a hardware/GPU BIOS issue. Are all other system voltages normal?
Yes, all the voltages except the gpu one's are fine, or so does report everest.
I thought a misread too, but since the temps go up like it were to be on 3d load most likely is not.
I'm thinking the PSU, but the spare that I own is broken.. I don't know.. I'll check with my old gpu card and see if it overvolts... but I doubt it, never had such weird issue before.
If the old card does not overvolt, I'll RMA it, thanks for the replys
I'm checking the ticket every now an then, so far nothing.
Also, I've been checking the gpu-z log more thoroughly, and the memory temps are almost hitting 90c, also the VDDCI reports high voltages too... if you want to take a look, I could upload the txt to megaupload or something.
Yeah, I understand what they are thinking, however the fact that it is correct and cool at first and then wrong and hot later makes me think it is not a GPU-Z bug. Just to check, do you have any other programs installed that can read the GPU voltages? (beyond GPU-Z and I guess CCC, though it doesn't report them). If so, try uninstalling them and see if the problem still occurs.
Only Everest, I'll try uninstalling it and see what happens.
Weirdest thing though, is that doing research, I found a lot of ppl trying to manually overvolt their hd 5870 that are failing to do so, and my card is doing it on it's own ¬¬
PS: Sorry for the English of mine, sometimes I forget things, as it's not my native language.
Oh, and the reason for the suggestion is that MSI said in their afterburner program that you should not have more than one GPU voltage reading app open at a time. I have no idea why they said that or if there are any side effects, but worth a try.
Indeed, every ideas are welcome,I tried uninstalling Everest and Speedfan,(although I don't use this one)so far the voltage readings are fine along with temps and clock frequencies of CCC.(I just tried that, so I'll give it some time).
I'm considering removing every program up to the date that the gpu-z log states voltage issues. Can't see how could affect though.
I'll report whatever the tech support of Sapphire report back to me, hopefully a new bios for the card will fix this.
Well, that was unexpected, so far so good!, card temps are fine, gpu-z shows no overvolt.(crossing fingers for it to not happen again)
Looks like a good advice there ext64. Now time to test it under some 3d stress. I'll get back to you on that, thanks for the help ; )
Yeah, unfortunately for this sort of problem, it could randomly come back (could just be hiding). If it goes for for a few days we might be on to something.
Hi,i also had this same problem on my ati 5850,volts shooting up to 1.65.i was running everest which was reading my gpu vrm voltage and also had voltage monitoring running in msi afterburner,i unchecked voltage monitoring in msi afterburner and everything been ok since.