Hi guys,
I recently bought a Vapor-X 5870 1GB rev.2.0 card (used), and found out that only Sapphire TriXX allowed overvolting it because it has a non-reference PCB. And it works too.
My first question is, what are the maximum safe voltages for a 5870?
TriXX lets me go up to 1.3V from 1.125V.
My second question is this. I had a crappy PSU when I bought the 5870. It burned after a few days. Maybe it couldn't handle the extra load. When I had it, I used to overclock and overvolt the card. I managed to get it stable at 940MHz @ ~1.18V. Just for the fun of it, I increased the volts to 1.25V and tried increasing the clocks. I didn't really stress test the card using Furmark, because Crysis seemed to detect the instability before Furmark. (So did CODMW2). So I tried those games. After few minutes (sometimes few seconds), I would get black screen and the video card fan would rotate at full speed. The only thing I can do is hit the reset button on the case, and sometimes even that doesn't work. I have to cut the main power and turn back ON, to get the PC POSTing properly. Now that I have burst my crappy 550W Chinese made PSU, I bought a Corsair TX850. I haven't been able to test the card with this PSU because I had to go away. So my question is, could that black screen + fan at 100% issues have come from the PSU not giving enough juice or the video card not being able to handle the extra volts? (maybe not just the GPU having too much juice to cope with, but also the VRMs - which people say are crappier than the ones on rev.1.0 if the card - not being able to give a steady power to the card?)
P.S.
At 1.25V, I could even run 3DMark11 at 1GHz. (Not loop though) When I pushed it to 1.3V as I got excited, it gave the prob.
P.P.S
TBH, I wasn't 100% accurate when I said 1.25V wasn't stable in games. It was stable at first, but it gradually became unstable (I mean, the black screen thing) and I got the same prob even at 1.2V. That's why I dropped it to 1.18V and I didn't get the issue while gaming. It was pretty stable. Only thing is, it eventually burned the PSU. Maybe the PSU was the culprit all along.
Yes, I know, I should give it a try and figure it out myself because I have the goods, but without knowing the answer to the first question, I don't really like to try it out. Besides, someone might have been in the same boat as I.

Reply With Quote
