I am trying to get an idea here of when one should be concerned about installing VRM heatsinks on their video card(s). Here is what I got:
After I lost some of my last system to condensation from the vapochill, I decided to migrate all the surviving parts on over to my unused 4U rackmount case and build a "New" gaming system. In order to keep costs down I decided to stay a few generations behind in technology and just overclock the hell out of everything.
For my two 768MB GTX460s, that max stable overclock turned out to be 900Mhz core and 1050 on the RAM. With everything fine tuned I was able to dial in these clocks stable with a 1.087 core voltage on card one and 1.1 on card two. I am running the stock EVGA coolers with a few simple modifications and also a "secret ingredient" and OCCT load temps seem to be 61C for card 1 and 55C for card two. I actually got the first card to hit 930Mhz on 1.087 which surprised the hell out of me, but when the second card still proved unstable at 910Mhz with a hefty 1.125v core I decided to abandon hopes of anything above 900Mhz and didn't push the first card any further.
Given the fact that I believe these are excellent temperatures for the overclock involved on the stock cooling I am left with the last concern of the VRMs. I don't want everything to be fine and dandy for a month or two then be in the middle of Crysis 3 and end up blowing a VRM off the board.
Would everyone here be confident in saying that the voltage mods warrant VRM sinks?
Bookmarks