GTX480 EVGA Hydro Copper O/C & Temp Test results

[gabe]

Thanks! Just wanna see if my card can stand up to a more "industrial standard" test as opposed to 10~ minutes of Kombustor, and it seems to be able to. And the "So far our 890 is the highest on record on water, isn't it?" and the following forum posts felt like a challenge.

I do see the Swiftech are better at GPU cooling then the EK full cover. However they are not available to purchase at least on this side of the pond.

Curious, is it possible to point an infra-red thermometer at the VRM section of the Swiftech? Wondering how well the heatpipe works to cool them.

VRM's are only relevant to OC in as much as they dictate an absolute operating limit. Let me give you an example: say the card manufacturer has a spec of 120C for their VRM's. They would hardcode the card to shutdown if the VRM's exceed this temp to prevent damage. Now, if you are running under this maximum spec, then the limiting factor for your O/C is the GPU, not the VRM's.

In other words, as long as VRM temps remain under their specified limit, they have no effect whatsoever on O/C. As to the card's life, we know (*) that component life is doubled for every drop in 10C. So we want to keep these component to run lower than stock in order to improve the shelf life of our cards, right? So let's calculate this. Say the stock temp is in the high 90's. I don't know what the shelf life of the component is, but let's say for the sake of argument that it's at least the same as the manufacturer warranty, which is what? 3 years? So let's do the math: if we drop this component temp down to 80C, we are at 6 years shelf life, down to 70C we are at 12 years, etc..

(*) drop in 10C is either equal to ~2% frequency increase, or ~double the life at same frequency