Quote:
|
Originally Posted by unregistered
Swiftech's posted data is all recent, comparable, and utilizes the same heat source; without question the present data is more accurate than the previous set, as the next will be improved yet again
|
If referring to an Intel TTV, I disagree.
The temperature of the heat-die is not known, the size of the heat die is not known, nor is the efficacy of the heat-die to IHS contact as a variable of waterblock base->IHS interaction known either.
IMO, the Intel TTV isn't too far from measuring the internal temperature of an oven by a thermometer on the outside. I know that Intel proposes it as a
heatsink validation mechanism, but even Intel states that once a heatsink has been validated on a TTV that it must also be qualified by taking actual CPU temperatures in a real system.
IMO, if one wants to simulate an IHS based scenario in a fixed and repeatable fashion, then the heat die and the IHS need to be a one-piece affair, thereby the die->IHS variable is removed and can be corrected by applying a correction factor based upon observed TIM behavior, but applying such a correction factor would only be for validation purposes.