Testing and Results
Testing was done in a spare computer I have in the room. Things were housed in a Corsair 600T SE and the door was on during any testing and reading of the temperatures. The room itself was kept at 25°C for the duration. To test the cards I used EVGA OC tool with a Furmark-like OpenGL 4.0 program. The card is controlled with MSI Afterburner for the overclocking and overvolting parts of the testing.
The testing is run for thirty minutes and temperature information is gathered from Afterburner for all the test results. In order to attain the idle temperatures, things were allowed to cool down and a reset was done and an additional ten minutes was waited to allow for the best possible results.
This is some terminology that will help with the reading of these charts.
Stock: GTX 470 with 608/837/1215 clocks and voltage of 1.00V
Overclocked: GTX 470 with clocks raised to 775/950/1550 and voltage raised to 1.050V for stability.
Stock Fan Profile: This is what the card is set at in BIOS level for fan control. It starts at 40% and ramps to 100% at ninety degrees.
Adjusted Fan Profile: See the image below. I changed the slope of the fan control
At idle, it doesn't really seem to matter much what the fan speeds or clocks were. I was able to keep up with and surpass the results I got with the original cooler. This isn't really all that surprising to me, as the body of the cooler hadn't changed that drastically.
Here the results are sort of hit and miss. I thought it may have been my using the found hardware, but a removal of the cooler showed me I had great contact. I really can't explain why this cooler reacted so much differently at load with overclocks or without. They don't seem to be up to par with the original cooler at all.
As for the noise levels, they are practically inaudible. While the cooler is in the case and the fans are running at full speed, I have to actually get my head inside the case, within 3" or so of the cooler to even hear them at all. Running them any lower can only make it better, and with the 40% or 1600RPM of the fans, I had to look to see if they were even running, as I could not hear them at all!