And the CGK Results are in!

[yo2tup]

a7v, tbird 750@1050mhz
artic silver
i took temp reading by using the thermal probe that came with the a7v. i cut of the plastic at the end of it so the metal thermal diode tip is exposed. i then placed the probe at the side of the processor so that the metal tip is actually touching the the side of the tbird core.


alpha pal6035
idle 38C, load 53C

and the moment you've all been waiting for.....
*drum roll*

Taisol CGK
idle 33C, full load 48C


not bad



damn that thing was a b!tch to install, i'm not taking that thing off. anyone want a pal6035? send me a private msg if you're interested. and it's very easy to install

 

[8008S]

Buh...I just ordered a CGK. I don't wanna fool around with another god damn clip again

 

[Mikewarrior2]

Hey, is your pal6035 equipped with a Socket A clip?


Mike

 

[yo2tup]

no, it's not. but i just retensioned it to hold against the processor tighter, and it works fine.

 

[thermite88]

yo2tup,

Your result of the two HSF comparison seems strange to me.

The temperature difference are 5 degrees C both at idle and full load. It is possible only if the thermal resistance DegC/W is changing with the heat dissipation level.

For example, assume that the CPU runs at 50 Watts at full load and 15 Watts at idle. (Yes, I know that the 30% load at idle seems high, but I heard that AMD's implementation of the HALT command is not very efficient.) If the temperature difference is 5 deg at full load, the difference should be 1.5 deg at idle, using the same thermal resistance values. The temperature rises vs. heat is supposed to be fairly linear for individual heatsink/fan combination.

Your result seems to support Mikewarrior2's argument that most heatsink measurements were done incorrectly.

 

[yo2tup]

maybe *shrug*..

all i know is that i am now able to overclock my processor 70mhz more, so i'm content

 

[Tetsuo316]

how long did you wait after starting up before you took the readings? that might have affected the results...

whether or not the results are "exact" or "absolute" doesn't really matter. I think that it's safe to assume that the cgk outperforms the alpha (is the alpha forged - i don't know?), and the proof is that he was able to gain a few more megahertz by simply switching the hs/fan.

on a side note, i thought that the three-prong clip for the cgk was supposed to be easier to work with than the old one-prong clasps...