A7V vs. KT7: CPU Temp differences....!

[thekingofpain]

I have almost the EXACT same results w/2 Birds at over 1g running side by side, KT7 registers consistant lower temps...(both with Superorbs, the faster one is the cooler one)

Hmmm, it is in my Clearcase tho, still unsure if the case material allows a certain ability to retain (or release) heat...

[This message has been edited by thekingofpain (edited 11-26-2000).]

[otheos]

I used the same case (Aopen mid tower) with enermax 430W dual fan PSU, and other than the different motherboards everything else was exactly the same. CPU is D700@900, 1 mem stick 128MB PC133 CL2, 2 HD's (1 Quantum 10GB LM+, 1 Fujitsu 10GB), 1 ASUS 40x CDROM.
The HS/F is a PAL6035 + Arctic silver.

In win98SE, I ran both setups for an hour running SETI@HOME.

KT7 said 41C
A7V said 49C

Both run @ 1.8V with 3.56V I/O.

Main difference is the A7V had the probe on the HS next to the core, the KT7 has the probe under the CPU (which I think is old fashioned since the cores are now on the top side of the chips).

So when KT7 users report their temps they read it off the bottom side of the chip which is cooler -> 41C only

A7V (1.01 rev only)reads from as near to the core as possible and hence gives 49C.

49C sounds more realistic, and this can be also told by the fact that at idle the KT7 read 28C and the A7V 36C. With 1.8V the A7V's temps sound more real to me.

Unless the A7V for some reason actually does run the CPU hotter than the KT7, Abit users should be carefull when overclocking.

BTW, I used BIOS 1003e2 for the A7V that enables the HLT commands (1004 and later don't and as a result of this idle temps are higher than 36C).

Now this test was not done under any kind of professional standards so its accuracy is within home standards. I tried to have the same ambient temp (with a thermometer) though.