> (AGP/PCI-E, PCI-X, all onboard sound/video/sata/pata/network,
> etc) connected thru CPU0. The only thing CPU1 (zero based, so the second
> CPU) has hanging off it is the memory belonging to its memory controller.
> That's the way my Tyan s2885 is laid out.
>
> > I run irqbalance, and irq's are
> My Dual Opteron comp usually has a CPU 1 temp that is 15 degrees F. less
> then CPU2, looking and GKRellm for a while though I tend to believe it's
This appears to be contrary to what others see! What is your motherboard, and
case? How do you heatsink it? What is the OS (Gentoo AMD64, I presu
> Int he Opteron server which is a rack there is only a 1.5C difference
> between both CPUs (one 43, one 41.5).
This is interesting, what is your motherboard and case? How did you measure the
temperatures? What is the cooling system?
Jason.
--
gentoo-amd64@gentoo.org mailing list
Others have posted that they see different CPU temperatures with different
motherboards. This leads me to believe that this is an artefact of the Linux
kernel causing the different temperatures through running CPU0 harder. Previous
emails have stated that CPU0 for various motherboards always giv
> > If we are all using the smae motherboard, then this may be possible. But if
> > they are different, it is morelikely to be a feature of the kernel. I run
> > distributed.net, so both CPUs are at 100%.
> loaded, but I think opterons do not 'power down' when not under load.
I am pretty sure tha
> > i wanted to ask if its normal that gkrellm shows about 45 degrees for
> > CPU0 and 35 degrees
> > for - i admit not strong working/used - CPU1.
>
> And I've seen diffs of 41 C and 36 C. I'd guess bad calibration
> of bad sensor. And it could be air flow - are there cables - like
> the power
