On Thu, 03 Feb 2000, Edward Schernau wrote:
> I see a significant temp difference, esp. in Windows95.
>
> Win95 temps: CPU1 29, CPU2 39, system 38
> Linux temps (running 2 RC5 clients) CPU1 35, CPU2 40, system 39.
>
> These are probably off a few degrees, since the way I've checked
> this is rebooting and hopping into BIOS ASAP.
>
> However, setting the alarm for 46 degrees only ever triggered
> 1 temp alarm. Anyone else see CPU2 hotter? Or is this just
> the variance of the board?
I see roughly the same variation in my boards here (2x366@>550) with temps
typically CPU1=34-35, CPU2=38-40, and board=38-41 under Linux running multiple
seti clients (using lm_sensors, Alphas/FEP32s, 40mm fans on BX, good
airflow...) I seem to remember several posts querying/confirming this at
www.bp6.com.
Dunno if it's more the design of the board or the way the boards typically sit
in cases, with CPU2 tucked into a corner under a PS. I'd be curious to know if
more even temps between CPU 1 & 2 correlated to better (not more, probably some
clever baffling would be needed) ventilation around that region of the case.
Maybe something like inward airflow both at the front of the case as well as
right near CPU2 while ducting Alpha exhaust directly out the case. I've seen a
couple of rigs like this but don't recall the reported temps.
Cheers,
K.
--
=- To unsubscribe, email [EMAIL PROTECTED] with the -=
=- body of "unsubscribe linux-abit". -=
