CPU temps

[Lonni J Friedman]

On Fri, Mar 7, 2008 at 6:03 PM, Alan Jackson <ajackson at oplnk.net> wrote:
> On Fri, 07 Mar 2008 17:55:20 -0800
>  Tony Alfrey <tonyalfrey at earthlink.net> wrote:
>
>  > Vu Pham wrote:
>  > > Tony Alfrey wrote:
>  > > [...]
>  > >
>  > >
>  > >>> Kurt
>  > >>
>  > >> Yes, but are 26C and 28C really the temperatures of your CPU?  This is
>  > >> lower than body temperature.
>  > >>
>  > >>
>  > >
>  > > As long as it is higher than room temperature then it is possible, I think.
>  >
>  > Yes, but it will be higher than the ambient air temperature in the box.
>  >   The finned heat sink itself will be higher than that and the CPU
>  > (sensor is where? - maybe on silicon in the chip?) will be higher still.
>  >   I see from specs that a Pentium IV can dissipate some 60-ish watts
>  > when operating under substantial load (not idle).  I see that a typical
>  > big fan/heat sink combination for CPUs has a thermal impedance of about
>  > 0.6 C/W.  So 60 watts x 0.6 C/W yields a CPU case temp of 36 C above
>  > ambient.  If ambient in the box is perhaps a little above room temp,
>  > let's say 35 C then the CPU is at 71 C.  I see that Intel specs a
>  > maximum case temperature of 85 C.
>  >
>  > So Kurt is right:  103 C is too high, but 26 C sounds as though the CPU
>  > is totally at idle.
>  >
>  > BTW, how do you make that little °C symbol from the keyboard?
>  >
>  > I don't know stink about computers but I know physics.
>
>  Hmmm... my dual core AMD runs a little hotter
>
>  ]$ sensors -f
>  k8temp-pci-00c3
>  Adapter: PCI adapter
>  Core0 Temp:
>             +122°F
>  Core1 Temp:
>             +111°F

since when is 122F higher than 103C ?


-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
L. Friedman                                    netllama at gmail.com
LlamaLand                       https://netllama.linux-sxs.org