[GLLUG] desktop freezing??

[Chris Chan]

The 5v rail is used on the vast majority of Pentium 3 and Athlon
boards, and the 12v rail is used on nearly all Pentium 4 and newer
boards, as well as some oddball Athlon ones. It's the ATX12v 2.1
standard that specifies the 2x2 connector should be on its own rail,
if I remember correctly.

On Nov 29, 2007 12:42 AM, Richard Houser <rick at divinesymphony.net> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> The +5V rail might have been used on this machine for the CPU, but on
> modern boards, I think it's actually one of the +12V rails (that's the
> little 2x2 connector).  One of the appropriate hardware standards (can't
> recall which) specifies such that this connector uses a different rail
> (in multi-rail systems) than the PCI-E power connectors to help balance
> the high draw devices out better.  For example, on my desktop, the CPU
> is on rail 1 and each PCI-E connector is wired to split the load across
> rails 2 and 3.  The +12V rails have a PCI spec limit of 20amp max per
> rail, so this is required.
>
>
> Chris Chan wrote:
> > Asus used quality capacitors on P3 era boards but cheaped out in the
> > Prescott P4 era where arguably quality caps were needed most. USB
> > powering issues would be my first guess since a lot of your devices
> > are bus powered (mouse, keyboard, webcam). Since both USB devices and
> > the CPU are powered on the +5v rail, your PSU's 5v rail is progbably
> > getting overloaded.
> >
> > On Nov 28, 2007 11:29 AM, Michael Rudas <audiotech50 at gmail.com> wrote:
> >> Benjamin Cathey wrote:
> >>
> >>> As far as Power Supplies go - the more wattage the better, correct?
> >>>  The system will only draw what it needs, right?  Or am I remembering wrong?
> >> Mostly right.  The efficiency and power factor ratings of
> >> higher-wattage PSs is kinda variable, but differences are (usually)
> >> minor, especially for quality units.  A 400W PS would not be overkill;
> >> an 800-watt one would.  I still think the PS is the problem, but check
> >> the caps, as well.
> >>
> >>> I figured Linux would have a message somewhere if the problem was overheating though.
> >> It was only a suggestion based on experience.  Linux can't read ALL
> >> thermal-monitor ICs, unfortunately -- but the BIOS can.
> >>
> >>
> >> ~~ Mikey
> >> _______________________________________________
> >> linux-user mailing list
> >> linux-user at egr.msu.edu
> >> http://mailman.egr.msu.edu/mailman/listinfo/linux-user
> >>
> >
> >
> >
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.7 (GNU/Linux)
> Comment: Using GnuPG with Mandriva - http://enigmail.mozdev.org
>
> iD8DBQFHTlFMUMkt1ZRwL1MRAnkzAJ9ohv7We1s+Pmb/2iOCbrJ9T1F3/ACgtK/X
> xWVndRm6qN0ksvYPsLlJhCI=
> =GQhP
> -----END PGP SIGNATURE-----
>



-- 
Chris Chan
AIM: crztales
YIM: crazytales562
MSN: cchan562 at yahoo.co.uk