General PC question

  • Thread starter Thread starter Golden Oldie
  • Start date Start date
Paul said:
Check the +5VSB rating on the side of the supply. It is limited
to 2 amps. That means 10W max from the supply. At 50% efficiency,
the max drawn from the wall, is 20W. It should be a little bit
less than that 20W number. At least some motherboard manuals,
give hints as to how much current is needed during sleep. So
that is how I arrived at the 10W-20W estimate.

Paul

I just measured my current computer. This is for a P4C800-E Deluxe
with P4 Northwood processor. It has 2x512MB DDR400 memory, and
I just put the computer into S3 suspend to RAM (sleep).

+5VSB 0.70 amps which is 3.5W inside the computer

Assuming my PSU is 50% efficient making +5VSB, that means
the computer uses 7W when sleeping. The 7W keeps the RAM
contents, and my LAN chip is still running in some state
as well.

Paul
 
Safe for you (fire etc.) or the PC?

I'd ssay it depends on the power quality - if spiky and horrible
(rural, generators, lightening, dodgy utility company) then if the PC
has no work to do, turn it off, else leave it on unless idle for more
than (say) six hours.

When PCs of the 2000-2002 era fail, they tend to do so due to bad
motherboard capacitors. I see a lot of these, and my impression is
that it is the number of powered hours, rather than power on/off
cycles, that hastens these failures - so I've moved away from "leave
it on" to "turn it off if idle for > 6 hours".
Should Al Gore be addressed for picking up the issue?

Al might want to ask Intel if their value-limited processors keep
power applied to the disabled parts of the chip.

For example, the same core might be left with 2M cache as a "Pentium
4" or have access to 3/4 of it disabled to make it a "Celeron".

If I were Al Gore (actually, even though I am not Al Gore), I'd want
to know whether this "dead" circuitry consumes power, and if so, why.


------------------ ----- ---- --- -- - - - -
The rights you save may be your own
 
Back
Top