On Wed, 8 Nov 2006 01:23:33 -0500, "Chris Buechler" <cbuechler at gmail dot com> wrote:
>Just because you have a 500 wt power supply doesn't mean it's using
>500 watts.  I'm no power supply guru, so it may use more than a
>smaller power supply would, but I don't think there would be much if
>any difference in actual power consumption.  I have one box with a 500
>wt power supply that actually draws 275 wt under normal operation, and
>another that draws 175 wt under normal operation (these are boxes with
>numerous drives, power draw if they were firewall boxes would be much
>less even on a 500 wt supply).

For well designed switching power supplies the efficiency curve will be fairly
flat over a range exceeding the upper half of the output power range with only a
small dip if any at maximum rated power.

>Every self-respecting geek should have a Kill-a-watt meter (google it,
>they're available all over and they're cheap) to measure power
>consumption.  I have a couple, they're great for UPS size planning and
>gathering power consumption info, for cost or other reasons.

This is especially true considering how inexpensive they have become.  I still
rely on my clamp on ammeter which given the poor power factor of most computer
power supplies is really only good for comparative purposes.