Re: Shut down or leave on?
On Aug 29, 2007, at 8:01 AM, Dan H wrote:
88 watts it used to be. A modern desktop PC will consume several
hundred watts of power, which is as much as it takes to make a room
"quite hot" -- do you think you could get your room hot by leaving
the lights on?
It might be worthwile to measure power consumption again, with
modern equipment -- you'll bes surprised.
Have you tried it yourself? I have. I was surprised how *low* the
consumption was.
Now granted, my machine isn't bleeding edge. It's a 2.0 GHz Xeon
with two hard disks, an NVidia GeForce 6, and a TV tuner. At full
whack it draws around 180 watts, but when the CPU is idle it drops as
low as 60 watts. (This is just the computer itself. I haven't
measured the monitor, but it's a flat panel so I'd expect it to draw
somewhere between 40 and 60 watts. I turn it off when it's not in
use.) I checked these figures because I *do* leave that machine on
all the time -- it's one of the backends for my MythTV system. I
wanted to see if it was cost-effective to replace it with something
smaller and more efficient. (I concluded it wasn't.) This did
convince me not to run any of those background computation programs,
like SETI@Home, though, since they effectively triple the power
consumption by keeping the CPU constantly busy.
Don't confuse the nameplate wattage of a computer power supply with
the actual power it draws. Most computer power supplies don't see
peak load, and if they do it isn't for long. Even the dual-Opteron
rack servers I deal with at work have a maximum power consumption of
only about 300 watts, and they draw significantly less than that when
idle.
It used to be you needed expensive equipment to accurately measure
this stuff, but nowadays you can pick up a Kill-A-Watt for under $50,
and it's highly educational. I've found a surprising number of
household electronic devices that draw exactly the same amount of
power when turned off as they do when turned on, making their power
switch little more than a placebo. My VCR, for example.
Reply to: