Re: Should 470629 be RC?
Steve Langasek wrote:
This is hyperbole. Percentage of CPU time is not linearly related to % of
energy usage
Perhaps you're unaware that Linux does no-op instructions
when it has nothing to do, and that reduces the power consumption?
I've run vanilla Linuxes without cpuspeed control, and you can watch
the CPU temperature change when you run a process.
Basic physics says if the CPU gets hotter, the energy to heat it up
came from the wall plug. See
http://monkey.org/openbsd/archive/misc/9808/msg00196.html
for an early reference from 1998.
Also, haven't you ever heard a computer fan rev up on a laptop
when you started doing something? Jeez! Be real.
The power figures were actually rather conservative -- I assumed
a increase in electrical power that was half as fast as linear,
just to be on the safe side. I also assumed a fairly modern mix of
CPUs to estimate the effect. (If everyone was running P4 machines,
the costs would be 3x larger.) So, it's not exaggerated hype.
(If I wanted to hype, I could have used an estimate of 50M Debian
users...)
It looks like I had this package installed on one of my systems for some
reason - but the CPU usage according to top was on the order of .3%, not 3%,
for a much slower system, which, BTW, has no power management capabilities,
so the net power consumption from having this process running is
approximately 0.
I can't say why you got much less CPU usage than I did, but my
3% figure came from averaging the result of top over a few minutes
when the system was idle (as it is, for most of the day).
Are you running Debian testing? Is it that you don't have
many USB ports? What makes you think that your single
measurement is any more representative than mine?
From http://qa.debian.org/popcon.php?package=libchipcard, roughly 1000
systems, or about 1.3% of systems participating in popcon, report having
this package installed.
I admit that Popcon show that this package is currently not
popular, but that's not necessarily conclusive. Popcon
mostly measures Debian Stable, and says little about Debian Lenny.
Also, one could make arguments that the
people contributing to the popularity contest are probably
sophisticated, and such people may remove inessential
packages). However, *I* didn't install the package
intentionally. Did you?
If you didn't and I didn't, don't you think that lots of other people
might have it also? It showed up on my Debian Testing system, and also
on my
two Ubuntu 8.041 systems.
Anyhow, even if we take the Popcon figure, we're still talking
$500,000 of wasted energy. Even if we *also* take your
CPU usage from top, we're still talking $50,000 of wasted energy.
So, even if you're completely right, isn't it worth fixing?
Reply to: