Linux uses less power than Windows?
James Knott
james.knott-bJEeYj9oJeDQT0dZR+AlfA at public.gmane.org
Mon Nov 15 03:58:55 UTC 2004
I seem to recall an article, in the Linux Journal, a few years ago, that
reported on the difference in battery life, when running Windows vs
Linux. As I recall, it was that "busy wait" that was the culprit.
Robert Brockway wrote:
> On Sun, 14 Nov 2004, Stewart C. Russell wrote:
>
>
>> * A modern desktop PC or lightly loaded server spends most of its
>>time waiting for the next interrupt. Interrupts come from timer ticks,
>>keystrokes, mouse motion, or data ready on a disk, modem, or network. In
>>the Microsoft system, the CPU "idles" at full speed and full power
>>during this wait time. When Linux is idle, the CPU halts in a low power
>>state.
>
>
> Believe it or not this one is true, but only for MS-Windows98 or earlier.
> Linux and MS-WinNT family use a special opcode (called IDLE or HALT,
> forgotten right now) while MS-Win98, MS-Win95, etc use a "busy wait".
> Busy wait keeps the CPU hotter (one person reported a 10C difference a few
> years ago).
>
>
>> * The Linux file system and memory management are more efficient
>>than Microsoft's, so the same application program runs with a lot less
>>disk activity.
>
>
> Honestly I think it would have a lot more to do with the purposes the box
> was being put to. A fileserver will use more energy regardless of its OS.
>
>
>> * Most Microsoft boxes directly connected to the Internet are
>>infected with spyware and trojans, and loaded with programs that are
>>supposed to defend against those things, which generate more activity
>>than the intentional software. Linux is pretty much immune to that stuff.
>
>
> Seems pretty marginal to me.
>
> So I'd score the article 0.5 out of 3 :)
>
> Rob
>
--
The Toronto Linux Users Group. Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml
More information about the Legacy
mailing list