On 03/08/2010 04:25 PM, Oswald Buddenhagen wrote:
> On Mon, Mar 08, 2010 at 12:12:04PM +0100, Holger Macht wrote:
>> You're taling about compile clusters, so you're obviously an experienced
>> user and technician, but people aren't. Nobody wants to hinder you from
>> setting things like that _manually_.
> riiiight ... because power users *want* to fiddle with the command line.
> that's why we use kde, after all ...
Ok, so in my opinion, this will results in the KDE desktop being _only_
for power users, because the common user will get it wrong. So that's a
design decision I don't have to make...
>>> a third datapoint is the wrongness of the assumption that the workload
>>> is fixed and should just be executed as fast as possible. this certainly
>>> isn't so, for example in interactive applications which refesh "as often
>>> as possible" there is always somthing to do. downclocking the cpu is the
>>> only way to conserve power in such a scenario.
>> Can you give an example where an application refreshes that fast that
>> you would notice the impact?
> an example would be an ide which is re-indexing in response to source
> code changes. if the indexing takes longer than entering two characters,
> then the first indexing will be simply aborted, having consumed only
> part of the power.
> a similar scenario would be imaginable for any cpu-intensive interactive
> application which immediately displays e.g. a wireframe or interpolated
> image in response to an action and then refines a display incrementally.
> no example at hand right now.
> games which try to make an as-good-as-possible physics simulation may
> also continuously run at 100%.
>> Frequencies are switched in microseconds these days.
> that's utterly irrelevant in this context.
I was talking about dynamic frequency adjustments, switching from lower
to higher frequency.
>>> also, perceived slowness of the machine has a psychological effect of
>>> not demanding too much of it, i.e., it changes the usage pattern to
>>> something more power-friendly.
>> You don't notice the difference from using ondemand vs. performance,
> i'm talking about powersave vs. anything performance-oriented.
Anyway...this has been discussed a hundred times on all kind of
different lists. I won't convince you. On the other side I have my
facts, my experience and read _a lot_ of opinions from linux power
management developers all over the world. In the end it's not me to
decide if this gets applied or not. I just think it's the right thing to
There are tons of information about this topic on the net, so either
someone who can commit this believes either me, or you, or builds up his