[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Patch: Idea for saving power on laptops
From:       Matt Newell <newellm () blur ! com>
Date:       2007-05-18 16:15:31
Message-ID: 200705180915.31626.newellm () blur ! com
[Download RAW message or body]

On Friday 18 May 2007 08:59, Guillaume Laurent wrote:

> Good luck in explaining this kind of concept to the average user.
>
> Come on, don't you think people would have thought of this already ?
> This just can't possibly work. How about a terminal or an app like
> kdevelop or eclipse ? As a user, the last thing I want is having to
> worry if minimizing a window isn't going to wreak havoc else where
> (oops, my build has been laying there frozen for hours just because I
> minimized the terminal I had typed 'make' in).
If the option is off by default and only enabled on a per-application basis, 
then there wouldn't be any problem with it happening accidently to kdevelop 
or eclipse.

> This is putting in the 
> user's hands some decisions which belong to the OS (allocating CPU to
> apps). And unless I see tests indicating otherwise, I'm pretty
> convinced the power saving gains will be barely measurable, because
> the OS already allocates CPU to apps according to their needs, so a
> minimized app won't eat much anyway (unless it's actually doing
> something useful).
>
I completely disagree.  Do you think it's the OS's job to decide that some 
stupid web page shouldn't be wasting cpu(energy) at a specific moment?  It's 
not, because the OS doesn't know or care if the window is minimized or 
maximized, or what it's purpose is.  The user on the other hand does.

What if it's the user's habit to leave 5 web browsers open with 10 tabs in 
each.  The user might not be looking at any of them, just playing music or 
something, yet they could be wasting tons of battery power, doing nothing for 
the user.

It could be argued that the application should be fixed to use less CPU when 
minimized, but in the meantime a simple to implement and activate feature 
that is properly presented to the user could be very usefull.  Your reasons 
for rejecting the idea don't seem very well thought out to me.

Matt
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]