On 2011-03-11 (March, Friday) 14:49:15 ischou wrote:
> Well, my problem is not exactly that my computer is noisy (though it can be
> at times), but that it seems noisier when I'm running programs in Wine
> than native XP
As I explained this may be a good thing. But to know for sure you need to
measure temperatures during same high load lasting at least 20-30 minutes in
both Windows and Linux. "Same high load" means ideally using same application
for testing in both Windows and Linux. "Personal perception" is very
inaccurate, you can't rely on it enough to make any conclusion in this case.
Example: running simple command like "yes &> /dev/null&" as much times as
you have CPU cores may heat CPU noticeably more than real-world application
that fully uses all cores, this is why similarity of high load is important:
different kinds of load will heat components of your computer differently, even
if total percentage of CPU usage in top for each load measures exactly the
same.
> my original question is whether Wine CPU usage can be
> selectively/temporarily throttled back by using tricks/hacks like
> minimizing the window.
Well, I already answered this question ("yes, sometimes it can be"). But I
can give you detailed answer: yes, sometimes minimizing application's window or
going to another desktop can bring down CPU usage if your program uses CPU a
lot for drawing GUI in its window (like Metatrader 4 when online with a lot of
graphs on the screen). But please understand that some inefficients in Wine can
lead to higher CPU usage than on Windows. Reverse is also true: sometimes Wine
is more efficient than Windows and will use less CPU. Only way to know for sure
is to profile your application on Windows and Linix+Wine carefully on exactly
same hardware. If you find noticeable inefficiency in specific Wine function you
can report a bug. But if you satisfied with performance of your application you
really shouldn't bother about this, simply use recommendations I already gave
to you to make your computer less noisy.
