[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: preload



Hello,

On Wed, Jan 23, 2008 at 01:48:50PM +0100, Linus Walleij wrote:
> Load time is something time does not measure AFAIK.

time can measure real wall time.
If it is supposed that the user will actually _notice_ (and
appreciate) the difference, then it should really be measurable as a
difference in real time.

So script the application to exit as soon as it starts and then do the
measurement.  (After a reboot, of course.)

For application which cannot be scripted to exit, use a trained
monkey, real (your son?) or virtual (expect).  In this case, you
really should do multiple measurements.  The number of measurements
depends on the spread of the monkey's reaction time.
(Make it so that the mean-root-square error is significantly smaller
than the mean value of the measured difference.)

Happy reboots!

Stepan Kasal


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]