Yup.
So I'd just get a more accurate number if I could just run this test on
a LZX-native machine that doesn't have around 90 WinXP-related processes
running....
-e
On Fri, Jun 29, 2007 at 1:20 PM, P T Withington wrote:
My point is that the local server could be perturbing the working set
of your browser.
AFAIK, the inittimer only times the startup time of the app, but since
it is a wall clock, any other process on your machine could be
stealing cpu from the app.
On 2007-06-29, at 13:00 EDT, Elliot Winard wrote:
I am running against local server.
App is compiled/cached on server.
I see the same difference if the browser is already running and
paged in or not running.
I'll try the same measurements on IE and see if I get the same
results.
-e
On Fri, Jun 29, 2007 at 8:29 AM, P T Withington wrote:
Are you running against a local server? Is your app already
compiled/cached in the server? Is the browser already running and
paged in? &c. &c. <inittimer/> is just a wall clock, so if other
processes/threads (in the server, browser) are getting time slices,
you time them too.
On 2007-06-28, at 23:16 EDT, Elliot Winard wrote:
I'm testing size and inittime differences between OL3.4 and
Legals. I noticed that the first time a browser hits a SWF during
each browser session takes longer to initialize. My Firefox cache
is set to 0 so I don't think it should be caching anything.
I'm using the <inittimer/> util to get applicaiton startup times.
Is this a case of Firefox not doing what it's set to do, or lower-
level caching?
Thx,
-e
---=---===-------
Elliot Winard
Sr. Software Engineer
Laszlo Studios
---=---===-------
---=---===-------
Elliot Winard
Sr. Software Engineer
Laszlo Studios
---=---===-------
---=---===-------
Elliot Winard
Sr. Software Engineer
Laszlo Studios
---=---===-------
