Greetings All,
Sorry about the earlier e-mail, I now see it was not
clearly stated.
Purpose:
1 Testing/demonstration page for different aspects of
a simple animation and to time those different aspects
running under DynAPI in reference to the challenges of
setTimeout.
2. Provide a benchmark of the timing of the
system/browser it is running under (it tries to
calculate what the actual play rate will be). As I
get more feedback from the users list on the
setInterval test, I will craft the code to try and
closely predict the playspeed for different
combinations that we support. The ultimate goal is to
provide code for DynAPI that will set the ms number at
a crossbrowser number that plays back at a uniform
speed independant of the platform/browser.
3. (Most important) Get users opinions about the
multithread code, or rather its affect on the
animation. Richard, if you noticed the playrate when
running under multithread, it actually allowed the
animation to play at a rate below the system timeslice
minimum (below 55ms on my machine, typically at 28
ms). That is the goal of this implementation of
multithread.
Thoughts about multithread and the animation in
general:
I too observed what Richard referred to as "jumpy"
playback. Pathanim and DynAPI do not make jumpy
playback, the ability of the platform and browser do
this to all DHTML code, it is a universal problem.
However, once I started watching the normal play
closely, I saw the same "hesitations" in the normal
animation. The multithread code just made them more
obvious. From that aspect, I will probably not pursue
this implementation. Does anyone have need of a
faster playback if it is not smooth and steady?
Yes, thread.js does give us "multithread" but it is in
reference to other animations running at the same time
under DynAPI. I probably used a bad choice of words
here, but multithread in this demo means that this
individual animation has two timing threads, one
triggered by the other.
Finally, this demo page was an attempt to apply the
principles/facts learned about setTimeout and
setInterval to a DynAPI example that everyone could
observe without having to code it themselves. I tried
to throw in things like an image and transparent
background to give the browser a little work while
rendering the animation.
Glad to know someone is looking at this stuff, my
off-hand comment in the earlier e-mail was slanted
towards the fact that there is little discussion of
DynAPI2 as an animation platform.
Still plugging away, I am in contact with the original
author of the setTimeout setInterval article at DHTML
labs and also with other writer/editors at
webreference.com.
Thanks to all who have responded to my requests for
tests so far. And thanks to Richard for asking for
clarification.
Cheers,
Dave C. "You Changed What?!?"
__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/
_______________________________________________
Dynapi-Dev mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/dynapi-dev
