>Geoff Canyon wrote/ schreef:
>
>> First, MetaCard surprised me by being able to move 50 32x32 images
>> reasonably well. It's not as smooth as JewelToy, but much better than
>> I expected. I'm using images with the fileName property set to a gif.
>> Has anyone tested to see if there's a better option?
>MMmm, did you try having buttons with the icon set to the ID of an (already
>imported) image?
I hadn't, but now I have.
I've now upped the number of images to 100, then to 200, in an effort
to break MetaCard's spirit. This is all on a PowerBook G3 400, which
while no slouch is not a fast machine by today's standard. MetaCard
was able to move 100 32x32 images around the screen and still look
pretty good doing it. 200 images got me to about 3 frames per second,
with noticeable jumps in the movement of the images.
I had to go so far in order to figure out which was faster: images
with a reference to an external gif, or buttons with an icon
reference to an imported image. The images are a little bit faster,
but only a little. One difference is that the images are noticeably
faster when some of them are moving offscreen. Obviously the smart
thing to do in this circumstance is to not move things that can't be
seen by the user. But still, if many of the images are offscreen,
images are faster than buttons.
Does anyone know of a way to reliably compare animation rates? In
SuperCard, I would use the nudge command. That way I would know that
the same number of "frames" were being animated no matter what and I
could time the results to compare. As it is, I just added objects
until the frame rate became visible, and compared that way.
Finally, I tried locking messages while the animation was happening,
and this didn't seem to have much of an effect, which surprised me. I
expected significant overhead to be reduced if MetaCard didn't have
to pay attention to what was happening everywhere and pass it on to
my objects. Either the overhead isn't significant, or locking
messages doesn't operate at a low enough level to help this, or my
testing was flawed.
Regards,
Geoff
Archives: http://www.mail-archive.com/[email protected]/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.
