I don't understand, then, why the unrolled version works. I'm changing the particle count there just as in the iterator version. The only difference is that in the iterator version, the histogram is forwarded through the iterator to the macro, as opposed to just going straight to the macro. Either way, it is (should be?) precisely the same data that actually reaches the particle system patch.
From experimentation, we've found that the particle system patch does strange things when time goes backwards for it -- in the unrolled version, time moves forward for every patch always, so it's totally happy to have changing counts. In the iterator version, it steps through time, then jumps back for the first iteration (patch time + 0). This is where it gets frustrated, since particles can be undead or unborn, and it doesn't know anymore because too much is changing (time and population).
It makes a bit of sense in programmer terms (it's difficult to create particles that were dead on the previous frame, if you don't know how many particles there were at the time), but to non-programmers it's totally weird. Or maybe it's weird to everyone.. Anyway:
Here's a fun demonstration to illustrate it (no iterators were harmed in the making of this composition)
awesome particle misfeature demo.qtz
Description: application/quartzcomposer
use the space bar to toggle between "forward only" and "back and forth" time traversal. Note the difference. Impress your friends. Maybe file a bug report :)
(The trick: The lfo on the second system has an amplitude of 0, meaning it never changes. However, the particle system doesn't know that, so it assumes it changes and does whatever re-init stuff it does on change. When time's going backwards, it forgets to re-create particles that were dead in "the future". This is the foundation of your random particle spray feature).
-- [ christopher wright ] [EMAIL PROTECTED] http://kineme.net/
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
