Hmm, that kind of exception is the reason that I made "RectHit" to replace the stock Hit Patch. I hadn't seen those exceptions in a long time, so I thought it was cleared up and used the stock Hit Test for this - and maybe it IS in SL, but just not Leopard. To my memory, it's because of the Image Mask function when running in Leopard (drats).
I guess I'll do more testing, and consider replacing the stock Hit Test with my own. I am interested in this running well in Leopard since it reproduces the Interaction method, more or less.... OK: I booted in Leopard. The Image Mask IS FOUL in Leopard, but works in Snow Leopard (10.4+). Here's an attachment that replaces it for those that want some kind of Interaction like function in Leopard. (Too bad, the image mask is a definite advantage of that method over the Apple interaction setup, and my simple hit test doesn't take that into account). -George Toledo On Thu, Aug 12, 2010 at 4:26 PM, Alastair Leith <[email protected]>wrote: > I get exceptions when I run your comp GT. I also get that Billboard > doubling in size when I re-run but only one doubling not endless doubling > like is the norm with that bug. > > Seems Leopard safe to me (?). > > 0x92bf4fbc: GFException > 0x92bf6ae0: GFThrowException > 0x92bb8250: -[QCImageManager > createPixelBufferFromProvider:withFormat:transformation:bounds:colorSpace:options:] > 0x92bb8164: -[QCImage > createPixelBufferForManager:withFormat:bounds:colorSpace:options:] > 0x92c87b60: -[QCHotZone execute:time:arguments:] > 0x92b75eec: -[QCPatch(Private) _execute:arguments:] > 0x92b76314: -[QCPort _execute:arguments:] > 0x92b75de4: -[QCPatch(Private) _execute:arguments:] > 0x92b76314: -[QCPort _execute:arguments:] > 0x92b75de4: -[QCPatch(Private) _execute:arguments:] > 0x92b76314: -[QCPort _execute:arguments:] > 0x92b75de4: -[QCPatch(Private) _execute:arguments:] > 0x92b76314: -[QCPort _execute:arguments:] > 0x92b75de4: -[QCPatch(Private) _execute:arguments:] > 0x92b76280: -[QCPatch(Extensions) executeSubpatches:arguments:] > 0x92b75eec: -[QCPatch(Private) _execute:arguments:] > 0x92b76314: -[QCPort _execute:arguments:] > 0x92b75de4: -[QCPatch(Private) _execute:arguments:] > 0x92b7619c: -[QCPatch(Extensions) executeSubpatches:arguments:] > 0x92b75eec: -[QCPatch(Private) _execute:arguments:] > 0x92b7470c: -[QCContext renderPatch:time:arguments:] > 0x92b7448c: -[QCGraphicsContext renderPatch:time:arguments:] > 0x92b73580: -[QCOpenGLContext renderPatch:time:arguments:] > 0x0000db2c > 0x92be978c: -[QCView render:arguments:] > 0x92bea600: -[QCView(EventHandling) mouseMoved:] > 0x95075e40: -[NSWindow sendEvent:] > 0x95048aa0: -[NSApplication sendEvent:] > 0x0000dc18 > 0x94fb58d8: -[NSApplication run] > 0x00003f84 > > > > On 12/08/2010, at 11:51 PM, George Toledo wrote: > > The last attachment may open up with an exception (but will still work > fine). > > This is the same attachment, sans exception (I honestly have no idea why > that QCFakePatch message occurs with the last qtz). > > -George Toledo > > On Thu, Aug 12, 2010 at 9:48 AM, George Toledo <[email protected]> wrote: > >> I've replicated interaction using built in patches, except that getting >> offsets to work inside an iterator is becoming puzzling to me (maybe because >> I've been looking at the patch too long). >> >> If I place in an iterator and power the offsets with interpolation >> patches, as standard, all the initial offsets are fine. What I'm not sure >> about is how to get it so that if I "add" x/y from the mouse to x/y >> interpolation offsets, x/y from the mouse ends up doesn't affect every >> iteration, just the one that is being interacted with. >> >> I'm not placing this in an iterator, just uploading it as is. >> >> It may actually be of significant use, as it adds the function of an image >> mask to the hit zone. It also restores initial offsets without opening and >> closing the composition (unlike the the stock Interaction). This method can >> also be rejiggered to support Z and rotation stuff if desired. >> >> If it occurs to anyone how to get offsets going in an iterator reliably, >> please post back to list. It may seem pointless at first, but if this can be >> made to work in an iterator it would actually be much more flexible than the >> stock interaction (and arguably already is for scenarios where it doesn't >> have to work in an iterator). >> >> Thanks, >> George Toledo >> > > <Interaction_GT Alpha 2b.qtz>_______________________________________________ > 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/qc.student.au%40gmail.com > > This email sent to [email protected] > > >
Interaction_GT Alpha 3.qtz
Description: Binary data
_______________________________________________ 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]
