The easiest way to do this is to put the whole thing in a constant sized render in image and mask that.
Sent from my iPad On Apr 19, 2012, at 11:34 PM, Alastair Leith <qc.student...@gmail.com> wrote: > Hi List > > I'm trying to make an image mask for a Billboard such that the masked area > stays static (relative to Viewer window) while the image of some text moves > up and down (as if) inside the bounds of a masked window. The problem I am > having is that there is no simple way to give an absolute position for the > mask while moving the image independently, like you say a mask in PS or AE > behaves. Typically the billboard patch scales a > rendering-destination-dimension sized mask image down to be (somehow) > relative to the size of the input image rather than use it as a mask the size > of the rendering destination (Problem 1). > > If I use an Image Transform Patch to translate the text image up/down on the > Y-Axis the Billboard patch basically ignores the change in bounds and just > centres the image at the same origin as determined by X pos, Y pos and image > size (Problem 2). > > I did make a successful composition that creates a mask the same size as > source image (an Image from String patch output) by 'screen' blending black > and white cropped images (doing math for the crop inputs to get positioning > relative white/black correct) then using a Mask to Alpha patch and inputting > it to the Billboard. By animating several of the cropping offsets and heights > and the Y-Position of the Billboard I can generate the desired animation of > text scrolling up and down inside window bounds. Problem with this solution > is it seems overly complicated, a little CPU hungry on my old mac anyhow, and > because the crop patch uses integer inputs, an animation that only moves the > mask say 33 pixels is noticeably jerky (only just) as compared to just > animating Y Postition of a billboard with cropping sprites all around it. > > Maybe animating a billboard with cropping sprites set to alpha-0%-black > (blend mode: replace) on higher layers inside a RII patch is the way to go > but I'd like to avoid RII patch on principle that's more overhead. I don't > mind using RII patch to construct the mask if it is static and wont > re-evaluate after initialisation but that just gets me back to Problem 1. > > Any assists/demos/thoughts gratefully received. I'm sure some of you have > been here before — I know I have :-) > > Alastair Leith > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Quartzcomposer-dev mailing list (Quartzcomposer-dev@lists.apple.com) > Help/Unsubscribe/Update your Subscription: > https://lists.apple.com/mailman/options/quartzcomposer-dev/asabatelli%40apple.com > > This email sent to asabate...@apple.com
_______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list (Quartzcomposer-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
