> > It would be interesting to learn where the edges are, and why, just out of > curiosity :) Does your ATI split at widths similar to my GMA950? >
I'll have to do a thorough check, but I believe it does. It definitely splits at some very odd points! The internal seams are from subdivisions (I'm assuming you know that now), > which happen arbitrarily. So clamping won't handle those. The affine > transform trick posted earlier on this list (regarding gaussian blur) would > probably remove the edges without needing to clamp. Yeah, I got the subdivision part. I'll have to try the affine transform. Am I right in thinking the performance impact is minimal? > The scaling I was referring to was the ROI rectangle, not the image. The > image size remains unchanged. I was curious as to why ROI needed to be set > so much higher than Tom's simpler (and intuitive) explanation -- all the > examples use CGRectInset(r, -radius), but we're effectively needing > CGRectInset(r, -radius*4). Anyway, this just tells CoreImage how many > pixels "over the edge" each region needs, so that it'll correctly sample > them instead of black. This is applied for each subdivision, so that each > subdivided region gets a bit of overlap. Right, thought you were wondering why I was scaling the video in up to 4k size (as in resizing it before processing, which doesn't make much sense). I don't know why the ROI enlargement needs to be so much bigger, but I've had issues like this when clamping the coordinates within the filter at times. If I clamp to zero, imagewidth it will remove the line at the bottom/left as it should but I've seen cases where a bigger than expected difference is needed at the top/right (like width-4 instead of width-1). I didn't investigate much but if I come across a filter where I'm still doing that I'll see if I can figure it out. Chris
_______________________________________________ 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]
