Moving between Matrix and GL: Best Practices?
hi everyone -
i would like to process a live video feed with gl.slab. i'd like to take the output of the gl.slab and feed it back into itself, using its compositing shader operators to manipulate the feedback.
i'm curious what's considered the best way of doing this - and whether there are any speed/efficiency benefits to going this route, rather than simply using something like jit.op.
thanks for any help!
You don't need to get your texture back to the CPU to make feedback, just add a blank jit.gl.slab or jit.gl.pix between your slab outlet and second inlet. You can see an example of that in the builtin tp.slide.jxs-help patcher (in Help > Examples > jitter-examples > render > slab-helpers).
If you want to process geometry with feedbacks, look at [jit.gl.tf].