Layers and compositing
Showmesh can display several video, image, or text layers simultaneously. The program is composited bottom to top using premultiplied-alpha over blending.
Ordering
Visuals are ordered by:
- lower Layer value underneath;
- on the same layer, the earlier-fired cue underneath;
- stable cue ID order as a final tie-breaker.
A useful convention is:
| Layer | Use |
|---|---|
| 0 | Background video |
| 10 | Main image or second video |
| 20 | Graphics or lower third |
| 30 | Emergency or blackout layer |
Leave gaps between layer numbers so new levels can be inserted later.
Opacity
Each visual layer's video.opacity multiplies its alpha. 0 is transparent and 1 fully visible. An image file's own alpha is preserved and multiplied by cue opacity.
Current limitations
- Layer position, scale, and rotation are not editable.
- Per-output crop and corner pin are not implemented.
- Full alpha in the final external program output is not complete.
- The GPU compositor exists in source, but every Windows/GPU combination needs a hardware rehearsal.
A crossfade needs overlap
If Video A is stopped before Video B becomes visible, there is nothing to crossfade. Keep both cues running for the transition and stop the lower cue only after the fade.