Following two previous threads (one dating back I think to 2008 or so) I am trying to build an FCPX transition in Motion 5 where Video A breaks apart into a 10x10 grid of tiles and each tile rotates in sequence to reveal Video B. However, I want Video B to also be rotated in as the back side of the tiles.
I have managed both things independently but can't seem to make them work together.
In my first attempt, I created a replicator with 2 cells and used each video layer as a cell source. The effect is perfect except that it's working on 100 small copies of my video when I want the full-size video to be broken into 100 tiles.
My second attempt created a replicator based on a rectangle shape and used the replicator as the image mask for the video layers. This makes Video A disappear in the desired format but Video B is either sitting still behind it (just a normal reveal of Video B) or is not shown at all depending on how I set it up.
As someone who has designed similar transitions for FCPX, I can tell you that there are actual "under the hood" issues with the way FCP interprets Motion replicator instances. You don't always get a correct translation between the two programs and in particular, this exact transition is one that cannot be properly accomplished with replicators (believe me, I've tried)
You can get the desired result using clones however, but there is a caveat: this will significantly limit the number of "tiles" you'll be able to add before the transition bogs down the program or causes instability. I would not attempt to do more than a 3-4 columns/rows in your grid.
Thank you, Andy. To be clear, are you suggesting make an image mask of the video layer to represent one tile and then make clones of that, or making clones of the whole video layer and then mask each one off to the size and position needed for each tile?