I have a shoot upcoming that I'm trying to prep that involves two 3D elements.

1. A 3D green/blue screen foreground element - think seated talking head shot against green or blue.
2. A 3D natural background element interior shot for the background. This shot will have no foreground items, some mid-ground items equal to the foreground distance, and deeper background items. Think interior of a large building lobby.

Both shots will be shot with mirror rig shooting parallel.

The issue really is one of dealing with 4 clip elements and a composite process. I would be curious as to how anyone who has done this already has accomplished it, or any input of ideas on how to do it. Post will be done on Final Cut Pro 7, compositing using Ultimatte Advantage plugin. I am testing out the Stereo 3D Toolbox now, but plan to purchase it if the workflow with that plugin will work.

I know all of the elements are there, I am just not sure what order to do it in.

Hi Bruce,

I'm working on a big project just like this now. Once you have selected your clip(s) in the timeline it is really best to move the workflow over to Motion.

The keyers in Motion are very good I'm also using the new PHYX keyers.
More info on my project here: 3D Fashion Shoot Uses Gaming Style to Make a Statement | Studio Daily (http://www.studiodaily.com/main/11932.html)

I will create a full tutorial using my material as soon as we have finished. In the meantime take the approach of editing your elements as four layers and then send them to motion.
In motion you don't have to worry about the silly way FCP handles frames with image wells. You can just drag from the layer below to the image well and sync will be maintained, even if you split the clip and move it.

Use one of the stereo3D preview modes in the plugin to fix your disparities and set your convergence, then select left only and copy and paste the sequence and select right only. Export those two sequences as your master outputs and you are done!

Interesting article Tim. I'm curious about one comment in it;

"our interocular had to be between 35 and 45 mm most of the time.” That appears to be a very narrow IO, less than the standard 65mm or 6.5cm generally regarded as normal eye socket distance. Were you putting everything in negative space? From the pictures it looks like your camera is very close to the subjects.

Back to this thread subject - I'm not that familiar with Motion, but I guess I'll have to start familiarizing myself. Does the green screen compositing happen for all four data streams independently or is it done after each 2-channel convergence is done? I can see a case for left eye FG+ left eye BG being composited first and then fused and converged to the right eye FG + right eye BG (composited) in your program. Or am I looking at this through the glass darkly?