I'm working with Newton Physics on choreographies with several hundred interacting objects, and I've made a small discovery that's saved me a lot of time.

It's sometimes desirable to make a single frame of a Newton Physics simulation into the choreography's static state. Export as Model will work fine with small scenes, but it tends to choke on large scenes, and if you're using a lot of instances of the same objects you'll just be filling A:M's memory with redundant geometry.

Instead, the following will work:
1. Set up your scene for the Newton Physics run.
2. Save
3. Simulate Newton
4. Step through and find the frame you'd like to freeze (say, 00:01:12)
5. Revert to saved
6. Set the choreography length to the position of the freeze frame (i.e. 00:01:12)
7. Turn Animate Mode OFF
8. Simulate Newton

With animate mode off, each frame the Newton Physics plugin saves overwrites frame 00:00:00, resulting in a static scene. Hope this is helpful.

Good trick!

I've always sorta done the same thing by running the simulation to get all the keyframes... then you can just delete the keyframes before and after your 'hero' key...

Can you give a visual example?

In this case, I'm only interested in the puzzle in its half-demolished state. This happened around frame 00:01:10, so I set the chor range to 00:01:10, turned Animate Mode off, and ran Simulate Newton. Notice that only frame 00:00:00 has been created.

With close to 500 pieces in the puzzle, manually deleting keyframes quickly bogs the interface down and becomes a slow, finicky and repetitive task.

Click to view attachment

I am still having trouble wrapping my head around this concept. Why would it be desirable ?

There's plenty of reasons why you might want it... one being that Newton can provide a nice 'scattering' of objects- as illustrated above'...you don't always want them moving...maybe you want them to be a still background. SpaceToast is simply showing a way to 'capture' a moment in 'Newton time' and freeze it.