Many good comments already.

My workflow is similar to what most have already put forward.

1. Render the iClone scene as PNG image sequence.
2. Use audio only as reference or for lipsynching.
3. Compose everything in the NLE, which is Sony Vegas in my case.

I use separate Vegas projects for things like credits, so that I can reuse them. In Vegas you can import another Vegas project and put it on the time line as a clip. I used that for the newscast in "The Lost Client", which consists itself of a sequence of clips. By using a separate project I didn't have to prerender.

I keep image sequences and dialog files organized by scene (which is usually an iClone project). Sound effects are not organized or duplicated per project and are reused as needed.

I time the dialogue for all participants in a scene in advance in Audacity, padding with silence where needed. Sometimes I need to make changes in timing or the speech itself, which then of course means redoing the lipssynching. However, the advantage of this approach is that it is easy to get the timing right in the NLE, as all dialog files for a scene start at the same point in time.

---

There's a lot of good info in this thread! So I won't go into too much detail, but my process is much like everyone else's:

- Each camera shot is a separate project in iClone (90% of the time, though I do sometimes use Camera Switch when I know what I want to see, and when, or when pesky particle emitters are involved.) And I keep each Scene in a separate folder (with from 1 to 20+ shots in each.)

- Voice Over audio clips in iClone are for lip syncing only, and usually get replaced in the editor with cleaned up, modified versions.

- I used to export as raw AVI, but HD space became an issue, so I now go with the vastly smaller MP4 H.264 in 1080p. (I have never tried the PNG image sequence route... not sure of the benefits?)

- As I build up the shots, I put them into my editor (Sony Vegas Pro) so I can get a feel of the flow of the film as I'm making it. Helps me identify what I might have missed when planning the film. Usually don't edit too much at this point.

- I generally have just the one Project file in Vegas, though I may create a second when it comes to SFX and VO and MUSIC tweaking. One project is easier to alter if I wanted to tweak the contrast/colour of the film, etc.

- When I have all the footage done, I edit and re-watch the film a billion times 'til it's right (or as close as I can get it!)

- Until recently, I would export the finished footage and bring into Sony ACID to work on the sound design, and then music. Now that I have Vegas Pro (and unlimited audio/video tracks) I found I can do the SFX in the same project. If I'm creating the music myself, I still export then use ACID, otherwise I bring it into Vegas and adjust it there.

This is not a hard and fast set of "Rules" on my part, but is generally the way I do things. I'm always open to trying new ways, especially if they work well for others.

https://marketplace.reallusion.com/iclone/author/biggstrek

The prime reason for me to use PNG sequences is that it is a lossless format with a smaller total file size than uncompressed AVI. It may in the end be a marginal quality difference, but common wisdom is that it's best to use lossless formats until final render.

Image sequences have the additional benefit that you could use a batch process on them in your graphics editor. Another advantage is that you can correct a single frame by rerendering the one frame in iClone, rather than the whole sequence. This was handy in IC 4 where trees wouldn't look right after a camera change.

---