Quote - You are best to render individual stills and then compost them using a video editor.

Directly rendering to a video format is a problem if the render hangs. You would have to start from the beginnig of the render sequence.

Rendering to picture files, you just pick up where you left off if there is a crash.

I hear this a lot and I find it to generally be true.  It's nice to be able to resume rendering if there is a power outage or crash. 

That said, I have been working on a project for a client in which Poser's lost ability to save videos in Xvid has been greatly missed.  The animations each took about an hour to an hour and a half to render, not a big deal if I needed to render one over.  But the space requirement was mind boggling.  For example, one sequence contains 90 png frames @ 872kb each, or 76.5MB of data.  The animation file compressed with Xvid is only about 240k.  That means the entire animation is less than a third of the filesize of one single frame.

Also, you have to figure in the time spent converting each set of frames to an animation and then deleting the frames.  I'd say rendering to single frames is best when each frame takes a long time to render and/or if there is a need to post process them in another app.  When creating a lot of short sequences that need no frame-level post production, rendering clips can actually be a more efficient use of time and storage space.