Leave 30 FPS to 25 FPS conversion until last?
 

What are the arguments against leaving the Compressor conversion from 30 FPS to 25 FPS (PAL) until the final stage in an FCP edit? The advantages are obvious in terms of not having to convert unused footage. The edit material would not involve any significant layers or special effects, but would go on to be graded in Color.

If all of your clips have a similar feel (say, a romantic drama), do the frame rate conversion last. That way, you'll convert as little as possible, avoiding discarded clips. It also lets you do all of your creative work without breaking it up with technical worries.

This assumes that the technique you use has scene change detection. One thing for sure, you don't want to blend frames across scene changes. You want hard cuts.

If some of your scenes are unique (a fast moving fight scene within a drama), you might want to convert that separately, since you might choose different parameters.

The approach I used in an international project with a wide variety of clips and formats was to edit each clip alternately onto timelines A or B. That allowed me to extend each clip by three or four frames. After all of the edit points were locked, I converted every clip individually, including these additional frames. That allowed the final edit to have clean transitions.

If you do the conversion last, you might review every cut frame-by-frame. In the cases that you get a merged frame at a cut, go back to your original content, grab one clean frame and insert it over the merged frame. That will ensure that your cuts are 100% pure.

Thanks Jon for that very useful advice.

I think there's an option in Avid Media Composer to 'consolidate' an edit, which saves all of the scenes complete with 'handles' of extra frames as specified. If FCP has a similar option that might be the way to go. This should overcome any dissolves issues and leave the flexibility for future tweaking of the material (in 25 FPS) even after the final edit or conform. The end result should be similar to your approach of A and B timelines. I'll have to check it out further.