"Film Look" is certainly in the eye of the beholder ... so here's what works for me:
-- use the FilmLook.js script to deinterlace
-- modify contrast slightly using color curves (gentle S)
-- change saturation; depends on the look you're after (I ususally desaturate a bit)
-- add a *tiny* bit of noise (gaussian, animated) to simulate grain
-- use a mask track to letterbox
The last part makes a big difference in "percieved" film look.
Another route, of course, is to use an external program like DVFilmMaker (www.dvfilm.com). It's inexpensive and works nicely.
This may be kind of a dumb question and for sure subjective, but what do you consider the "film look" to be? In other words what is it EXACTLY that you don't like about digital that you're tying to remove/mask by making your project look more like film?
As others already said for me, just a slight reduction in saturation (try the range.925 to.975) on the Color Corrector or tweaking the older levels filter slider and sometimes a little blur makes the video look less harsh and just a bit softer. Or too the older HSL filter, cutting back saturation just a bit.
Thrillcat: a few effects like "glow" or "light rays" might overflow the limits of your frame and affect these black top&bottom bars you have in 16:9.
Therefore, using a mask ON TOP of it all really *cuts* your picture, *effects included*... you then get a clean, unaltered black frame (be it 16:9 or other size: you can decide how far you want to go. And you can even use some color of your choice instead of these "standard" black bars...).
Personnally, when I edit, I use red bars and pull down the level to around 50%: I can instantly see what I'm cropping if I use different media like stills + 4:3 + 16:9 stuff. At the end, I put clean black bars...
You can pre-Pan/crop or post-Pan/crop the filters if you play with the little arrow on the filter's keyframable properties track. There you can assign if the filter is pre-crop or post-crop.
I'm with Futz -- I find it easier to use a mask track because I have two versions of my mask: one uses red letterbars set at 50% transparency, the other uses black bars, of course (both are PNG files). I put the red bars on track one and can then adjust other tracks [with track motion or pan/crop] if needed to better frame shots within the viewable area. Once everything is setup, I drop the black bars onto track one as a take and use it for my final output.
Because I like to set the background color in preview window in black therefore it's a little bit easier to spot-check the framing. And at 50%, the hue provided by red is also more "slicing" through most images I have in my shooting.
Well, I guess it's kind of "each one its own"... :/