From: David Tetzlaff (email suppressed)
Date: Fri Feb 18 2011 - 13:11:46 PST
> Most recently I was accepted by a festival, only to learn that they do not show
> 16mm film and if I spent money on an expensive transfer they would show it.
A festival not taking 16mm may be a bit surprising, but there's a kind of logic to it the way things are going. On the other hand, demanding an _expensive_ transfer is inexcusable. As I wrote here awhile back, festivals need to lower the technological/economic barriers to makers by accepting work in a variety of formats that can be generated by end-users w/o expensive lab services. I offered to help any exhibitor set-up a working system, and had no takers. Perhaps, if they get thousands of entries anyway, they could care less who or what is being excluded and why.
If the expense in your example would have come primarily from getting something that only exists in 16mm into an electronic form, as I noted in another post I sent earlier today, all you need to do an acceptable telecine is a standard 16mm projector and a 24fps-capable video camera: for example a Panasonic DVX100 for NTSC, or a Canon XG-A! for HD. (Anyone connected to a film-art scene in a decent-sized city should be able to borrow one of these from some kind person...)
Use a matte white projection surface. It doesn't have to be a screen per se. I use poster board. Darken the room.
Position the camera as close to the line of the beam as possible to minimize keystoning. This generally means putting it immediately to the right of the lens housing, where it will tend to block the projector controls. You'll need a sturdy tripod that will resist moving when slightly bumped, and a head that lets you get the tilt not-quite-locked so you make tiny bump adjustments to the framing before locking it down. If you don't have a ball-level, a three-way photo head is good as it makes lining the horizen easier than fiddling with the legs.
It's a matter of choice whether you frame the camcorder so the film image fills the viewfinder (this will invariably create a small amount of crop), or leave a little cushion between the edges of the film and video frame. I leave a cushion since you can't get back what you don't shoot. If you're shooting to HD, you'll be fitting the 4:3 film image inside the 16:9 video image, leaving pillarboxing on the sides (which you'll whack off in post)
Run the projector w/o film, and set white balance on the camcorder based on the light reflected from your screen.
Use manual exposure on the camcorder. Getting the right adjustment can be tricky as the exposure latitude of the film will be such that parts of it will likely be too dark for one video setting, others too light. Avoid bloom-out (use zebra stripes or a waveform monitor if you have one... there's software for this). For a reasonably short film it's easier to bracket exposure and do several full takes than to stop and start making adjustments.
If you can get a live feed from the camcorder into the computer, via Firewire, you can capture the footage directly without putting it on tape in-between. (Avoiding tape has a couple advantages). In general, you'll be capturing using the default codec of the camcorder (DV or HDV) since the camcorder is doing the digitization and the computer is just capturing the data stream.
Once the film has been captured in the form of a digital file or files, edit, adjust the image as necessary and so in inside your video software. Pull in the sound track from the final mix you used for the film. It's easy to adjust the duration of the digitized audio, in chunks large or small, to restore sync if it doesn't line up right at first.
Create a master of the finished digital version, and then you can convert it to a variety of forms for distribution: burn to DVD, Blu-Ray, HD-DVD, bump back to DV or HDV tape, create Quicktime or AVI files in an assortment of codecs and/or resolutions...
(If you don't already have software, there are cheap or free tools available to get all this done. You don't have to shell out for an Apple, Adobe or Avid Suite license by any means.)
Doing all of this is a bit of a job, yes, but nothing compared to the kind of work that actually goes in to shooting and editing a 16mm film in the first place. You don't pay fees, you don't wait on the labs schedule, and you control the results.
Of course, the image isn't as good as a scan, and I suppose for some kinds of work it just wouldn't do. However, having looked at scores of student pieces done this way - shot on 16mm, telecined as above, finished in FCP, screened on a 3chip DLP video projector - I cannot recall one instance where I found the transfer quality wanting.
FrameWorks mailing list