Intermediate Method (part 1)

Let's now study the way most films are made today :

  Strangely enough, films that look very complex on the screen, such as The Matrix, are shot in most part without time code. The result, though, is highly professional isn't it ? Just like thousands of other films shot in the same conditions. The editing of such a film would be painful, except for masochists, if it were performed on a traditional editor. You'd better use a video or virtual digital editor for that.

Don't worry, again, I'll explain...


Nothing changes (regarding the previous case) on the shooting.

A slate is performed for each shot, as we've explained in case n°1, to enable the re-sincing of picture and sound in the editing process.

What happens next ?

Picture process

.

Sound process

The négative is developped.

The takes that the director decided to print are printed in positive, and will be shown as dailies during the shooting.

The negative is then set aside, it will be manipulated again once the editing is done.
The takes that the director decided to print (their sync sound), as well as the wild tracks recorded on the production recordist's own initiative, are transfered to a Direct to Disk audio editing system (PC, Mac or dedicated Workstation).

To gain time, the sound editor can already sort the wild tracks he'll use or not, isolate the set silences, etc...

A video transfer of the chosen takes is given to the picture editor.

It's the key to any modern motion picture production.

The object of the video transfer is :
*   to enable the editing on more flexible systems than the traditional film editor : video (VTR or DtD) and audio (DtD) editors.
   
*   to link the pictures and sounds to a single time code, to end the post-production on a common basis.
     

  It is a transfer of the chemical picture (photographic film) to a video medium. During the transfer, a time-code is striped alongside the pictures. The TC start adress is arbitrarily chosen by the lab. The relations between this time-code and the original negative's key-codes are saved on diskette by the lab. Each video tape (duration : 60 or 90mn) has it's associated disk, the whole package been returned to the production company that asked for the transfer.
 
  NOTE : When you look at a film on television, or on video cassette, it is a new transfer you're looking at, this time of the completed film, once it has been released in the theatres.

As long as the sync sound of the shooting hasn't been synced back to thses pictures, the editor can't move forward in his work.

For the following processes, the sound editor needs
-   either a VTR (video tape recorder) of the same format as the video transfer, to read these tapes...
 
-   either a transfer on hard disk, of the time-coded pictures of the video transfer, in a format his audio DtD editing system can read (Quicktime, for example).

We'll study the case where the work is based on a VTR, which is still the most frequent case, and by far, the most complicated.

Picture process

.

Sound process

.
The time-code of the VTR (hence of the pictures) is fed into the TC input of the DtD system.

As soon as the VTR is put in playback, the DtD locks to it, at the same adresses. Ideally, the DtD should have a VITC input, which will considerably increase the speed of syncing the sounds, as we'll see below...
 


The slate of the first shot on the tape is located, using still and slow-motion via the "jog/shuttle" of the VTR, or, delicious luury, by remotely driving the VTR from the DtD via "9-Pin Machine Control" (an almost universal transport command protocol, at least in the video world, also called RS422).

Two possibilities can present itself :
* The DtD understands VITC. In this case, when the VTR's playback head is precisely located on the frame where the two sticks of the clapper-board make contact, the corresponding time code adress is constantly sent on the VTR's VITC output, towards the DtD, who will then place his own "playback head" at that same adress.
* The DtD does not understand VITC, it can only read LTC. In that case, when theVTR's playback head is precisely located on the frame where the two sticks of the clapper-board make contact, the LTC cannot be read by the VTR (since the tape doesn't move !). Luckily, the human being in front of the VTR (that's you !) can read the corresponding adress that's been burnt-in at the bottom of the screen by the lab. You just have to locate the DtD's playback head to that same adress, by typing it in manually.

As we've just seen, there are two human interactions necessary with the use of LTC (visual reading of the adress / manual typing of the adress on a keyboard) we'd rather do without when there are hundreds of shots to resync : hurray for VITC !

NOTE : For a DtD that can only understand MTC (Midi time Code), which is generally the case with cheap DtD systems, there are inexpensive little black boxes that convert LTC into MTC and vice-versa. You can then work with this DtD as you would have done with the LTC system described above. But beware, these small DtDs have other drawbacks as we'll later see, I doubt they can autoconform sound for example
The sync sound recording of that shot is listened to : to check that the vocal announcement corresponds to that of the clapper-board inscription. Then the clapper's noise is located :
either in audio, by scrubbing to the exact position of the impact...
either visually (which DtD is no longer capable of displaying waveforms these days !?!?!) by spotting the typical transition of the clapper closure in the waveform.

Any DtD that's worth buying will enable you to place a marker at any position within a sound file. This marker will of course be placed above the previously spotted clapper closure.

Any DtD that's worth buying will also enable you to insert the audio sound file into the playlist * so that it matches the playback head position of the DtD.
  * We'll later see what an EDL is. For now, the term "playlist" may be more familiar to you. Let's say, to simplify, that it's a group of files, with their specific playback parameters.

And "voilà" : the first shot is synchronized. If you rewind the VTR a bit, and playback, the sound will play in sync with the picture. On with the other 1499 shots of the film !!!!!