Intermediate Method (part 2)

Once all the sync sounds of the video transfer tapes have been resynchronized... Abracadabra ! Even though the shooting has been done without time code, the pictures have been linked to an arbitrary time code by the lab, and now, the sync sounds are positioned at the same time code adresses... Just as if the shooting had been done with time code.
But the post-production process is far from being ended.

Picture Process

.

Sound Process

.
The sync sound recordings must now be transferd on the longitudinal audio tracks (A1 et A2) of the video transfer tape, so that the picture editor can move on with his work.

To do so, the analog outputs (stereo) of the DtD are connected to the A1 and A2 inputs of the VTR. I recommend you use the Dolby NR function of the VTR to make the sound a little better (it's a type C noise reduction system).
 


     
 
An IN point is programmed on the VTR, a few frames before the first sound element to be recorded. The edit mode should be INSERT + A1 and A2 (in ASSEMBLE mode, you'd wipe out the video !!!)  

 

view of the edit mode selector

  All you have to do now is press the EDIT button on the VTR. The machine will perform a pre-roll (rewind approximately 5 to 10 seconds), then go to plaback mode. the DtD will then lock to the incoming time code, therefore in sync with the pictures. At the IN point time code, the VTR will go to record mode on the two longitudinal audio tracks. If you'd checked your recording levels before the transfer, you can now go for some coffey while the sounds record themselves onto the video tape.

NOTE : You're supposed to record a 1 kHz sine at 0 VU on the audio channels during the color bars at the beginning of the tape so that the picture editor can align his equipment, or simply as a reference for any future application from that tape.

It's child's play to generate a 1 kHz sine wave from the DtD, and to set its playback level so that it matches 0 VU on the VTR.

The problem is that this 1 kHz had better also reflect the RMS level of your sync sound, so that it two, can be correctly recorded. Since the dB FS * value to which the 0 VU corresponds is arbitrary in the digital world, all I can do is give you an average value, a value that should work most of the time.

  * Full Scale : the negative scale typical of the digital world. It is scaled in negative, from 0 downwards, because it is the highest possible electrical level the A-->D or D-->A convernters can represent..

If you generate your 1 kHz between -8 and -14 FS from the DtD, so that it drives the VTR at 0VU, everything should go fine during the subsequent audio recordings. Indeed : most of the motion picture live sound recordings are very dynamic. If the peaks have been recorded at 0 dB FS on the DtD (which should be the case), the average level (RMS) of the signal should hover between -8 and maybe -20 FS in worst cases.

The picture editor can at last get the video tape(s) back to start working on the editing. Since several days, or even weeks, can go by before his work can in return be used by the sound editor (the sync sound will have been manipulated along with the pictures and will therefore have changed positions !), what to do with a DtD hard disk filled with hours of sound ?

The ideal solution is the removeable hard disk drive. I am against the use of hard disks for long term memory applications (more than a few months) but here, since the manipulation time window is rather short, they really are perfect. The session (all the sync sounds that have been resynchronized and the wild tracks) can entirely be contained on a set of disks (maybe even one big one) and set aside while you work on other projects.

Ideal solution n°2 : a backup on MO disks.

Let's admit you don't have a removeable hard disk drive (poor bastard !!!).

then you have to "empty" the content of the DtD's disk drive to a medium capable of recording audio and time codeb at the same time. Since you've painfully managed to resynchronize the sound with the time code of the video transfer tapes, it would be a pity to lose all these time code relations once again.

  What you need for that is a time-code stereo recorder (or a multi-channel one, if the sync sound had been recorded in such a format). And if possible, to avoid unnecessary conversions, prefer a digital recorder. It can very well be an analog recorder, like a 1/4" central time-code recorder for example, but the sound would go through two conversion stages : D-->A and, later on, A-->D... Not really what we're looking for nowadays !

This machine will record the digital output of your DtD as well as the painstakingly found time code adresses. This medium will later on enable the "re-loading" of the DtD's hard disk drive with the audio and its time code adresses, as if nothing happened, as if you'd stayed on the same project in the meantime.

Note that both the sync sound transfer to the video tape and the "emptying" of the hard disk drive to the "backup media" can be done simultaneously. (see drawing on right)
 
       
  The medium obtained here should be called "autoconform medium". As it is most of the time a DAT tape (yuuuck !!!), they are therefore often named "autoconform DATs".