October 1, 1997

I spent an unnerving afternoon checking the movie's synchronization at Rich Cohen Productions, a little post-production company recommended by Fran at Zounds. Rich lined up the answer print and the magnetic track on a flatbed with a dim little screen and left me there to watch the film. It's rather difficult to detect if you're one or two frames out of sync under any circumstances, and especially so under such adverse viewing conditions. I concluded, rather shakily, that the first reel stayed in sync; I thought I saw some little problem in the middle of the second reel, and by shifting the sound track around I sort of decided that the sound might be a frame too early. Nothing tragic, I guess. The third reel looked okay to me for most of its length--then, near the end of the film, came the first moment where it was easy to check sync, when James Vidos slams a deck of cards down on a table during a poker game. Lo and behold, the sound was clearly two frames ahead of the image.

What to do about all this is hard to say, especially as I trust my perceptions so little under these circumstances. To the best of my ability to determine it, the third reel starts in sync and drifts gradually until it is two frames out at the end. The other reels might be the same, and then again they might not. Cutting the mag track to remove a frame or two will be difficult and expensive, and I don't have a clear sense of where the sync is lost, or even whether it is lost on the first two reels. Rich, a post-production veteran, doesn't think the problem is too severe, and advises me to tell the lab to retard the entire sound track, on all reels, by one frame. The idea would be to split the difference on the two-frame drift, so that the sound is one frame behind at the beginning of the reel and one frame ahead at the end of the reel, and hopefully imperceptible on either end. Of course, there is the possibility that reels one and two aren't drifting the same way as reel three. But Rich says that the sound lagging by one frame isn't so bad a thing, and somehow looks better than when the sound is one frame in advance. I have to think about this, and get advice from the sound lab.

There is, of course, the issue of how the sound managed to drift, when the process is supposed to be so regulated at every step. I think that the track was in sync when we we made the DAT tape at the end of the mix sessions, but I wouldn't want to bet a large sum on it at this point. It probably won't be worth the effort to find where the problem crept in.

Click here to read the next diary entry, here to read the previous entry, and here to go back to the main menu.