Stupid Sync Trick & Tests

[Elliot Lee]

Hi,

Just thought it'd be neat to share some results from a bit of testing I did today. I'm planning on using XH-A1's to shoot a wedding in a couple of weeks, and I didn't want to spend oodles of extra money on XH-G1's and Ambient Lockit's, so I'm testing to see how many issues I'm likely to have with syncing the footage between cameras.

Stupid timecode syncing trick (not my invention, but worth repeating):
So you don't have a fancy G1 with the genlock & TC connections, but you want to get your XH-A1's sharing the same timecode. To do this, go into the timecode menu (usually custom key #1) on all of your cameras. Then line them up facing the same direction, and use the remote control to set up free run timecode on all four simultaneously. Once you finish setting the starting timecode for free run (through the remote), all the cameras will be in sync. Theoretically... :)

Using this starting point, I tested to see how well the cameras would stay in sync in various conditions, as measured by using a clapper. The first test I did was recording with all cameras for fifty minutes. The second test I did was turning the cameras off for a while, then turning them back on briefly to record another clap. After this I imported the resulting footage into FCP and put markers on all of the claps for both tests.

What I found:
On the recording test, the timecode stayed in sync to about a frame on 3 out of 4 cameras. The 4th camera's timecode for the ending clap was exactly 3 seconds off (I verified that the starting clap had the same timecode as all the other cameras). Because that camera was much less out of sync for the sync-after-shutdown test, I suspect something really odd happened here (tape dropout?). The seconds-and-frames numbers of the clap on the four cameras were:
40:19, 40:20, 40:20, 37:21

In the sync-after-shutdown test, the cameras all stayed in sync within just over a second or so. The seconds-and-frames numbers of the clap on the four cameras were: 56:18, 57:05, 56:07, 56:16

What I didn't test:
. How environmental conditions affect the sync. For example, would moving one camera to a warm place, and another one to a cool place, cause them to drift out of sync?
. Whether sync drifts much when the camera is on but not recording. (I.e. is that an acceptable way to keep the cameras closely in sync)
. Whether removing the main battery from the camera contributes to drift. I suspect this is true, because the camera that drifted the most in the sync-after-shutdown test was the one that ran out of battery shortly after the recording test. But it'd be worth verifying more directly.

My conclusion: If you just have to have tightly locked multicam timecode, you'll have to pay a lot of money. If you want a setup that has a good chance of staying in sync for a less demanding multicam shoot, this trick should work fairly well.

Hope this helps,
-- Elliot

[Pete Bauer]

Not sure what happened with the fourth camera; maybe it didn't "see" an IR signal for a couple seconds or something. But yup, this is definitely a useful trick for multi-cam that doesn't require dead-on genlock. Some further info at:

http://www.dvinfo.net/canonxl2/articles/article11.php

and

http://www.dvinfo.net/conf/showpost....8&postcount=43