Reliability of Jam Timecode

E

Exact

Veteran
I've just done some testing with a 3 camera setup using jam timecode.

The first test everything was sync'd perfectly.

The second test (after re-sync) two cameras (both slaves) were 1 frame earlier than the Master.

The third test gave Camera 3 2 frames early and Camera 2 1 frame earlier than the master.

Is there any specific sequence to follow to get better accuracy?

The second two tests were done over about 10 minutes and were the least accurate. The first test (where everything stayed in sync) was about 30 mins!

Any ideas? I have a 3 camera shoot at the end of this week and I would like to get this nailed if possible :)
 
Smifis said:
Jam timecode?

Whats wrong with a clapper board nowerdays?
Click to expand...

Not possible in this instance without having the cameras rolling for 20-30 mins before the event starts due to where the cameras are positioned. They can't see a common position :)
 
Who says you have to do it at the start :)

What I've found myself syncing up cameras all over the place.

Lets say two cameras are withing audible range of a common object that makes a distinct sound, lets say, someone dropping a glass. sync those two cameras up and then with the last camera, find another point.

There are so many creative ways of accomplishing a 3 camera shoot without getting sync issues. :)

It all comes as part of being zero budget all my life.
 
Thanks for the replies - I've been using alternative methods for a long time ..... but .....

Since the HMC 'has' this facility (my previous cameras didn't) I'd like to try to use it..... so is there anything I could be doing wrong?
 
The thing you might be doing, is either turning one of the cameras off, or toggling between camera and playback mode. Doing either of those things will throw the sync off by a few frames.
 
Barry_Green said:
The thing you might be doing, is either turning one of the cameras off, or toggling between camera and playback mode. Doing either of those things will throw the sync off by a few frames.
Click to expand...


Thanks for the reply Barry. In the very first (30 minute) test I did indeed change battery (as a test!) and it maintained perfect sync. My reaction was Wow!

The second two tests, both of which resulted in being out of sync by one or two frames, the record button was pressed as soon as I exited the menu for slave sync on each of the slaves and on the master as soon as the cable had been disconnected and the menu exited. So, on these two I know there was no power off at all. Strange....
 
Exact said:
I've just done some testing with a 3 camera setup using jam timecode.

The first test everything was sync'd perfectly.

The second test (after re-sync) two cameras (both slaves) were 1 frame earlier than the Master.

The third test gave Camera 3 2 frames early and Camera 2 1 frame earlier than the master.

Is there any specific sequence to follow to get better accuracy?

The second two tests were done over about 10 minutes and were the least accurate. The first test (where everything stayed in sync) was about 30 mins!

Any ideas? I have a 3 camera shoot at the end of this week and I would like to get this nailed if possible :)
Click to expand...

Well no camera keeps accurate timecode and don't believe anyone that tells you that they do.

For some reason, they continue to use cheap and inaccurate clocks. If you had a watch that kept time as bad as they do, you'd throw it away.

You didn't say, but did you re-jam the timecode between your second test and first ? Or did you jam the cameras, record right away. And then some time later recorded a second event ?

The cameras will drift out very quickly. I've seen it happen within few mins.

The onyl way (expensive) way to guarentee timecode is to use a more accurate external timecode generator that you jam and then leave connected to the cameras.

Such as this

or this

With one of these connected you should be able to get less than a frame drift over 8 hours.

The good news is that with data based cameras, they only look at the timecode when they *start* recording. So if you have a 1 hour event, they check the timecode for the first frame and then frame count within the file to work out the rest of the timecode.

So if you jam and can start recording within a few minutes then you should be fine.

jb
 
John Brawley said:
You didn't say, but did you re-jam the timecode between your second test and first ? Or did you jam the cameras, record right away. And then some time later recorded a second event ?
Click to expand...

I re-jammed the time code between each test and started recording immediately. A simple hand clap was used to check synchronization, and this was within a few seconds of starting to record on the third camera - so maybe a minute or two from start to finish.

I understand that time code can drift, but I didn't expect it to drift by 2 frames within a minute or two. That was my concern really. If it had drifted by two frames over 8 hours I would not have given it a second thought - but over a couple of minutes I would have expected it to be synchronized perfectly.
 
You must log in or register to reply here.
Back
Top