5D Sound Solution by Mark Messenger at Crews.tv
The Canon 5D and other members of the Canon DSLR camera family are getting out on video shoots more and more, thanks to their image making qualities. But this creates a problem for audio. Recording through the mini jack microphone input to tracks that will be compressed on the whim of the camera is hardly going to create audio of a quality I would like to put my name to.
I recently landed a job, only to be told it was to be shot on the Canon 5D.
At first I wasn’t concerned. My thoughts were to simply treat it like a film camera and slate everything. But I was told the job was a behind the scenes, observational style documentary. Using a slate every time we rolled just wasn’t going to be practical.
Some form of double system recording was needed – without a slate. That’s when I began my search for a solution that allowed us to button on and off frequently, and at the same time provide something that the post-production guys could deal with efficiently.
Efficiency was very important – we were planning on providing 50 plus clips per day. This would require good metadata on each clip, so that the editors could quickly identify each one and marry it up to its corresponding sound.
It was clear that some sort of batch auto sync system would be ideal because it would mean the editors didn’t have to spend a lot time and money syncing each clip individually.
At first I wasn’t concerned. My thoughts were to simply treat it like a film camera and slate everything. But I was told the job was a behind the scenes, observational style documentary. Using a slate every time we rolled just wasn’t going to be practical.
Some form of double system recording was needed – without a slate. That’s when I began my search for a solution that allowed us to button on and off frequently, and at the same time provide something that the post-production guys could deal with efficiently.
Efficiency was very important – we were planning on providing 50 plus clips per day. This would require good metadata on each clip, so that the editors could quickly identify each one and marry it up to its corresponding sound.
It was clear that some sort of batch auto sync system would be ideal because it would mean the editors didn’t have to spend a lot time and money syncing each clip individually.
The easiest way to do that was to have identical time-of-day time code on each clip and corresponding sound file. But the Canon 5D doesn’t have a time code generator or time code input. This meant that the time code would have to come from the sound recorder.
This eliminated the Zoom H4 sound recorder and Sound Devices 552 mixer combination because an external time code source would be required. I wasn’t sorry because I had tested the Zoom recorder and found it very limited and lacking in functionality with quite a few traps – it’s a fairly crap recorder if you want to use it as professionally.
As well as that, the 552 mixer doesn’t seem to have metadata input and just increments each file number as 1,2,3,4…. etc. The chances were high that audio files would be given the same name across card changes and different days. This could confuse the editors and waste time in post – especially if there was more than one crew working on the same day.
This eliminated the Zoom H4 sound recorder and Sound Devices 552 mixer combination because an external time code source would be required. I wasn’t sorry because I had tested the Zoom recorder and found it very limited and lacking in functionality with quite a few traps – it’s a fairly crap recorder if you want to use it as professionally.
As well as that, the 552 mixer doesn’t seem to have metadata input and just increments each file number as 1,2,3,4…. etc. The chances were high that audio files would be given the same name across card changes and different days. This could confuse the editors and waste time in post – especially if there was more than one crew working on the same day.
After some thought we came up with another idea: to use Echo Digital Audio sync software. But this presented potential problems too. Again you would have to sync up each clip individually, and it appeared you needed to separately to press both camera and sound on and off button at nearly the same time.
The next idea was to use a fairly unknown technique of sending time code from the sound recorder and putting it down on the audio track of the clip within the camera. When the editors get hold of the clip they can use the “read audio time code” function in the “special” menu of Avid. They then use this time code as a sync reference for the separately recorded audio track.
The next idea was to use a fairly unknown technique of sending time code from the sound recorder and putting it down on the audio track of the clip within the camera. When the editors get hold of the clip they can use the “read audio time code” function in the “special” menu of Avid. They then use this time code as a sync reference for the separately recorded audio track.
I found that I had to pad it back quite a lot as the time code comes out a roughly line level and the 5D input is at mic level. If you sent it in too hot then there was a good chance that it might distort the time code making it hard for the Avid to read.
We did some tests and got the editors up to speed (something that you need to do before starting to shoot).
It worked great. The editors seemed to love it as it meant they did not have to waste time syncing each individual clip, they could batch auto sync saving the production money.
Mark Messenger is an Auckland film and television sound recordist.
No comments:
Post a Comment