Thinking about this workflow
 

For my next project (about 75 minutes) I was wondering: how to acquire the sound?

I came with this workflow:

The boommic will be connected to the Zoom H4. Everyday at the end of shooting the sound will be put on a harddisk. It will be archived and named, so it will be easy for the editor to pick the sound.
For reference, the build-in camera mic (xh a1) will be on, so the editor can sync the great audio, recorder with the A1, with the video.
That is: after he's done cutting the video.

Is this a good workflow? Please let me know.

Ivo:

Use a sound report everyday. Fill it out, use some type of slate and call out each scene. Make sure the information is written down immediately if possible. It's always more painful going back and reentering the information. Put the sound report in a safe place. If you use a computer, you can always enter the sound report information in a spreadsheet and then generate a PDF from it. This can be stored with the audio files. But make sure you have the sound department fill out the sound report and use a slate. I know that sounds odd, but I've shot with directors who didn't like to slate. Later when it came time for them to do post production, they simply didn't use any of the 24-bit audio I had on my Deva, instead they used the audio on their tape I fed them because they didn't have any idea how to piece things together without a slate.

Otherwise your workflow sounds OK.

Wayne

Agreeing with Wayne in all respects.

Also, the H4 will give you sequential filenames only. A sound report that matches filenames to scenes, shots & takes is essential for the editor. You can't do that at the end of the day, it has to be done as you go.

The H4 sync runs off a little. A clap at the head and the tail will make post sync much easier on longer takes.

Thanks for the advice. But regarding the takes won't be longer than 5 minutes (it's narative film), is a clap at the end necessary?

Hi Wayne, when you say use a slate, are talking about a clapper board, whilst reading it aloud?

Also how do you do the report? I have ben struggling to find time as a one man band to do it but fear for what post will do to my work if I don't.

Cheers

Yes, the clapper board is the slate, and again, yes you would have somebody (not you) read it aloud:

"Scene 4a, Take 3" then clap it. You obviously would be rolling prior to this.

As far as reports go, I cheat a bit because I use the Deva, so all metadata is entered as I go along. I enter the scene and the unit automatically counts for me. Now I have gotten into trouble when I have forgotten to reset the scene number. Also things like circle takes are automatically placed into the metadata if I press the circle take button on the recorder. After I'm done for the day, I mirror all the audio on a DVD-RAM disc and put that into a computer. Using BWF Widget, I generate a sound report as a PDF file, put it onto the disc and turn that in. A sample report can be seen here (this one is a bit crazy because we did both English and Spanish, so the same scene was shot twice, once as 4 and once as 4sp):

http://homepage.mac.com/wayneb/DSHS_Day3.pdf

You can certainly do sound reports manually and Location Sound was nice enough to provide a PDF of their sound report at one time, which others have archived on the web. You can find it here:

http://comm.unc.edu/mmlabs/helpfold/soundreport

Using this sound report, have a PA fill out the scene and take numbers for you. You're right, as a one-man operation, it's hard enough to keep things in order, so you do need help. But the reality is if you don't keep track of audio, they certainly aren't. And guess what? When it comes time for the post production of the project they will use the easiest audio available, which of course is what they already have on tape.

Nearly every camera I've worked with has audio which pales in comparison with what I can give them from the Deva. In fact, some of the rental cameras I've dealt with have some serious audio issues requiring the Deva audio. Last night, I got a chance to watch a Nova program that I worked on, they had issues with the camera audio and so my backup audio on the Deva was required. Ironically, I offered them a disc before the producer left town, but he didn't want it. A month later when they had things in production they found some glitches in the tape, so they called me for the audio. I had to overnight it to the production house (nearly 6 hours of audio!)... The ironic part of this story is 6 hours of interviews and experiments with apes ended up being less than 3 minutes of the program... I guess you never know what they will keep and what they will toss away.

Anyhow, hope this helps.

Wayne

Strictly speaking not necessary but it can be a very big help. Sync drift is always a potential problem, especially with consumer equipment like the Zoom recorder. With a sync clap at both the start AND end you have two precise reference points for length on both picture and sound should you need to stretch or compress the audio clip to keep it in sync with the video.

Saw that show last night .... fascinating stuff, congrats on your piece of the work. Hoping to get a chance to work on that sort of gig someday!

I sent the producers a nastygram last night as I was left off the credits. They may have only used less than 3 minutes, but they did use some of the material. But hey, at least they paid me.

Wayne

How do you sync sound on the last clap? Do you have to change speed?

If it's possible, I would use an audio recorder that at least has a time of day clock for the files. The H4, unless something has changed, can't even set a specific date and time when record was pressed to associate with the files. If you use a recorder that can accurately record the date and time, along with the careful sound reports and slating already mentioned, that would be very helpful in matching the scenes and takes with the camera's time of day clock later on.
In a typical production, scene, angle, and take info can become very confused if you are shooting quickly. Generally I prefer to have the camera's timecode called out verbally instead of the scene or take number. That combined with matching time of day clocks and the sound reports can keep things from becoming confused later on even though you're not using actual timecode on the audio device.
As long as you have good guide audio on the camera and slate every take, "syncing" isn't a major problem. Keeping the files organized and matched up is the real difficulty.

A lot of the better NLEs have feature that can change the length of a clip, making it longer or shorter, without changing the pitch as a simple speed up or slow down would. Of course there's a limit how much you can get away with. So you line up the clap with the picture at the start of the clip and then scroll to the end and see if the picture of the sticks coming together in the tail slate matches the clap on the audio. If it is, you're in sync through the whole clip but if it's not, you can see how many frames it's off and that'll tell you how much you need to change its length to make sure everything is in sync from start to finish.