3

Hey guys,

Basically all the work I have done so far in University has been post production audio. I take clips from films, strip out all the original audio, and try adding it all back in.

I've now decided the time is right to venture into the production aspect of the film and join forces with another section of my university "Film and Media". I'm a Masters student and I have offered a Masters Film student the great thrill of having me as their post-production sound designer. However, to advance my skills I really want to be involved with the filming also!

I have a shotgun mic, a DAT, XLR cables, Pro Tools, MBox2 etc etc BUT my main question is. .how do you actually go about shooting?

Are the audio and film always two separate entities?? (we won't be using 3.5mm jacks on the HD camera as we want amazing sound). Is it just a case of before every single take we start recording video (hitting record at exactly the same time - him on his camera, me on my DAT), and audio and shout "TAKE 5" and then line up the audio and video in Post production?

Thanks a lot for any advice!

3
  • Hello again. Ok the project is nearing it's beginnings and I have a script. Now before I plan the post production I'll share the technique being used (due to budget mostly): Boom Mic mixed with SQN and recorded onto the Camera. So do you have faith in me to get some good dialogue recording?! I hope so. Does anyone have any final tips they could offer if they have used this type of practice? Obviously I'll be using some other mics to get room tone etc while ensuring dialogue doesn't overlap sound effects (trying to ensure nothing but dialogue at any one point to keep post production clean). Tha
    – user464
    Jul 27, 2010 at 20:52
  • Yeah I was a bit nervous too, but after doing two locations you're starting to see what's constant in the environment and what changes that you have to adapt to. I'd suggest you capture ROOM TONE with the boom mic in a similar position as the one you had for the take. If you point it in a different direction you'll pick up a differently coloured environment. If you can, do re-takes if you think audio is not satisfying or do wild lines on the side with just the actors. Make yourself proud of your recordings :) Jul 27, 2010 at 21:26
  • Out of curiosity, what mic will you be running with? Jul 27, 2010 at 21:26

11 Answers 11

7

Wow wow wow! One thing shocks me horribly:

"Is it just a case of before every single take we start recording video (hitting record at exactly the same time - him on his camera, me on my DAT), and audio and shout "TAKE 5" and then line up the audio and video in Post production?"

Don't shout, voice is not transient enough. Clap with your hands at least, if not a clapperboard, but make sure the clapperboard is on camera. In post, you'll then sync the peak (clap) in the audio file with the frame where the two hands (or sticks) hit each other. Also, whatever you use as your clapping machine, it should ideally stay closed after the clap and not rebound, otherwise there might be no frame, in your video, of the two clapping things touching. There might be a frame just before and just after (when it's not fully closed) and this makes syncing harder. If your clapping device remains closed after the contact, then you'll have a clean visible clap to sync your audio to.

Second, I'm no authority in this discipline, but the 1/4 inch jack vs XLR thing shouldn't make much difference. A TRS jack offers the same balancing as an 3-pin XLR.

Now booming (cause a boom should be your main mic, radio mics as a backup if you're lucky enough to have any). From the Latin booma, boomae, boomorum...

  • Ideally you'll boom from the top down rather than the bottom up. That way your mic is further away from the chest and closer to the mouth and you get less low freqs from the chest.
  • Try to point your mic not at the mouth directly, but rather about 6/8 inches down on the projection axis. Give the lows enough space to get formed.
  • !WARNING! If you're shooting a CU but the scene also requires a wider shot, you don't want to get your mic any closer during the CU take than you'll be able to get it in the wider shot. Otherwise you'll create an huge change in perspective (really close vs quite far) in the audio for the same scene. You'll curse your fool self in post. !WARNING! Nathan disagrees and has a valuable reason (cf. his post further down).

I'll edit this as it comes to me! Feel free to ask questions :)

EDIT:

  • I assume you'll be the only audio guy there. And if the shoot is not too long or too intense, you'll likely remember what take was what and you'll know your material when postprod time comes. But don't take any bet, try to keep track of each take using a Sound Report. A sound report tracks, in a table, each take by providing sensible info: **scene/take number**, **file name** (if appropriate), **comments** (f*or this one I'd like somebody else to take over as I've never had the chance to write anything interesting there...*)
  • Regular backups could save your butt (twice a day if possible).
  • If you're recording in noisy environments (city traffic, air con...), do wild lines with the actors. Find a spot around where you are that's quieter (but that still have the sonic character of the actual set) and record your actors without the camera, without moving. This saves you from doing ADR, and it's already matched to the audio from the takes since it's the same recording environment.

You also might want to check out John Purcell's Dialogue Editing for Motion Pictures, cause all that changes from what you did before is that you'll be working with production sound this time. So document yourself properly on these, and I recommend John's book, but as I said I'm not an authority, so you might feel reassured if I say that many people do recommend it ;)

SECOND EDIT: I just remembered that Iain McGregor keeps telling me to clap at the start and at the end of the take to optimize your chances if ever your devices were to go out of sync!

4

Sound Reports were mentioned earlier. I second that. In Post production, you are going to rely, more than ever, on proper labelling of takes. Make sure that the name you give a scene on the clapboard is the same name you give the audio file. Otherwise, in post, your editor is going to have a hard time matching sound to picture. Worst case scenario, you'll be the guy digging through your audiofiles to find the ones that match the edited scene. That should not be your job. Your video editor should be able to quickly find the appropriate audio for the scene based on the file naming convention you have setup before the shoot.

4

Your a brave man, Aaron. Fortunately, I think location audio requires a good set of balls anyway especially when you need to tell the director that the awesome scene he just filmed was ruined by the garbage truck that just pulled up down the down the block. So don't get too worried. You're saving yourself a lot of headaches just by thinking of these issues ahead of time.

Since you have just one Mic for the shoot, do you have a wingman? Someone who is also familiar with audio that can operate the boom while you watch levels on your mixer? You can do it solo, of course. But it helps to bang these issues out with someone of equal or, perhaps, lesser skill. Be mindful of how far you need to be from the actors to get consistently good audio. If you're behind the camera and the actor(s) are starting their dialogue as they walk into frame 15-20 ft away, then you're going to get more of the reverberation from the room before you get their direct voices. Many new filmmakers do not appreciate how sound is captured so be prepared if you get a couple quizzical looks as you figure out a way to manuever the your shotgun to the actors without getting into the shot.

OH and don't leave anything to post! Try to capture all the audio as best as you can during the production. Don't leave anything to ADR. ADR should be a decision that's made in Post not during production (ideally).

Good luck on your shoot man. I wanna know how it goes.

3
  • Thanks for all that info! Yeah I'm NOT doing any ADR AT ALL. . .I'm making sure all the dialogue is PERFECT!! It's the other little bit's that I'll redo that I mightn't catch with the single shotgun mic (I'll be holding the boom and monitoring on headphones!) such as the odd little sound effects that are between dialogue and not on axis with the mic (so I'll do over the original sound with some room tone then dub in the sound effect . . I hope!). You raise yet more concerns with me about movement of the actors. I'll have to sit down with all of my directors for each of their . . . Jun 23, 2010 at 22:35
  • Projects and tell them that we need to plan each shot and imagine where I'll be able to stand to keep perfect recording as the camera moves without the boom dipping into the shot! (I want all of this pre planned so that shooting is relatively straight forward). I am very nervous but I'm sure once I meet the directors and we plan the production and discuss what direction the film is going in, where it's going to be shot, how long until filming, and then eventually the location planning I'll be ready to just go to the set and carry out our strict plan! I want most of the shot already in my head! Jun 23, 2010 at 22:37
  • Still very nervous about it though. If I need any more little hints and tips I'll be sure to come back here before we actually start shooting (it's very much in the pre-production stage now with script and planning etc) as I'm sure I'll think of more questions/problems that I may need help on before shooting :) Thanks for all the help so far though :) Jun 23, 2010 at 22:39
3

Hello Aaron,

I've not done much in the way of film shoots myself, however, I have worked on numerous shorts. What camera and format is being used to shoot the film? Tape formats such as Digibeta or HDCAM offer excellent sound quality, they are used in mastering for film/television. Just because the sound is recorded onto camera doesn't mean it will be bad.

The only advice I would offer is to ensure you get clean dialogue. Make sure you speak up if you have a problem with a take and get the director to reshoot something. ADR is grand but can be tricky to match production sound.

Secondly make sure you record a wildtrack of the location where the film is being shot. It will be extremely useful to patch over things and fix any problems in the dialogue edit.

If you have a wildtrack and nice clean dialogue then your job in post will be made much easier.

Good luck!

3

If it does go out of sync, I doubt it would be even noticeable. I know some devices have crap clocks but still, in two or three minutes per take, I'd say not much can happen...

As for the workflow, there are several ways to start. 1. You're lucky, you've got DigiTranslator and you can open OMF files. In this case you're golden. The audio will be synced in the picture editing station and you'll get an OMF file with the audio nicely laid out in it. 2. You're not so lucky and won't be able to open OMF's or the picture editing station can't create OMF's. That means you need to sync audio with each take PRIOR to the start of the editing. If you don't edit the video with the production sound synced you're buried even before you start. You'll end up with the takes split into dozens of cuts and syncing using the claps won't be possible anymore, you'll have to sync to plosive sounds if you're lucky enough to have some.

But again, grab John Purcell's book, it's all in there.

For the rest of the workflow: the picture edit is happening with the production sound attached to each take, so the picture cut already sounds as best as it can for now. If you're not using OMF's I suggest you wait for the picture to be locked before you start editing sound.

Then I guess you know about the usual structure: edit, PREMIX, mix. Don't skip the premix. Work like a pro, don't start going around stuff ;) After all, you're trying to learn (just as I am).

2

Cheers guys, I had to Google OMF's as I haven't worked with those.

Before I have uploaded a final clip into Pro Tools and just added Audio tracks to do the project and then exported it as a Quicktime film (to hand in and upload to my site).

Yes I see what you mean about scenes comprising of several cuts and thus me losing track of all the audio takes needed (as I might need small sections from each take).

The initial editing of the visual aspect of the film will be done in Final Cut. I'm thinking that once my video editor imports all the video takes we've meed. . .I will import ALL of the audio takes in running alongside the video (so we'll have all the unedited footage audio and video synced and ready to edit). This way I can let him edit the entire film while always ensuring the audio is edited at exactly the same time? Does this sound like a reasonable approach? Then once the film is edited and we have the final film edit + dialogue. . .I can import that movie into Pro Tools, put ambient sound (that i'll have recorded in the scene space) over everything that isn't dialogue and then put in all the sound design of footsteps and all the other things that aren't related to audio into the scene.

This is genuinely frightening, lol. ALL of my stuff has involved working on a final clip with a clear canvas without having to worry about the film editing process at all! :D (I put two little demo's on my website if you want to check it out / criticize me! :D). I think it's going to be a challenge to try and take out all the sounds I want to add via Sound Design (say in a scene an actor talks but a door opens as he's talking - it'll be difficult to decide whether to keep that door sound or try and remove and totally redo it. . .as you'll still hear the ORIGINAL door noise as it'll be present on the dialogue track). Scarey - but you have to put yourself out there to get experience so I'm just going for it! I'm going to check out the books given also! Thanks for the answers so far.

2

Hi Aaron, I think what you mentioned above seems a little more complicated than it needs to be.

Do you have access to a small portable mixer? I think recording sound straight into camera would save you many headaches. Don't worry about quality, as long as you're going into camera via xlr (which an HD cam should have), the quality will be fine (just make sure the sound is going into the camera!). And you'll save time synching everything in post (which is actually a picture editor/editor assistant's job btw).

On the offchance you can't record into the camera, you can fix things in FCP. Once you've put your audio in sync with the video clip, delete the camera audio and link your audio to the vision. I'm not sure about the exact command, but you can make it so your audio will move with the vision.

One more thing; you probably already have some kind of relationship with the director, try and get on the good side of the 1st AD (or equivalent) as well. They're the ones who you need to yell at people when they make noise during your 30 sec of room tone!

I was in your shoes a few years ago, go for it! Film shoots can be long and gruelling, and sound can be rather downtrodden, but it's an important learning experience!

3
  • Thanks a lot. I've been reading more and more about dual set ups, and it does sound complicated. I'm sure I can get access to a portable mixer. If I plug the Mike into the mixer - then plug the mixer into the Camera is all the audio recorded directly to the camera's tape thus allowing the director to edit all the scenes the way they want it? I could then record the room tone directly to DAT away from the filming aspect and do that kind of thing post-production along with adding post production to a completely finished film project. Perhaps this is the way to go as opposed to the dual model. Jun 24, 2010 at 14:00
  • Sorry I should add - if I plug the mic directly (via mixer) into the camera. . .will it overwrite the on board microphone and thus make the audio of the recorded footage the "external mic" footage and completely ignore the onboard mic? Jun 24, 2010 at 14:06
  • Yep, you may want to get the camera op's help with this. You should be able to set the audio inputs on the camera. Your mixer should have a switch that makes it send tone, so send tone to the camera and adjust the levels on the camera so it reads -12dB. If recording 2 channels, make sure they're going to 2 separate channels on the camera. If you're only recording 1 channel, send that channel to both left and right channels on the camera, and set one to -12dB and the other to -20dB as a safety, to catch any peaks. Jun 24, 2010 at 15:51
2

I would respectfully disagree with the advice about keeping the boom mic farther out on close up shots as opposed to wide shots. When you go to do your dialog editing, the close up shots with the mic closer in will sound much better than the wide shots, because they will be much closer to the action and have far less distracting background noise.

A lot of scenes are primarily close ups and medium shots with maybe a wide shot at the beginning to establish perspective. As a contrary way of looking at this, maybe you can get the bulk of your sound coverage from the close up shots, when the boom mic is closer, and sync that close up sound over the wide shots in post. It really is all down to how the coverage and editing is. If you watch big movies you will sometimes hear a perspective shift as the camera angle changes. As long as it's not too distracting, I would optimize for good sound quality. As long as you don't get the mic element closer than about 12" from the person's mouth, you won't really run into any problems with proximity effect.

Besides, if anything gets the ADR treatment in post, it's most likely to be the wide shots, on account of too much background noise. You don't want your whole movie to sound like that, do you?

One other thing... prepare a proper shooting script before you start filming, and stick to it. It's so much easier if you know just from the audible slate at the beginning of a take that this is scene 3A take 4, than trying to guess where this file came from after the fact.

2
  • I agree with you Nathan, I'll edit my post to put a warning. Jun 25, 2010 at 9:30
  • Thanks for this , I didn't even think about problems arising from the "proximity effect". I'm definitely going to do a shooting script in advance - I really like to plan ahead so will be going through all of that with the directors. :) Jul 8, 2010 at 13:00
1

What brand of camera is it? If it's something like a Panasonix HVX, then you could send the sound from the mic to the camera using the XLR cable especially if you don't have a mixer anyway. You need to do a practice run long before the shoot starts.

2
  • i should have said, "may" need a test run. Jun 23, 2010 at 1:42
  • I'm not sure what the camera is. I've yet to meet any of my directors - it's all very much in the planning stage (I like to be prepared lol. . .I'm talking about the events even before I know ANYTHING about the video I'm recording! :D). Jun 24, 2010 at 14:05
1

I'll just say a word on syncing when you're recording dual system. For a professional film shoot, there's a step at the end of each day called syncing dailies. The dailies are all the takes from the day.

I think it's usually the production assistant (or the 1st Assistant Director maybe?) who's in charge of taking each video take, finding the audio take that goes with it (fortunately, as a production mixer, you created nice sound reports that show the scene/shot/take number and relate it to the file name in your recorder) and syncing up the two.

Once the production process is over and postprod is ready to start, the picture editor will have solid picture with good audio. He/she can then take it away. When he/she's finished, he'll create a LOW RES render for you as well as an OMF.

The advantage of the OMF is that when you import it into Pro Tools, you end up with all the discrete takes nicely laid out across several tracks and you can extend the handles of your regions to reveal the whole take if you want.

If you were to work with a simple rendered audio then the only audio you get is the one that was rendered. You get no handles to play with (which means no J-cut for example) and no flexibility. The trick here is to get all the raw audio takes into Pro Tools and start syncing up manually, using the rendered audio as a guide track. To do this, you can invert the phase of the guide track, slide your new take along and when you hear nothing (phases cancel each other), you're synced to the guide track!

3
  • Hey. I've read up on the syncing of dallies before (it's mentioned in a man called David Sonnenschein's book) - I'm hoping to avoid that as the takes should be relatively short and thus sync free! As for the OMF - I always export in AIFF. Should I start exporting from Pro Tools in OMF? Are there any clear differences or advantages? Jun 23, 2010 at 22:41
  • Pro Tools doesn't do OMF's natively. What's more, you're more likely to use OMF's to build your Pro Tools session than exporting OMF's from Pro Tools (I don't even know if that one is possible). OMF is not a file format, it's a "standard", a nomenclature. It's a sort of archive that, when opened, will place all the regions on the timeline where they have been edited. What you call "exporting from Pro Tools", is it "bouncing to disk"? Also, I don't understand what you mean by sync free short takes - all takes must be synced as long as there is audio and video recorded on separate media. Jun 24, 2010 at 0:30
  • What I mean by short takes was that each take of the film shouldn't be longer than 10 minutes (and I've read that audio going out of sync can sometimes occur after 45 mins of footage so hopefully my footage is so short that sync problems won't occur. In terms of the Pro Tools OMF I meant this: kenstone6.net/fcp_homepage/images_basic_export_using_omf_jordan/… Is this what you mean ? I've never exported to it as that format is completely new to me. Jun 24, 2010 at 14:08
0

If your camera has one input that'll be one channel and a stereo onboard mic would be likely to correspond to two more channels. To use several mics in one input you'd have to use a mixer and then send a mix of your several mics into the camera that'll be recorded on one single track. DON'T DO that. If it's your first, keep it simple. If you'd do that, the noise of all your mics would add up and mash up all your work. And everything that's recorded onto the tape will be presented to you that way in your editing environment in discrete tracks (as long as it's recorded on discrete channels).

5
  • Ok. Thanks for that warning - I was just concerned that I'll have a boom mic for the dialogue but also I'll need a few other mics that are stereos to record everything other than dialogue....perhaps plugging all those tracks into a mixer and then into Pro Tools as seperate tracks would be much better than my original dat machine plan.
    – user389
    Jul 9, 2010 at 8:58
  • During the takes a single boom should suffice. I'd keep stereo recordings for later. One boom is already a lot to focus on if it's your first gig. Don't be too daring, just do it well. If you don't have a mixer and your recorder only takes one input/mic, just go there with one boom (it worked for me for my first project: vimeo.com/10400372 pass: adelaide). It also was my first postprod and I did my best in the two weeks I had. But the booming worked well, I only wish I had recorded wild lines for the last scene! Jul 9, 2010 at 12:01
  • Cool. I'm going to give your video a watch this evening. :D I'll stick to one mono boom focusing solely on the dialogue. Then I'll record some room tone. . .and add the rest later. :) I meet with the 7 Masters Students tomorrow. They're pitching their films to me and I pick the one I think will be most interesting. Jul 13, 2010 at 22:03
  • Just watched your film! It was really great. :) You only used a single boom for the dialogue? Were all the traffic sounds captured by the same boom or added post? Was the only thing recorded with the boom the dialogue? Jul 13, 2010 at 22:34
  • Yeah I had no ADR so all dialogue you hear is all prod/alternate takes. That was all shot with an NTG-3 through a Shure FP33 mixer to be recorded on the Panasonic P2 camera. Traffic noise was already very present on prod tracks so I did my best with expansion to get rid of what I could, then I laid stereo traffic recordings I captured with my Tascam DR-100's onboard stereo mic. The real pain was the SFX part, I had to go out in town to record whatever I needed and the timeframe was very short. I did several all nighters in 2.5 weeks,e.g. to record water dropping on a brick a 4am in my kitchen. Jul 14, 2010 at 3:56

Not the answer you're looking for? Browse other questions tagged or ask your own question.