Treat it like you would anyother live recording.

They picked the space - you did not so you get what you can given the situation. You might have to add a little reverb (TC 3000 4000 or 6000 would be good for this application). I have done some of this in our local PBS station with somewhat similar acoustics and using a DECCA Tree and it turned out great.

Best of luck and PLEASE let us know how it turns out

5flagsaudio wrote: This room is DEAD acoustically.

Before worrying about the microphone selection, how big is the orchestra? Will they be able to hear themselves in that space? Without the 'acoustic glue' of reverberation to hold all the sounds together, I wouldn't be surprised if they end up using some foldback (just to make your life harder!).

I would treat this like an outdoor gig. Abandon all notions of purist stereo techniques, because this is not a purist stereo situation. Instead suspend mics over the different sections of the orchestra. This is similar to how they record orchestras for film scores, by the way, and most people would agree that orchestral film scores can sound pretty good - different to traditional stereo, but it works and people have learnt to accept that as the sound of an orchestra, especially when it is accompanied by video footage. The expectations are different.

Also, I wouldn't be too fussy about which microphones to choose, from a pure sound quality point of view. You have a problem to solve, so start thinking in terms of problem solving rather than sonic beauty. You have some NT5s and small ATs that are good for hanging, so use them. You can always EQ and so on because for this type of thing, a processed sound is not so bad, as long as it still represents the music.

I would send each section of the orchestra into a slightly different variation of the same reverberation algorithm. For example, I'd send all the strings into, say, a larger plate emulation. The violins and celli would go via a send with, say, a 20ms predelay. Violas and woodwinds into the same plate algorithm but with a significantly shorter predelay (say, 10ms) to make them sound a bit further behind the violins and celli. Put percussion into a smaller plate... I'll let you figure the rest out (when two sounds go into the same reverberation, the one with the longer predelay will sound closer), but try to be smarter than just pouring it all into one reverb algorithm.

Definitely multitrack, because they may ask you to remix it later for video, which usually means slightly emphasizing whatever instrument(s) the camera is focusing on.

And finally, because you're working with a video crew, be sure to send their feed through a stereo isolating transformer so that you are electrically isolated from them. Otherwise, you are likely to end up with an earth loop and because a) it is an audible problem, and b) the video crew will argue they didn't have it until they connected to you (!), it will be seen as your problem to fix.

Your room looks a lot like the Hal Prince theater at UPENN here in Phila. I've done a number of things in that space, from choirs to small orchs to soloists. I agree with you, it's not pretty and you'll have to work much harder than you'd expect compared to other venues. It'll be very boxy, to say the least.

I agree with the advice you've been given so far, and if I had your mic choices, I'd feel pretty confident about it as well. It's not a nice reverberant hall, but the viewers will know that as soon as they tune in. Not your fault at all, and if you went overboard creating the world's best mix, it might actually seem phoney, contrasted with the limited studio space. (Is the staging & lighting as limited as your audio options?)

I agree with you about using as many discretely placed spots as they'll allow. Something has to give SOMEWHERE, and it's not all that horrible to have a mic showing here or there anymore. People expect to see a little technology from time to time. You may be able to get the music director to rule in your favor for any disputed turff.

I recently just did a similar type of gig (but much smaller, and not live, but Tape-delayed) for a local PBS TV station, shot in an art gallery, and happily, the artist was one step ahead of me in terms of the audio. They specified that the TV crew got raw audio as a temp, with the stipulation that we'd give them the good stuff a week or so later. (We even did a few retakes afterwards, so we'd have some alternate takes in the can to fix a note here or there...) Nothing is going on the air without our final approval. (How often does THAT happen, eh? )

So, a "raw" feed went to their cameras & switcher, so they'd have the best audio possible for the moment, letting them have a nice stereo mix, for their temp & editing purposes. We're now in the midst of post production now, where I've created a much better mix, adjusted the gain, EQ, etc. of each mic as needed, along with a bit of reverb and some tiny amounts of peak limiting to get the best gain without sounding squashed. (This is baroque chamber music, so we've got to make a small tiny ensemble full enough to sound plausible onscreen without being hyped.)

As soon as the artist gives me an EDL, we'll make the final fixes and send it to the TV station for layback on the video track. (At 16/48, of course....)

If you're going out live as-is, you won't have this luxury, but hopefully you'll at least be off and isolated in some back room somewhere, with monitors you can trust, and a little bit of limiting & reverb dialed in? If there's any chance at all of giving them a better mix for layback before it airs, I'd fight tooth and nail to arrange it, even if it means a little extra work for you. (Your client - the orchestra - will love you for it.) Chances are good there will be some offline editing before it airs, and you may have a shot at convincing them of using the better mix in post.

Good luck with it, and do let us know how it ends up for you.

Thank you all for these things to consider. The orchestra will be rehearsing for the first time in the space 1 week from sunday. I expect that I will find out then if they will demand any reinforcement, which I've suggested against. The caveat: they're not rehearsing the same piece with the exact same personnel. Rather, they are rehearsing the program for their upcoming concert. Their next rehearsal will be 5 weeks from this one, with the right personnel and the right piece. Unfortunately, from there, it's another 3 weeks before they will actually perform the piece in studio.

The orchestra should be around 80 members, but that's just a guess. I believe I've decided to hang a mic array like I normally would for an orchestra, but also hang sectional mics as some of you suggested. That way I'm prepared for either circumstance.

To clarify one thing, while I said I was mixing live to video, it is not live to air. This is just one 25 minute segment of an hour documentary about the life of an orchestra player in my city. Therefore, I'm certain i'll be able to do any remixing that is necessary, although, I'm not quite sure how it will be synced with the video again as I've no experience in that department.

Again, thanks for your suggestions.

Anyone doing video these days is likely to be doing NLE stuff, so dropping in a new copy of the audio should be a breeze for them. I haven't run into tape (linear) editing suites in a long long time now.

Little by little, the "old guard" of difficult, inflexible & costly production work is disappearing, and IMHO, in this case it's a good thing. More are more folks I run into are hip to Avid, Final Cut or Vegas. It's all good, and handing someone a remix of an audio track should be easy for them, as long as you haven't done any changes WITHTIN the timeline, so that it stays in sync from beginning to end.

You might, however, want to make sure you get them the good mix before they do any cuts of the time line of their own. (say, between movements, etc. to remove lulls or tunings). The quicker you can get them your new mix, the better. (Obviously, make it 16/48, so they don't have to do any dithering or SCR - that stuff you want to make sure is done properly - by YOU - before it gets to them.)

I've had one instance where something wasn't right...

Doing just as you mention Joe - handing an uncut mix to the video guys of a 1 hour performance. The audio-only version ended about 4 seconds sooner than the video version did. Over the course of the 1 hour, something slipped and little by little the Audio and Video lost sync.

Since I ran this on two separate recorders and both of them maintained sample sync over the entire concert, I assume the fault was with the video recorder.

I personally wasn't too worried though since I provided my audio as a favor to the video guy and wasn't contracted to do it by the client.

Luckily, this is a single movement tone poem. It lasts about 25 minutes. I know they use Final Cut as a NLE, but this being a new experience for me, I didn't know the onus would be on them to do the syncing of picture to video.

I have mixed audio for video before, both for this station and other purposes, but it was always live to tape. I will discuss the possibility of remix with the music director and the video director to see where that leads. All of the musical events we've done in that studio have been live to tape, so it'll be a new concept for them.

5flagsaudio wrote: Luckily, this is a single movement tone poem. It lasts about 25 minutes. I know they use Final Cut as a NLE, but this being a new experience for me, I didn't know the onus would be on them to do the syncing of picture to video.

If I remember correctly, Final Cut can have multitrack audio, is that correct? (The Digital TV students at the school I teach at use it with multitrack sound, but perhaps they're running ProTools or something else with it?)

If they have multitrack audio capability, consider this:

a) As Joe said, do all of the audio at 48k, and ask them what word size they'd prefer. 16-bit for finished stuff, but if they have audio mixing capabilities they might use 24-bit.

b) Consider submixing into 'stems' (basically film sound speak for 'subgroups'), keeping the different sections of the orchestra - and soloists - on individual stems (stereo or 5.1, as required) that all combine into the final mix. That way, they can gently push up the volume of individual sections, highlighting them as the camera focuses on them. This is remarkably effective when done in a subtle manner. Without it, the sound remains static and simply feels wrong as the camera moves around, especially when the camera zooms in on a soloist and the sound doesn't change at all.

But anyway, all of this takes time and costs money...

As for syncing, the smartest thing to do there is ask someone in their crew how they intend to go about it. Perhaps they'll just be using clapper boards, which work quite well in these days of fully digital everything. But perhaps they'll want you to sync to them and capture the code in your recording. If so, talk to someone who knows what they're doing with it! Timecode itself can be a can of worms, I'm afraid, so many variations and so on.

You might also want to get on board with their take numbering system, so that the different takes of audio are named in accordance with the different video takes. They'll possibly be using a naming system that includes scenes, takes and slates. Depending if they're using the US or European system, you'll find takes are labelled with a combination of numbers and letters. The film and video people know exactly what all of this means, and that's how they can quickly locate the appropriate take of a scene or whatever. Best thing to do in that case is ask one of their crew for the full take 'number', either immediately before or immediately after each take, and make sure you label the audio accordingly.

If not, you're going to look rather unprofessional (from a video point of view) when they're saying, "Okay, we need the audio for take blah-blah-blah", and you're jumping from take to take, auditioning a moment and saying "Is this it? Or maybe this one?" A little advanced planning here to make sure you're all speaking the same language (in terms of naming conventions) will make a huge difference downstream.

If you're really good, you might also keep track of which takes are 'circled'. When the producer/director/whoever decides they have a good take, they draw a circle around the take number on the log sheet, to show it is the one they're most likely going to use. By the way, the circled take is not necessarily the last one they do - as with audio, they might do four or five takes but then decide that the second or third was the best one, thereby circling it.

[Forgive me if you already know this stuff, by the way...]

Update:

The orchestra rehearsed in the studio last night. I put up 16 mics, My main array, a woodwind stereo spot, a spot on the harp (which wasn't included in this rehearsal) a spot on the timpani, a pair of spots on Violin 1, pair on Violin 2, pair on Viola, 1 spot each on cello and bass. I ended up not using the spot mics, with exception of the WW and timpani. I'm also assuming I'll need the harp spot.

If I can figure out how to post a sample of the recording, I'll do that later in the week.

On another note, it looks like this will be a live to tape mix with no chance of remix, but I'm going to multitrack it anyway.