Choir recording

Posted: Wed Mar 30, 2005 5:01 pm

Cucco wrote:

Sorry, I hate to sound like I'm poo-pooing, you're still cool in my book. Thumbs Up

J.

Cucco, I feel I am learning all the time from you as well as the rest of the experíenced people on this forum. Hit me on the head when I believe I am right without beeing it (and my wife tells me I need quite a bit of that).

Gunnar

Posted: Wed Mar 30, 2005 5:55 pm

I just got back from setting up the choir project. I ended up putting the ORTIF with two cardioids about 10 feet in front of the choir and just for an option put the Blumlein about another 15 ft back from the ORTIF both about 2 to 4 feet above where the heads of the choir will be.(thats as far as I could go back with the Blumlein and nowhere in the back ,without being in the way, to put stands for room reverb without going to extremes to hang from a 40ft cathedral ceiling.) Just having my partner walk around the choir area talking and singing and yelling bad renditions of Beatles songs I found that I may well be able to use a mix of both pairs although I have to say the Blumlein sounded very nice by themselves with a good amount of room verb and still very present. I was very pleased to hear what seems to be a very smooth stereo image from left/ center / right in terms of amplitude and freq response from both pairs. Tomorrow will tell the tale. Well , going on and on but I wanted to give all you guys a very BIG THANKS of gratitude and if you were here the drinks would be on me. I'll drop a note when its over to let you know how it turns out. Once again. Many Thanks

Posted: Wed Mar 30, 2005 8:02 pm

Well, given that you've done the recording, I would think that it's a bit late to give advice since you've already done some of the work, but...

Since you are also doing pipe organ directly behind the choir - I would definitely try blum out front and then spot the choir with the cardioids. Unless your rather small choir is good at shouting in tune, they'll need some help. Sure, the organ won't sound as good as it possibly could, the Blum setup will yield decent results with it and then bring the spots in where help is needed.

J.

Posted: Thu Mar 31, 2005 8:32 am

It would be easy to lower the front cardioids to bring them closer to the choir . I'll do that. This choir has been together for 12 years and are conducted by the Director of Choral Activities at the U of WIS so things are well under control as far as balance between the organ and the choir and we record the rehearsal and then the live show 2 days later so I do have a small window at the beginning of rehearsal to make small adjustments. But like you said earlier, The Conductor will be there to work and not watch us moving mics around. Thanks for this last minute tip. It wasnt too late. I record the rehearsal at 5 today. Again Thanks for your help!

Posted: Tue Apr 05, 2005 8:36 am

Well the recording is done and it turned out nicely. My question now is .......The songs are extremely dynamic in levels. How would you approach getting gain on the overall and calming the peaks without squashing the performance? I have a good grasp on comp/limiting but mainly with rock/blues and was wondering if someone has specific suggestions for choir. Thanks

Posted: Tue Apr 05, 2005 8:56 am

MMMmmm... Dynamic Range.

For peaks, it's perfectly acceptable to use a limiter. However, I'll qualify this quite a bit.

Only on extreme peaks should the limiter be hit. Otherwise, you will want most musical peaks to remain under the threshold of the limiter.

Personally, when I mixdown a live recording here are a few of the guidelines I use for dynamic range.

1. I like my room noise floor on the recording to average at around -65 - -70 dBfs.
2. I like my loudest selection to hit 0dBfs. Any "non-musical" peaks such as extreme cymbal crashes (not to say cymbals aren't music - simply, they can be excessively loud and if not tamed, can cause the recording to sound too soft. This goes for occassional BD hits, Tam-Tam, picollo, brake drum, etc.) may be limited or gain-riden down to a more moderate level.
3. I like my lowest sounding articulation to start as low as -50dBfs or lower. Of course, this seriously depends on the type of music. A disc full of Sousa marches just won't have this kind of dynamic range. The Mozart Requiem would have even more.

In general, I try to make it sound as dynamic as the concert/performance. You would think this would be easy - set the volume so you don't clip and you're cool, right? Nope, a little bit of gain riding here and there is essential. Most the time that conductors sit in as producer, I constantly hear "can you make that section quieter?" Musicians under the influence of concert adrenaline will have a tendancy to make things louder than usual. Including pp sections that really need to be super soft.

Very rarely do I ever use a compressor. Occassionally, I'll use it on wind ensemble/bands and just recently, I had to use one on orchestra. The pic was so forward and bright, I had to use a multi-band carefully tuned to tame her down.

Other than that, minimal dynamic processors are required.

I hope this helps.
J.

Posted: Tue Apr 05, 2005 9:10 am

Yes it helps. I'll give it a go. Thanks

Posted: Tue Apr 05, 2005 9:58 am

I work with extreme dynamics manually. I will insert long crossfades (linear) and then gradually change level as needed.

Beyond that, for classical work, parallel compression can work well as well as gentle limiting.

--Ben

Posted: Tue Apr 05, 2005 10:19 am

I always "Test out" any limiting I do to check if I can hear bad stuff creeping in before I commit. Usually, cymbals, plosives, even mouth-ticks (if it's a close-mic'd situation) often create useless peaks that should be tamed.

I'm just working on a jazz quartet piece that's just gorgeous; starts out small and quiet, and builds......and builds.......and builds, to something that's actually TOO big, so I have to tame the drummer a bit, separate the music track from the applause at the end, and GENTLY (with limiting) reduce the peaks that are making the dynamic range too wide and in general not usable for much in the "Real world".

If I can get those peaks tamed WITHOUT tampering with the overall sound, or messing up the rest of the transients, then it's a "go".

I usually use a very fast attack & release, and set the limiting for a brick-wall peak that I don't want to exceed. Then I process and listen, a/b to the original. Working in 24 bits or 32 float, this usually works quite well, and there's rarely any artifacts when done properly. In the worst scenarios, I'll use a multiband instead. Once I'm happy with it, I bring up the track to my nominal levels and move along to the next issue. Thumbs Up

This topic sponsored by: