Chorus and Piano

Recording Techniques, People Skills, Gear, Recording Spaces, Computers, and DIY

Moderators: drumsound, tomb

Post Reply
Fieryjack
steve albini likes it
Posts: 385
Joined: Wed Jun 16, 2004 9:25 am
Location: New York, USA

Chorus and Piano

Post by Fieryjack » Fri Nov 18, 2005 6:28 pm

Hi: I recorded a 50 person chorus and solo piano....my mic selection seemed okay, but I'm not entirely pleased with my results. I had a pair of MC012s on the piano...sounded really nice, and a pair of AT 4033s on the choir, spaced pair, about 10 feet away.

I went into a Focusrite Octapre, then into my 828MKII....levels were fine coming in on all my meters, etc, never hit the red. However, when I listen back, on the "heavy" choral parts where they are singing loudly, there is a kind of distortion I hear....

I have no idea what caused this....am I too close on the mics? (even though my levels were good)? Maybe its the way they really sound with all the SPL belting out at the same time? Maybe my mics are shot? Low levels sound great...peak levels are questionable.

Professor
ghost haunting audio students
Posts: 3307
Joined: Wed May 07, 2003 2:11 pm
Location: I have arrived... but where the hell am I?

Post by Professor » Fri Nov 18, 2005 7:18 pm

I'm going to go out on a limb here and say that your levels looked fine because they are measuring the average volume level across the whole spectrum, but if you were to throw on a real-time analyzer, you would see a really big spike around 600-800, and again around 1200-1600. Those are your sopranos talking to you... or more accurately, screaming at you. If you have some form of RTA-style analyzer available as a plug-in on your DAW software, then kick it on and see at which frequencies the peaks are hitting. You may be able to tame the distortions in the mix by applying some EQ notches to those frequencies. For release projects, I will often automate the EQ so it is applied right to the offending frequencies (and their harmonics) right when it's needed.
You are probably hearing one of two common types of distortion from the nice ladies. Since you identified it as "distortion" and you record at all, I'll figure that you heard the crispy, crunchy, crackling sound of levels that are too hot. That's because there is a particular frequency band that is hitting into the red even though the rest of the material is quite safely in the green, and causing the average meters to read in the green. It could be clipping at the mic electronics (which I would consider less likely, though you could certainly kick on a pad to be sure), or it could be clipping at the preamp electronics (which is likely in your case) or it could be clipping at the A-to-D converter (also quite likely for your case). The solution there is just to record at lower levels, and make sure that you are not enhancing the problem by using a mic with a natural bump in the range where your sopranos are screaming.
The other kind of distortion is when the levels get very loud at different pitches and you start to hear a hollow, bell-like chiming of weird frequencies that don't seem like they are part of the singing. This is intermodulation distortion which is actually an acoustic phenomenon as well as a less common electronic one. Drummers know this one from when they bash and crach on their cymbals in a small room for a while and suddenly think they are hearing the phone ringing. Quite literally it is the different pitches creating 'beat frequencies' that are equal to the sum and difference of the pitches and their harmonics. For example, one soprano screaming at 800Hz and another at 1.2kHz will create intermodulation tones at 400Hz (the difference) and 2kHz (the sum), and since 800Hz has a strong harmonic at 1.6kHz, that will reinforce the 400Hz difference with the 1.2k, etc. Add to that the sound of 8 or 10 screaming women, with all their vibrato (vibrati? ) cooking and you'll have lots of intermodulation distortion also screaming in your ears.

But I'm guessing you had regular ol' distortion in a narrow enough band that it didn't read into the red on the meters.

-Jeremy

Fieryjack
steve albini likes it
Posts: 385
Joined: Wed Jun 16, 2004 9:25 am
Location: New York, USA

Post by Fieryjack » Fri Nov 18, 2005 8:26 pm

Wow, Jeremy...thank you for the insight...I never would have suspected that and I guess I've put entirely too much trust in my metering for my own good.

It certainly makes sense, because those mics always work fine on other applications (i.e. guitars, straight vox, etc.) that that aren't recording multiple frequencies at high levels.....(i.e. chorus of basses, sopranos, etc.)

I'll cut back my recording levels next time by about 30-40% and that should do it. Thanks again.

Professor
ghost haunting audio students
Posts: 3307
Joined: Wed May 07, 2003 2:11 pm
Location: I have arrived... but where the hell am I?

Post by Professor » Fri Nov 18, 2005 9:52 pm

Yeah, I've been meaning to write out a long diatribe about dB-VU and dB-FS for some time now, but haven't gotten around to it yet. See, I am a child of digital as well, and I see those meters and I want to push that level right up there to the top, and walk this terribly uncomfortable tightrope between that wondefully strong signal and that obnoxiously ugly distortion when it crosses the line from -1dB to 0dB.
And while I can't pin point to a single event, I kinda kept seeing the references, the numbers, and considering the reality of analog vs. digital metering - not recording, but metering... and I kind of reached this epiphany of "0dBVU = -18dBFS". Not that this is some kind of original concept of important scientific formula, but it made me reconsider what I'd been doing. 0dBVU is the zero at the top of an analog meter, and when tape guys talk about pushing into the "red" they are talking about pushing maybe +3, +6, maybe even an extreme like +10dB beyond that line into the headroom of any particular device. But that 0dBVU corresponds to only -18dBFS on my digital scale. So if I want to go balls deep on a recording and hit +10, then I should be setting my gain so that the loudest peaks on my recording hit around -8dBFS which ain't even close to the top of my meters.
And that is exactly what I've started to do. After all, the original reason for sending the strongest possible level to tape was that tape was a noisy medium and you wanted the best signal-to-noise ratio. Well chances are that if you're using modern professional gear, then the noise floors are insanely low. Your mics, preamps and recorders have such low noise that the loudest noise source in your signal path is probably the ROOM itself, and it doesn't matter whether you crank the gain at the preamp, the recorder or in mixdown, that room is going to be where it is.
So I track much lower these days, and I simply turn up a little more in mixdown. In the end, it leaves more room for level in the mix and in the mastering, and it really makes the tracking much less stressful - especially in live location recording.

-Jeremy

Fieryjack
steve albini likes it
Posts: 385
Joined: Wed Jun 16, 2004 9:25 am
Location: New York, USA

Post by Fieryjack » Sat Nov 19, 2005 7:14 am

Great stuff...great advice. Thanks again.

Post Reply

Who is online

Users browsing this forum: No registered users and 58 guests