Overheads and room mics mixing and recording thoughts

As I’m mixing things that I (and others) recorded I notice that sometimes the room mics and overheads are unbalanced.

For instance, with overheads, the snare can be a little off-center, and sometimes even the kick is leaning to one side more than the other. I usually place overheads as a spaced pair and not in XY so the overheads really are just there to capture the cymbals. However, because the OH are a spaced pair and are positioned above the cymbals, the kick and snare end up off-center. It’s probably possible to place a spaced pair so that they capture the cymbals but the rest of the kit is still balanced, but these things can be fixed in mixing too. A high pass at around 250Hz would basically eliminate the kick, and a really fast compressor will work to eliminate the snare from the overheads without washing out the cymbals.

I noticed that some of my room mics have a similar issue – the kick and snare aren’t centered. The way I almost always position the room mics is on the floor and so that they create an equilateral triangle with the kick. This should guarantee that the kick is centered, but I guess sometimes the microphones aren’t symmetric with regards to the kick. However, this positioning will always lead to the snare leaning to one side or the other. Since the room mics serve as a full image of the drum kit, this isn’t something that can be fixed during mixing. So maybe a better method is to keep the microphones symmetric along the line that runs between the kick and snare. One way of doing that would be to position the microphones in front and behind the kick. Another is to have the microphones in a non-equilateral triangle formation.

With all this being said, I haven’t really noticed the drums being asymmetric once the close mics are brought up, so maybe this isn’t something that a lot of time is spent on during quick sessions.


Recording Brass Quintet

Yesterday was my first time recording a brass quintet. In fact, I think it was my first time recording brass anything!

We recorded in a church that was relatively big – maybe 35ft wide and 50-60 ft long, with gable (symmetric) ceiling that was maybe 40ft tall. The room had a lot of hard surfaces – brick, wood, and some glass. As a result it had a little bit of slap back/flutter echo. It wasn’t too terrible, but noticeable when clapping my hands. I also found some fabric baffles there and placed them in front of the windows.

The band was seated in a U shape, left to right from audience perspective:
1st trumpet (D)


Tuba (“bottom” of the U)

French horn

2nd trumpet (C)


I positioned to Coles 4038 in a Blumlein array at 10′ from the line between the two trumpets and 7′-8′ high. I also added a Sony C48 as a spot microphone on the tuba. It was positioned 3′ in front of the tuba and maybe 6-7′ tall.

It all sounded pretty good! Before I settled on this arrangement, I tried the same array but a few feet back and that sounded too far, almost like recording from another room. The tuba spot mic sounded fine, not as full as I would have liked, but enough to focus it.

Here are a few things I learned in the process:

  • When recording brass for classical music, the idea is to get their projection in the room. The recording is supposed to mimic a performance, so the microphones should be outside the line of sight of any of the bells. In other words, the instruments shouldn’t project directly into the microphones. That also means that the sound will be more balanced.
  • However, the group was seated, and the trumpets’ bells were facing the floor because they had to read the score. At the end of the recording the trumpet players listened back, and they said that there was a lack of articulation and some things they were doing (like tonguing) didn’t come through very well. I myself noticed that they’re facing the floor halfway through the recording and realized I should have lowered the microphones by about a foot.
  • French horns shouldn’t be mic’ed, and that makes sense. The horn is behind the player and sometimes they stick their hand in there! It’s supposed to be reflected back.
  • If you record classical music, it’s a good idea to be able to read and follow the score! It makes communication easier.
  • ORTF can be good for recording small groups. It’ll have a more direct sound and reverb can be added later.
  • A lot of people prefer spaced omnis for classical music. I don’t have a pair of matched omnis.

One thing I’m realizing I did wrong is that I didn’t listen to the whole group play in the room. I listened to a few individual instruments (trumpet, trombone, horn) and I walked around clapping and listening to the room, but I never stood where the microphones were and listened. I should have! That’s the most important listening test! I have to remember that for next time.


I will add a short sample soon.



Guitar recording

This might not be helpful to anyone who reads this but me.

Recorded guitar today. Guild S-300D into an Orange Tiny Terror into a 1×12″ ported cab (similar to TL806).

I used a Coles 4038 going straight into my little sound interface’s preamp (Avid one track thing).

I aimed it pretty much at the center of the speaker, and at a foot away it picked up too much of my picking and sounded a little thin (I wasn’t playing loud). At about 8″ away it was a little too boomy. Will try 10″ away next time.

UPDATE January 8th, 2015:

Same setup again (all of it). I kept the microphone 10″ away from the amp and I got the same sound. At 10″ away there’s already a lot of bass because of the proximity effect. It’s not a bad sound, might even be a good sound, but it doesn’t sound exactly like what I’m hearing in the room. I should mention, though, that I play quiet enough that I can hear the acoustic sound of the guitar along with the amp. That could contribute to why it doesn’t sound the same. Anyway, I moved the microphone maybe an inch or two away from the center (off-axis) and things cleared out a little more. Not as boomy anymore. That’s something to keep in mind.

Recording and Mixing Hungry Man’s Eater

Hungry Man is a band I’m in and Eater is our first record. I recorded it almost a year ago (with Jon San Paolo’s assistance/tape op-ing) and mixed it a couple of weeks later. I’m not happy with how it sounds and want to put down in writing what I don’t like about it and what I should have done better. If you read this and are interested in listening to the record, you can hear it here.

I think it was recorded well, and frankly, a monkey can setup in Electrical Audio’s studio A and get usable results. The problem was mixing it. Here’s what I think I did wrong and what I’d do differently next time.

– We mixed loud. I didn’t have an SPL meter with me, but I think it was at the very least 85dB SPL loud. This was dumb for a couple of reasons. One is that our ears got fatigued fast and we took no breaks (mixing was done in a day) which made them even more tired. The second is that it was my second time working in that studio (not EA), so I didn’t really know how things sound there, and I think the room was interfering with what I was hearing. Also, when we mastered Bob Weston gave me this great advice – he said to mix at the volume the music will be listened to. With all that being said, the Fletcher-Munson curve says that our hearing flattens as things get louder, so maybe loud is better?

– The bass is too loud. I think that this goes back to monitoring too loud (although it contradicts what I said about the Fletcher-Munson curve), but also to the fact that I play bass and probably had the inclination to push it up. In the room it really didn’t sound that loud. If I were to mix it now, I’d drop the bass down by at least 2dB.

– The snare sounds wonky. This is probably my biggest mistake. The snare sounded beautiful when we recorded, it’s a 5.5 Supraphonc chrome over brass that has an explosive sound. However, sometimes it disappeared in the mix, so I set on fixing that by pushing some mid frequencies. I think I overdid it which made it actually lose some definition and sound wonky. A better idea would have been to compress it a tiny bit. That would have helped it stick out. FWIW, I got the mid frequencies advice from a fellow engineer who does amazing work, but I executed it wrong.

– The kick is.. weird. It’s very full and has this insane attack, but it’s too, I don’t know, pop-punk? What was going on was that I gated the beater side microphone to get rid of the snare there, and I mixed it too loud with the front side microphone. I think the fact that I consistently mixed it wrong with the front side mic is maybe a testament to how I wasn’t hearing things as they actually were.

– The drums disappear sometimes. I could have maybe gotten away with compressing the drums. Just a tiny bit so they don’t drown.

– The record is dark. Hear how dark it is? It was actually darker when we mixed it. I should have eq-ed the overheads and added ~1.5dB at around 10k. Maybe same thing with the room mics. Definitely should have added high end to the guitars and cut some of their low end. I think that part of the bass being loud is that we were competing over the same frequencies a little bit.

– There are room mics on the vocals, and I gated them so they open up when the singer yells. Thing is, the singer always kind of yells. It’s not like Heart Shaped Box by Nirvana where Kurt Cobain all of a sudden yells while the band maintains the same dynamics. Gating the room mics is a cool effect, but doesn’t work on every record or every song, and I should have known better.

– There was a room mic on the guitars that we ended up not using. I honestly don’t think there was room for guitar room here, but a big part of why we didn’t use the room mic is because it was unusable. It just sounded flimsy. Like someone accidentally left a microphone open, and it sounds bad and off-axis. What happened is that the guitars were in Center Field and they weren’t very loud. I mean, they were loud, but they weren’t absolutely filling the space with sound. The room mic was placed very far from the amps, maybe 25ft or more. It was dumb. Placing it closer I could have gotten something useful.

– We mixed 8 songs AND re-did vocals for a couple of songs in one day. And of course it’s not like we got in and started working. I had to calibrate the tape machines, set up microphones for singing (it’s quick, but still), then break them down, and only then start mixing. Also, I got in a car accident (wasn’t my fault) on the way to the studio, and while I was physically fine, it did cost me 30-45 minutes of dealing with it. I was also thrown off by the whole thing a bit.

To sum what I should have done differently in the mix: lower the bass, boost the high end on the drums, boost the high end on the guitars, don’t mix in so much of the beater side mic, don’t gate the room mics.

Also, next time I mix a record I think I’ll tell the band to get there half a day after me. This way I can listen to everything and take notes on what might need work, and be more prepared. Also, mixing 8 songs in one day is a bad idea. Two days of mixing would have made this so much easier, physically and mentally, and would have given me more time to get things to sound right.