So when we listen to sounds IRL, there's a natural HAAS effect as well as natural comb-filtering.
The HAAS helps us localize sounds, but does the comb-filtering help us, or is it simply a kind of 'noise' to our brains?
And if the comb-filtering inherent to the real world is beneficial, and recording with a dummy-head is better sounding (meaning sounds are easier to localize, and less fatiguing for the brain) then would the filtering inherent to an AB recording not likewise be 'beneficial'?
Is there any research on this?
As an example, here's a drum-track I'm trying to mix, recording in AB using Earthworks small diaphragm condensers.
If left alone, the kick is *almost* centred, but you can shift phase slightly so that kick is centred, and it'll effect the localization of everything else slightly. I would think centering the kick would be ideal here, because otherwise the comb-filtering will remove some of the fundamental (kick is tuned to 110Hz, snare to 165Hz IIRC).
It sounds like you are more interested in the phase effects of a comb filter than the frequency effects. The phase does indeed have a real effect on the perceived source direction in a stereo reproduction system.
In a real environment, the phase response of a comb filter embeds clues as to the distance and direction of the surroundings near the source, such as walls. The ear is not naturally sensitive to this, it only becomes apparent when the comb is frequency-swept. But the brain perceives them regardless and uses them to reconstruct the environment, much like a blind person can use the audio reflections from clicks and whistles to determine that he is approaching a wall.
In terms of recording, I don't know where that leaves you. Unless you want to create the illusion of a wall...
Well, I think what that means is that phase-aligned drums are not theoretically *cleaner* than non-phase aligned.
Certain sample libraries, for example, will phase-aligned the mics (close and room mics), but maybe this isn't a good idea. Maybe an AB with each drum being more or less out of phase in each mic actually *is* idea (both in practice, and in theory).
I'm thinking that in the future, I'm going to record drums using AB mics (but lower, to pick up a balanced amount of kick), and phase-align according to the kick, but just embrace the fact that every other drum is out of phase.
I agree about phase aligning. It removes a lot of the sound of the space in which the recording happened. It might be helpful in poor acoustic spaces or where mic positioning was not optimal. But otherwise I consider it a mistake, a matter of over engineering.
Comb-filtering occurs because of delays and in the real world that comes from reflections. If there are a lot of them and they are the right length they sound great. An example of this would be the reverb you hear in a concert hall. On the other hand when rooms aren't constructed with acoustics in mind they often times have unpleasent comb filtering. Clap you hands in a small room in your house and you'll probably hear a metallic ringing. When you record in places like this that comb filtered sound is imprinted on your recording and you can't remove it. I'd say in general comb filtering isn't helpful and will just cause you recordings to sound like they were recorded in a tin shed. When it is helpful its usually referred to as reverb or ambiance.
In a nutshell, human ears perceive comb filtering and the world very differently from a microphone. We usually have two ears separated by a good acoustic absorber (the head), as per your binaural example. We also move about a lot to build up a sonic picture. Even a tiny movement can make a big difference in comb filtering. Mics are stuck in one place, so you're stuck with the comb filtering in that particular location... it's not always very natural, and not at all good if you're stuck with a bad sound.
In practical terms;
It's not possible to make all drums perfectly phase aligned in all mics. Getting one drum aligned in a given mic can make another drum misaligned in it. So I'd agree that there isn't much to be gained from phase aligning drum mics, unless it's the answer to a specific problem such as the one you've identified.
I've used phase alignment of overheads for a similar problem and it worked well. Just check that you're not making something else unacceptably worse, e.g the snare.
Saying that, in the files you sent, to my ears at least, the snare is both substantially louder and more off centre than the kick. IMHO that's a more serious issue than the kick, unless of course you want the snare to be where it is.
Do you have any close mics on the kit? These can help pull a drum closer to the centre. If you're not after a wide stereo image, you might consider also narrowing the stereo width, or even going to complete mono for the drum overheads. Maybe even just use a single overhead mic that gives the closest sound to what you're after.
Much of this depends on the music, and how loud the overheads will be in the mix. Drums being off centre may matter a lot, or not at all.
There's a lot of room sound in these, and the comb filtering / room sound is not fabulous. I agree that moving the mics lower and closer to the kit will give a better result.
AB micing often comes with these problems. Comb filtering can make certain frequencies or even instruments quieter when summed to mono. Always worth checking the sum to mono when using a stereo pair. Have you tried ORTF or XY - they suffer a lot less from these issues.
Reading Jennifer's response I had a crazy idea... given the difference between mic and human perception I wonder if anyone has ever experimented with MOVING mics during a recording! So for example directing a stereo pair a little way towards a soloist etc.
I can see all sorts of practical issues but as a principle it seems... at least novel.
I can see all sorts of practical issues but as a principle it seems... at least novel.
I remember a discussion on a DIY forum from years ago. Someone had the idea that a kind of a robot to move mics around would be interesting. You wouldn't have to walk back and forth between control room and live room to change mic placement. Some were convinced this would be ideal for reamping and drums.
Some experiments were done and there was a robot on the market for a very brief period. In the end, it was simply too complicated and too expensive.
Moving mics while recording was very impractical, since the motors moving the mics made some noise and vibrations.
It still lives on in motorised stereo booms for mics that are suspended from the ceiling in great halls. It's used mainly for classical recordings. But even that is dying out, as it is far simpler to put extra mics in the other positions.
Of course, big studio's have human studio assistants to move mics.
__________________
In a time of deceit telling the truth is a revolutionary act.
George Orwell
Reading Jennifer's response I had a crazy idea... given the difference between mic and human perception I wonder if anyone has ever experimented with MOVING mics during a recording! So for example directing a stereo pair a little way towards a soloist etc.
I can see all sorts of practical issues but as a principle it seems... at least novel.
There's always moving the source relative to a fixed microphone, sometimes much easier!
In the very old days of just one microphone recording entire ensembles live direct to vinyl, soloists used to move closer to the mic when it was 'their turn' and back off when not.
There are also moving / rotating sound sources such as leslie speakers where the effects of changing comb filtering can be heard in a good way.
And micing an accordion, where you'd best get it right at source or you can end up with a nightmare to mix.
given the difference between mic and human perception I wonder if anyone has ever experimented with MOVING mics during a recording!
I attached my GoPro to my guitar at rehearsal on Sunday and you get ^that for free. Outside of the specifics of comb filtering, just a gentle reminder that everything we hear, ever, is a product of phase interactions.
__________________ Music is what feelings sound like.
Some of my old band recordings have moving mics (I'd hold a mic and run around, experimentally). Our first Album also had swinging mics simply because we couldn't afford mic stands, so all the mics were hanging from the ceiling.
@Jennifer, thanks so much for that reply! I really prefer wide sounds, and personally like the sound of AB over XY (maybe just for the width). I'm trying to create an image of being 'in the room', so the snare should be to the right, the hat slightly more to the right, floor tom to the left, and kick roughly center. The idea somewhat being an unrealistically-wide binaural sound. We only had these 2 mics recording unfortunately.
I was thinking that phase-aligning to the kick would
1 - centre the kick
2 - make the kick louder / less combing on the kick / stronger attack on the kick.
I'll have to do some boosting at 110Hz to bring the kick out for sure (so the kick is tuned to A 110Hz, the tom to E 165Hz, and the floor to B 123.75Hz (pased on simple ratios), so we can play the other instruments in Pythagorean A minor/major, and tom-fills harmonize with the dominant chord.)