David Mellor: So you want to record in stereo? You have two microphones. How are you going to do it? I'm David Mellor, a dabbler in audio in my boyhood and teens, a dabbler in professional audio ever since. I say dabbler because one thing I've learned for certain over the years is that there's always more to know, both practically and theoretically, but we all have to start somewhere. So I'm going to use this podcast to tell my younger self at a stage when I knew a bit and I was ready to learn more how to record in stereo. I'm going to narrow this down just to be clear. I'm thinking about vocals and acoustic instruments, choirs, bands and orchestras. As much as I like to create irritating background music for my YouTube equipment reviews, using my digital audio workstation software and synth plug-ins and it sounds a bit like stereo, I'm sticking to the real thing here - real sound, real music, real stereo. That's what I'll be talking about and I'll be giving you some demonstrations later that I recorded in Abbey Road Studio two, where they totally, absolutely do not dabble.

It's interesting to consider firstly, what actually is stereo. Well, what's mono to begin with. It's funny that ever since I first heard of this thing called stereo, people have been telling me that mono means one - one microphone, one recording or broadcast channel, one amp and one speaker, but a Greek-speaking commenter on my YouTube channel corrected the error of my ways, or my ways as I have been taught. Mono or μόνος means more qualitatively, alone. That's what my commenter said, I don't know anything about languages other than how to order a beer, or two beers per favore, but I quite like alone to describe single channel audio, something missing, needing a companion, another channel. The problem with mono and I don't see any other problems, is that it's not how our ears work. We have two ears that allow us to hear the direction from which a sound source is coming, localise the sound source we can say - localisation. How this works is threefold. A sound coming from the left will strike the left ear before the right and the brain can interpret this timing difference. Also, the head will shadow the right ear from a sound coming from the left causing a level difference and a frequency balance difference. Interestingly, the shape of the pinner, the ear lug if you like, plays a part not only helping left right localisation, but up and down too. Think of evolution. As the primitive lemurs we used to be, we needed to know whether the threat of being eaten was coming from a bird above or a lizard below. But we won't bother with height, that would have to be another story for another day. But mono, as I said, it's not how our ears work. We can enjoy it as we can enjoy a black and white film, but it isn't like real life. I'm just going to throw in here that there are indeed mono enthusiasts, just as there are black and white photography enthusiasts. I respect your choices, but I'm going with stereo and color.

As a first step to recording in stereo, it might seem obvious to mimic the human hearing system. Two ears, one on the left, one on the right, with a thick, dense head in between. Someone told me the other day to stop talking about myself. So I had a go at this a long time ago. I typed miniature microphones in my ears and made recordings of instruments and ambient sounds if I could find my old recordings. Made on a lovely Nagra IV-S I'd demonstrate them, but you'll do a whole lot better searching online for a play called The Revenge by Andrew Sachs, who if you're old enough, you might remember as Manuel in Fawlty Towers, but there's no 'k' about this. It's a superb example of binaural recording and the play consisting entirely of ambient sounds with no dialogue is brilliant. I slipped a technical term in there - binaural. I'm going to skip over this quickly and say that it really only works on headphones or earbuds. It just sounds odd on speakers, but it's there to try and it's fun. You don't need an actual human head, your own or otherwise, just something that mimics its sound insulating and absorbing properties between your mics spaced an ear-to-ear distance apart.

I slipped another term in there almost accidentally. Stereo. We are back to Greek again, and I think I'm on solid ground if I say that it means, well, solid. It certainly doesn't mean two and it's a bit of a myth that stereo has always meant two channels. Stereo in sound means creating a solid illusion of direction and space. Three channel stereo was a thing, particularly in cinema with left, center and right channels and as we know, a solid has three dimensions, so this is the ambition of stereo, however many channels you use to create a solid three-dimensional illusion of hearing voices, instruments, choirs, bands and orchestras from loudspeakers and today we normally attempt this with two channels, feeding two loudspeakers. Any more than three channels though, well, this is surround and beyond my ambitions for today. Another throw-in - some experts will opine that whatever you are getting from your headphones, it isn't stereo. Do you know for today, I'm just going to ignore that.

Here's a good question - how many microphones do you need to record in stereo? Well, if you're recording an orchestra, there's no limit to how many you can use if you want to, but you can make a fully fledged stereo recording with just two. Just two, one for the left and one for the right. There are a number of tried and tested ways you can configure them. I'd have to be correct here and say that you can make a better recording with more than two mics, unless of course you mess it up, which I can tell you from my own early experiences that this is very easily possible to do. But it's well worth saying that if you place two microphones in whatever configuration in front of an orchestra, you'll have a perspective problem. The instruments at the front, violins and cellos usually, will be louder and clearer than those at the back, winds, percussion basses. And who wouldn't want to hear the violas better? The solution is to raise the microphones so that they're more nearly equidistant from the front and the back. But then the problem is that this isn't how a real life concert-goer hears things unless they're flying on its trapeze. Having said that, for a small group of voices or instruments, two microphones are very often enough and very often better than if you multi-mic and try and blend things artificially into a cohesive whole. And if when you're starting out on this kind of recording, you may well only have two mics so you're getting yourself off to a good, solid stereo start. See what I did there?

As you'll know if you listen to my Sound On Sound podcasts about microphone polar patterns, shameless plug, there are four basic polar patterns, polar pattern meaning how sensitive a mic is at different angles. Actually, it's a seamless spectrum of patterns all the way from omnidirectional to figure of eight, but we normally consider just four or five. So we have omnidirectional, which in theory is equally sensitive around the full 360 degrees. I said in theory because there tends to be less high frequency content to the rear and it tends to be the case in general with all mics that the high frequency response drops off at angles away from directly head on to the diaphragm. At the other end of the spectrum is figure of eight, which is equally sensitive across all frequencies at front and back, but is completely insensitive to sound coming from the sides.

I am going to play you a walk around of a single figure of eight microphone. I'll be speaking to the mic from head on, then 45, 90, 135 degrees to the right, then directly at 180 degrees to the back. I could walk the rest of the way round and I did on the day, but it's just the same in reverse so I'll save you the time. I'm speaking in Abby Road Studio two and yes I know, I got the words wrong.

I wandered lonely as a cloud that floats on high o'er dale and hill, when all at once I saw a crowd, a host of golden daffodils.

And so we can go back into history, back to the 1930s, when Alan Blumlein invented stereo. Of course, who invented what often differs from country to country. Try television on that. But from all the references I can muster the arrows point to Blumlein as the genius we would have thanked if he'd lived to see the incredibly wide application of his work. What Blumlein invented and I say invented rather than discovered, because all of this starts from theory, is that two figure of eight microphones placed as close to each other as possible, pointing left and right, angled at 90 degrees will capture the direction of the sound source, provided though that the source is within 45 degrees to the left and 45 degrees to the right. And when this is played back on two loudspeakers, preferably with the speakers and listener positioned at the three corners of an equilateral triangle, those exact directions will be convincingly reproduced. Notice I said convincingly, not faithfully. What will be faithfully reproduced by the speakers are the amplitude and phase of the sound waves recorded from each instrument, which isn't just good, it's great. Fantastic. But, and here's the but, I said that the microphones need to be placed as close together as possible. Coincident is the word we use, but our ears are most definitely not coincident and so we have, I'm not going to call it a problem, but a disconnect. The coincidence cross pair, a figure of eight microphones, doesn't tell our ears the whole story. It's good, but not a hand rolled cigar direct from Havana. So what are we to do? Compromise of course. The recording engineer will compromise, the producer and label will approve his or her choice and the listeners get what they get and without further information they don't know what compromises have been made.

Let's look at and listen to these compromises, starting with the coincident crossed pair of figure of eight microphones angled at 90 degrees to each other. Classic Blumlein. The coincident crossed pair of figure of eight microphones is as sensitive to the rear as it is to the front, so for a live recording you'll be picking up plenty of texture, shall we call it, from the audience. Under studio conditions, this can be perfectly okay, but there is still an issue. Remember that I said that the figure of eight crossed pair only works between 45 degrees to the left and 45 degrees to the right and from what I've said, it'll work that way at the back too, in opposite polarity but that's unlikely to make much of a practical difference. What about the sides, 45 degrees to 135 degrees on the right and mirrored on the left. Oh dear, they're out of phase. Sound that strikes the front of the diaphragm of the right mic strikes the rear of the left, so they move in opposite directions. This is bad, but not insuperable. All you have to do is make sure that there's nothing that you want to record within these rogue angles. It's true that reverb from the room will come from all around, but it's usually such a mush that it doesn't matter, so some care needed.

Time for a demonstration I think. What I have are two Schoeps microphones from the Colette series, the CMC6 Amplifier with MK6 capsules, switch to figure of eight. They're mounted vertically at 90 degrees to each other with the capsules almost touching. It's as close to theory as we can get. I'll go halfway around from 0 degrees in 22 and a half degree increments all the way to 180.

I wandered lonely as a cloud that floats on high o'er vale and hill
When all at once I saw a crowd, a host of golden daffodils
Beside the lake, beneath the trees
Fluttering and dancing in the breeze

So what we have is precise positioning from zero degrees to 45 degrees, then the outer phasiness over the next three examples, then back to precise positioning from 135 to 180. Of course, since we only have two channels, what was recorded at the back sounds as though it's coming from the front. Useful for radio drama apparently. I'm going to play 90 again because that and 270 are the worst positions regarding the phase issue. I will just give you the hint that this should sound bad. I know from experience that my YouTube viewers will frequently listen on their mono phone speaker and wonder why they don't hear what I'm talking about. My Sound On Sound audience on the other hand, I'm not flanneling to say you're more sophisticated because clearly you are or you wouldn't be here, but just checking, just in case. So 90 again, a couple of times to be sure.

I'm probably making this out to be worse than it is. If you're musicians or radio drama actors, are within plus and minus 45 degrees to the front or similarly to the back, then you are golden and you'll get a very nice sound, good localisation of sound sources, but not unfortunately, entirely natural.

If you like me, are bothered about the phase issue, there's a simple solution for that and it will also reject audience noise to an extent in a live recording. The coincident cross pair of cardiod microphones instead of figure of eight. I think it's fair to say that most microphones have the cardioid polar pattern or an approximation. They are unfortunately sometimes called unidirectional but in fact they're sensitive all around except from the back. They're most sensitive to the front and just 6dB down at the sides. So we could say that they're less focused than the figure of eight, but they don't have that pesky rear sensitivity. The popularity of the cardiod microphone is an indication that it really is a good all round choice. I did mention other polar patterns earlier, the usual suspects being Omnidirectional, SuperCardioid, and HyperCardioid. You can use them for stereo, but here I'll just stick to Cardioid. I'll mention Omni again in a bit. And so we have the coincident crossed pair of cardioid microphones, point one to the left, one to the right, diaphragms as close together as possible and you're recording in stereo. No phase problems and not so much noise from your coughing, cough sweet unwrapping live audience. Now the difference is that because the cardioid isn't as focused as the figure of eight, you might want to angle the mics more widely. You can read upon a whole lot of theory on this and a whole lot of maths and you'll probably come up with an angle of 120 degrees between the mics. Without saying what's wrong or right, what will happen if you use a narrower angle is that the stereo sound stage will seem narrower on playback and instruments or voices in the center will seem a little louder. Wider than 120 degrees and you might hear what's sometimes known as a hole in the middle where the level is lower. Even at an exact 120 degrees, which should be okay in theory, the common feature of microphones that they're more directional at high frequencies means that there could possibly be a slight dulling of the sound in the center. You may therefore compromise at 100, 110 degrees or so. In reality the differences are hard to hear, but well worth considering even so.

Time for an example. Here I am back in Abbey Road Studio 2, wandering around a microphone reciting poetry instead of recording Sergeant Pepper's Lonely Hearts Club Band. This time the capsules are the Schoeps MK4 cardioid. I'll go from 0 to 180 in 22 and a half degree increments.

I wandered lonely as a cloud that floats on high o'er dale and hill
When all at once I saw a crowd, a host of golden daffodils

What we should hear is the direct sound decreasing and the reverberant sound increasing as I move around from front to back. Theory in practice, if we don't quite hear this, my guess it's down to the asymmetry of and the reflections in the room. But we hear mostly, pretty much mostly what we should. In a real world project obviously we don't want to be using the angles towards the front and 45 degrees or so to the sides, but it's good to know what's going on all around, I think.

Now if the coincident cross pair of cardiod microphones is so fantastic, there must be a fly of some kind in your chicken soup and there is. The mics are coincident, your ears are not. So at the level of each individual air molecule at your listing position, everything is pretty much perfect. It just doesn't sound the way it would if you'd been there at the session or concert. So maybe we should space the mics. There are all sorts of theories on how this should be done, bearing in mind both the spacing and the angling of the mics but let's consider one that has already been theorized for us, ORTF, "Office de Radiodiffusion Télévision Française" s'il vous plaît. In this configuration two cardioid microphones are separated by 17 centimeters and angled at 110 degrees, just a little more than the width of the human head and the compromise angle I mentioned earlier. It sounds like this:

I wandered lonely as a cloud that floats on high o'er dale and hill
When all at once I saw a crowd, a host of golden daffodils

So now we have not only directionality, we have a sense of space too. Whether Blumlein would've approved is a good question that we will never know the answer to but for me, while I like theory and accuracy, I do also like that sense of space that ORTF provides. And it's worth bearing in mind that there are other theories on how the mics are spaced and angled. Basically, it's more or less space wider or narrower angles. Invent your own combination, just how you like it.

I have one more and I'm going to give you some comparisons at the end. Over the years I've known this configuration as spaced omnis or omnidirectional mics. That's fine, but spaced cardioids in my opinion, works just as well and has the advantage of rejecting unwanted sound from the rear. So the idea is to space the mics parallel to each other so you're not using angling because you couldn't with omni and let the spacing do all the work. The question of course is how wide is the spacing. The answer to that question is, you've guessed it, it depends. It depends on the width of the group of instruments you're recording and on the acoustics of the recording space. One important point though is that if the spacing is too wide, then when you play back on speakers, you'll hear sound from the left, sound from the right, but not much from the center. A hole in the middle as it's called, again sometimes corrected with a third center mic but I'm not going that far here. What I have are two cardioid mics, parallel spaced about a metre apart. They sound like this:

I wandered lonely as a cloud that floats on high o'er vale and hill
When all at once I saw a crowd, a host of golden daffodils

Well this is different. Localisation really isn't all that precise, but there's much more of a sense of space. I'm gonna say that some people will like it, some won't. For myself I'd rather like the ORTF configuration, and maybe I'd add a feeling of being there with an additional pair of widely spaced ambiance mics. So many choices.

Lastly, I'd like to compare these configurations directly. So I'll give you a medley of all four configurations at 0 degrees, i.e. head on then at 45 degrees, which is normally as widely as you'd place your singers or instruments.

So, head on coincident figure of eight, coincident cardioid, ORTF spaced cardioids:
I wandered lonely as a cloud that floats on high o'er vale and hill
When all at once I saw a crowd, a host of golden daffodils.

And at 45 degrees:
I wandered lonely as a cloud that floats on high o'er vale and hill
When all at once I saw a crowd, a host of golden daffodils.

Bear in mind that the MK6 capsule in its figure of eight setting has a rather different sound texture to the MK4 cardioid but other than that you can hear the variation in the precision of localisation, differences in the sense of space and quality and quantity of reverberation from the room.

You know, I was fascinated by sound when I was a boy and I'm still fascinated now. What we've covered here mixes theory and practice, but only in a very limited set of conditions. The universe of sound that's out there goes way beyond, but the principles and foundations provide a solid start of a great stereo recording. I'm David Mellor and the next time I use one of Mr. Wordsworth's poems as source material, I promise to get the words right.

Thank-you for listening and be sure to check out the show notes page for this episode where you'll find further information along with web links and details of all the other episodes. Oh, and just before you go, let me point you to the soundonsound.com/podcasts website page where you can explore what's playing on our other channels.