Layering the Cake

FOR ME, making a record is like painting a picture in sound. If I have said that before, the reason is that I cannot over-emphasise it. That is exactly how I think of it, and I suppose the feeling started way back when I made those first records with Peter Sellers.

Not only are we painting sound pictures, but our palette is infinite. We can, if we wish, use any sound in the universe, from the sound of a whale mating to that of a Tibetan wood instrument, from the legitimate orchestra to synthesised sounds. That may be why, of all painters, my favourites are the Impressionists - Renoir, Degas, Cezanne, Monet, Van Gogh, Sisley. It’s surely no coinci­dence that they seem to match so well, almost as visual counterparts, the music of my favourite composers, Debussy and Ravel.

The fascination of recording is that you really do have an unlimited range of musical colours to use. That’s one of the main reasons why I enjoyed working with the Beatles so much, because our success won me artistic freedom. For so long I had been the maverick at EMI, always wanting to try new things but seldom given the proper backing to do it. Now, at last, I was able to say: ‘Let’s have a go at this. It doesn’t matter about the cost - we’ll just try it.’ The bosses could always have stopped me, but by then they didn’t want to kill the goose that was laying golden discs, and I was able to explore what were then the more curious corners of innovative recording: multi-speed tech­niques (like recording something and doubling it up in speed), lifting a tape off and recording it backwards, unusual sound-effects - every possible way of building up a picture in sound.

To do those things a record producer, like any craftsman, needs tools. As I indicated earlier, the first tool that turned him into a craftsman was the advent of stereo. The primary idea behind stereo was to give the listener a broader experience; true stereo is the use of two tracks to give a spatial dimension between two speakers. But I soon real­ised that even where we were not issuing the records in stereo, the stereo recording technique could still be very useful to the producer if he simply used its two tracks and mixed them to make a mono record.

That was the very beginning of multi-track recording. Its importance was absolutely fundamental to the pro­ducer, because now he could deliberate; he could change his mind. He didn’t even have to record both tracks at the same time. What’s more, he could record without loss of quality. With only one track, he either had to record the voice at the same time as the backing, or else dub one tape on to another, in which case he lost one generation of sound quality.

With two tracks, it became feasible for the first time to record the basic rhythm and to add the vocal track later, in sync with the previous recording. The advantages of this became evident when I started making hard rock records. Up to the time of the Beatles, most records had been fairly medium pop, for which Jim Dale was my answer to Cliff Richard and Tommy Steele. But, round about the end of the fifties, the beginnings of rock arrived in England, inspired by people like Elvis Presley, and our records were getting harder, our rhythm sounds more definite. This led to much experiment. John Burgess, for example, who today is my partner, was recording Adam Faith, accompanied by John Barry, and they tried ideas like close-miking the strings, resulting in the strange sound I described earlier. Strange or not, the records sold.

I found, starting with Jim Dale and moving through the Vipers skiffle group to the Beatles, that, if I recorded all the rhythm on one track and all the voices on the other, I needn’t worry about losing the voices even if I recorded them at the same time. I could concentrate on getting a really loud rhythm sound, knowing that I could always bring it up or down afterwards to make sure the voices were coming through.

In addition, I was able to get an even harder sound, and still retain clarity, by the use of compressors. A compressor does just what its name suggests - it squeezes the sound. The loud sounds are pushed down a bit in volume and the quiet sounds up. This was especially necessary where records had a wide dynamic range, where the peaks and troughs of sounds were widely distributed, because of the problem of the needle jumping off.

The late fifties and early sixties were a great time for compressing everything. The compressor was a bit of a new toy, and often people would overcompress. I remember describing the effect of that as being able to hear things punching holes in the sound. If you had a very heavy drum track, for example, and a voice that had to go on top of it, you could compress the whole thing so much that the drums, the loudest thing on the record, would be regulating the volume. You could only hear the voice when the drums weren’t impacting. A drum has a very high-impacting sound, one which decays rapidly, and as soon as it started to decay the voice would appear, as if by magic, rather like Pepper’s Ghost in the old-time music hall.

Compressed or not, the advantages of being able to add more tracks to your original sound soon became very evident. You would be able to concentrate on the individ­ual sound of each instrument at much greater length than you could afford to give it during the recording session. Musicians’ time is money, and economics dictate that you try to record as many titles as you can in a short time. You can’t afford to be too self-indulgent about getting perfection within that time. But obviously, if you can play around with the sounds afterwards, the whole situation is changed. It was clearly an innovation that had to come, but it would need capital investment, and EMI weren’t too keen on that. I was forever saying: ‘When are we going to get some more tracks in our studios? We’re behind the times. Let’s bring the studios up to date.’ I got a bad reputation at EMI as being the guy who always wanted a bit more - and never got it!

So perhaps it was with some desperation that, amaz­ingly, EMI sent me to America in 1958, to see what the opposition was like. The ostensible reason for going was I had had a moderate success with Ron Goodwin’s ‘Skiffling Strings’. For reasons best known to themselves, the Americans retitled it ‘Swinging Sweethearts’; it had some success there too, and Ron Goodwin was sent over to do a promotional tour. I was to accompany him.

Capitol Tower, on Hollywood and Vine in Los Angeles, is a thirteen-storey building shaped like a stack of gramo­phone records. I was hugely impressed. Capitol, who had been bought by EMI a couple of years earlier, were riding on the crest of waves as dramatic as those which thump on to Malibu Beach - Sinatra, Nat King Cole, Peggy Lee, Stan Kenton. Voyle Gilmore, one of the producers, invited me to a Sinatra recording session in studio C at Capitol Tower. The songs were eventually to be issued on the album Come Fly With Me, and it was all to be done live. Sinatra was backed by Billy May’s orchestra, normally a sound concocted from brass and slurpy saxes. But Nelson Riddle had just enjoyed a run of successes, and Sinatra liked the sound of Riddle’s strings, so Billy May’s set-up was augmented by a string section.

Sinatra, accompanied by Lauren Bacall, his current lady, was the complete, efficient professional. He came into the control room, listened to the tracks being run over, and hardly interfered with the engineer, the pro­ducer, or Billy May, except perhaps to say, ‘Take it up a bit, Bill,’ or something of that sort. Mostly he would come in and say, ‘O.K., that’s it for me.’ He did very few takes, and they got through five titles in about four hours.

The only sour note came at the end of the session, when Voyle Gilmore showed Sinatra the design for the cover of the album. Sinatra exploded. He called Gilmore every name under the sun, and disappeared still fuming. I, the country cousin from England, couldn’t for the life of me understand what all the fuss was about. Then I was shown the cover, and I did understand. It’s the cover that was eventually issued, and it shows Sinatra in his broad-banded hat with a TWA airliner in the background. Sin­atra’s view was that Capitol were ripping him off by doing some private deal with TWA. The innuendo was that if the record went out like that the Capitol executives would get free publicity, free fares, or free something-or-other. Whatever the truth, Sinatra’s attitude was plain - TWA were getting a free advertisement out of his face. Soon after that he left Capitol and set up his own Reprise label.

I returned to England with a head full of ideas. The first thing that had struck me had been that the monitor level in the Capitol control room was much louder than anything I had experienced in London - but in spite of that, the sound was much cleaner, much crisper. I had a careful look at the techniques they were using in their studios, at the kind of mikes, at the acoustics. I remember thinking: Gosh, we’ve got an awful lot to learn at Abbey Road.

The second great difference was that, compared with our two-track recording on quarter-inch tape, they were using half-inch tape to produce a three-track recording. The reasoning behind this stemmed from an inherent distrust of the ‘sound from the middle’ which was a development of film techniques and had so astonished people when they had first heard stereo. Their first thought was: Well, if we’re going to have more than one track, let’s have three, and be sure we’re getting a good sound in the middle. That is something which you cannot do in stereo using only two tracks. The reason is that if you record your backing on one track and split it between the two stereo speakers, then the backing, being equally divided, will appear to come from the middle; and when you add the voice, splitting that also in order to make it come from the middle, the overall effect is mono. On the other hand, if they used two tracks out of their three to get a nice stereo sound from the orchestra, they could then use the third track for the vocal, splitting it between the two stereo channels to put it on its own in the middle, and raising its volume where necessary in relation to that of the stereo backing. It made a lot of sense at the time.

I may have returned to England full of enthusiasm for these new ideas, but it was to be five years before EMI did anything about it. When the Beatles arrived in 1963, I was still forced to record all their early songs using twin-track to produce a mono sound, with the rhythm on one track and the vocal on the other. It was a technique that was to rebound on me many years later.

Bhaskar Menon had become the president of Capitol Records in America. As a student and trainee he had spent a while with me at Parlophone, learning my end of the business. In 1976 he phoned me (I was living in Los Angeles at the time) to say that he was proposing to issue an album railed Rock and Roll Music, using some of the early Beatles numbers. He asked me if I would approve the tapes before they went out, since they couldn’t get hold of any of the Beatles, and I was the only other person of whom they could think who had been involved. So I went along to listen - and was appalled.

EMI were terrified of the Beatles, who had issued an edict that the tapes must not be touched in any way. No one was to ‘mutilate’ them, and if they were reissued it had to be exactly as they were recorded. EMI had taken this absolutely literally. They had put the tapes on a transfer machine and were going to issue them just as they were - but in stereo! The effect was disastrous. Where I had made the original recording using one track for rhythm and the other for voices, we had put the rhythm track down in volume in order to avoid distortion. The result was that the voices sounded terribly forward. Not only that, but when they stopped singing there was an awful lot of ‘dirt’ coming up on the vocal track, because when we recorded live we had left the mikes open - which of course hadn’t mattered in mono. But in stereo - ugh! And, of course, all the voices were on one side, and all the backing was on the other.

‘You can’t let that go out!’ I said.

‘We daren’t touch it,’ they said, ‘because the Beatles wouldn’t like it.’

My response was: ‘Stuff that for a lark. Let’s do some­thing about it.’

I spent two days re-dubbing the tapes, aided by the sophisticated new equipment which had never been avail­able in those early days. I filtered out the sound of the bass from the rhythm track, and brought it into the centre. I brought the rhythm track as a whole away from the edges. I brought the voices to the centre, raised the volume of the rhythm, and compressed the whole lot so that it sounded more of an entity. With a bit of echo to bring them up to date, they really sounded quite tolerable.

Of course, it wasn’t really my job. I’d long since left EMI, and I wasn’t getting paid - since they were early records, I was not even receiving any royalties on them; I just wanted to make sure that our work didn’t get mutilated. But that didn’t prevent fierce rows breaking out with the people in England. Roy Featherstone, head of EMI records, said that I shouldn’t do it, and that they would have terrible trouble with the Beatles. ‘Well, get hold of the Beatles, and tell them I’m here at Capitol, and see what they say,’ I told him. But they wouldn’t, or they couldn’t. The Beatles all had pretty heavy protective shields around themselves anyway. At least the record went out in America in its modified form, so that it was acceptable, but I have a nasty feeling that it went out in England in the original version, which is horrifying.

That was a boomerang from the twin-track days. But at last, towards the end of 1963, with the success of the Beatles adding weight to my continual demands, the EMI bosses decided to join the world of modern recording, and we got four-track. It had taken a long time, so perhaps it is ironic that Abbey Road was to become the best studio in the world - while Capitol had hardly changed. I went back there a couple of years ago, and that same studio C, where I first saw Sinatra at work, now seemed antiquated. The sound emerging from it was so old-fashioned. I felt so strongly about it that I told Bhaskar Menon: ‘For God’s sake, why don’t you do something about that studio? The studio I once thought was so great hasn’t really changed in the last ten years!’ It is now, I am happy to report, equipped with a new Neve desk and NECAM automation. But for us, in 1963, four-track was a huge leap forward. It was on one-inch tape, which was the Continental stand­ard, and we adopted it rather than the American system simply because we thought three-track was too limited, Four-track was to be our standard for many years. Nat­urally, once the advantages of extra tracks had been real­ised, we didn’t limit our demands to four. We wanted all we could get. But the manufacturers couldn’t provide us with the goods; the European manufacturers said, quite rightly, that tape quality at that time didn’t allow them to reduce the width of each individual track, and that therefore four tracks were the maximum that one-inch tape would accept. Still, we were constantly pushing them. The record producers always provided the spur, and it was up to the manufacturers to try to satisfy their demands.

When we first got four-track, it was an enormous relief not to have the worries about how many generations of sound quality had been lost which we had suffered in twin-track. But even four tracks are very few. I still have the work-sheet of the recording of ‘The Night Before’ with the Beatles. It tells me the time I started the session, which was 2.30 in the afternoon, and the way we went about it. Drums, bass and rhythm guitar were on one track; George Harrison’s lead guitar on another; the voices on a third. The fourth track was reserved for any little extras, like piano overdubs or backing voices. It was little enough to play with. After all, if you’re dealing with a group, and you want some simple thing like having them sing little ‘oohs’ in the background, that requires an extra track. If you double-track it to make a nicer sound, you will need two extra tracks.

It doesn’t take a mathematician to calculate that four-track was still terribly limited. The only way we could get round it was to dub from one four-track to a second four-track - and that, of course, meant losing sound quality. Still, it was a start, and it gave us two great advantages. Remember, four-track is simply a way of saying four individual mono recordings. The problem was how to mix one against another in perfect synchronisation. The film people had done it for years, but then they were using sprocketed tape, which was easy to sync. In the world of ordinary tape there were no such mechanical aids. What four-track gave us was simply four separate recordings which, being on the same tape, were locked together physically, so that when you mixed them to make your final product, they had to be in sync.

The second great advantage was that we could play with the sounds on each particular track regardless of what we might or might not subsequently do with the others. For instance we could equalise the sounds - that is, cut the bass, cut the top and boost the middle. We could introduce varying degrees and types of echo. It wasn’t limited to the ordinary echo-chamber echo, but included tape-delay echo, and our own invention, which we called Steed, a mixture of tape-delay and echo-chamber. We could com­press one track to give more punch to the sound. Later, there were techniques like artificial double-tracking, where the machine did the work, rather than our having to get the artist to sing against his own voice.

Perhaps most important of all, if we weren’t satisfied with one track, we could replace it without having to do the whole performance again. We could hedge our bets. I suppose it was that ability to deal with the tracks one by one which first made me think of the process as being like making a layer cake. At the bottom you put a nice heavy layer of sponge as a base. Then you put a layer of jam. Then a layer of cream. Then you top it with more sponge, or icing, or whatever. Recording is like that. Your first track is always your rhythm track. Nowadays, the various components of the rhythm will be split into different tracks, hut in the days of four-track the first one took the drums and the bass. To that you would add, on a second track, the harmonies, which might be played by guitars, piano or something else. The lead voice would go on a third. The fourth track would be for the extra little bits - what today we call the ‘sweetening’.

So now you had your four layers, linked together phys­ically on the tape, and you could play around with the volume and the dynamic range of each track to your heart’s content in order to achieve the balance you wanted. What is more, when you came to the final stage - putting those four sets of sound on to a stereo record - you could place them wherever you wanted within the stereo picture, the sound picture that people who bought the record would ultimately hear. As I explained earlier, that was accom­plished simply by the proportion of the sound you allotted to each side of the stereo picture from each of the four tracks - the proportion that would eventually emerge from each of the two loudspeakers at home.

Fired with enthusiasm for these new techniques, like children wanting more sweets, and more, and more, we soon found that the four-track system ran out of tracks all too quickly. We had to find ways of getting more, and the only means of doing that was to dub from one four-track tape to another. And we still wanted to retain our ‘stereo picture’.

One technique was to mix your four original tracks down to what would normally be the finished two-track stereo product. You would then put those two tracks on to a fresh four-track machine, which would leave you two spare ‘open’ tracks to play with. What you couldn’t escape was the loss of one generation of sound quality in the process. Again, if you were very courageous, and if you wanted a very heavy sound on one particular track, you could dub all four original tracks down to that one, provided you kept that sound in the centre of the stereo picture. Now you had three other tracks with which to sweeten it. But for the most part I would dub four tracks down to two, giving me two extra tracks. When it later came to the making of the Sergeant Pepper album, that technique was taken almost to absurdity, to accommodate all the gim­micks and weird sounds, I needed every track I could lay my hands on. And I didn’t have them

Nor was I to get them until 1967, well after Sergeant Pepper. That was the year when eight-track arrived. By that time, tape quality had vastly improved, and we were able to have all eight tracks on the same one-inch tape we had used before. But, and it is an important ‘but’, the width of each track was halved. Now you might think that since we had started off with two tracks on quarter-inch tape, there should be no reason why we should not have eight tracks on one-inch tape. There was a reason: noise.

As I explained earlier, when each track passes the record­ing head, not only does it receive the input from the microphone or microphones that are being fed on to it, it also received the effect of the various imperfections in the tape itself. These imperfections set up magnetic impulses within the recording head, which are passed back on to the part of the tape which is being recorded. That is ‘noise’ - hiss, crackle, what you will - and is obviously undesirable. Two other things follow. The narrower the track, the more relative noise is induced in the recording head; and equally, I the more tracks you have, the more noise you get in your final product. That is the price you pay for multi-track working. Because, don’t forget, a stereo recording is all boiled down to two tracks in the end, whether you start with two or four or a thousand tracks. The noise, unfor­tunately, does not disappear in the wash, or in the mixing. If you start with four tracks, you are going to end up with the noise from four tracks. If you start with a thousand, you are going to end up with the aggregate noise from a thousand.

We haven’t reached that stage yet! But when the tran­sition came from eight tracks to sixteen, and subsequently to twenty-four, which is now pretty well the standard equipment, the noise problem could not be ignored. Some­thing had to be done about it. The man who came up with the answer was an American named Dr Ray Dolby. His invention was called the Dolby Noise Reduction Process. It was first marketed in this country, because although he was an American he found the British much more ready to listen to his ideas.

The technology for his process had probably existed for many years, but it hadn’t been used for the simple reason that it wasn’t needed. The European tape-machine man­ufacturers were always very conservative. They designed machines to cope with the existing tape technology. That is why, when they built the first four-track machines, they went straight from a quarter of an inch to one inch, rather than a half, in order to keep well within the tolerance limits of the tape then in use. But when we came to sixteen tracks, on two-inch tape, it became very evident that they could no longer cope with the problem. We were getting far higher background noise than we had come to expect. One way round that was to load each track to its maximum sound level, ‘drowning out’ the background noise. But that caused problems in the studio. Imagine, for example that you had a very quiet flute passage in a piece of music. You would put that on its own separate track. Then you might have a very heavy drum track, a very heavy bass guitar track and a very heavy vocal track. And perhaps a quiet track with a string ensemble on it. If you had recorded the piece naturally, then the bass, drums and voice would probably be fairly well balanced simply by playing them back at normal level. And if you wanted the right balance for the quiet flute, and the quiet strings, they would be way below that level.

But that would allow the noise to come through on those tracks. The answer would be to load each individual track up to the point at which it was saturated with sound, though still well recorded and free of distortion. But if you did that, the flute and strings would be uncomfortably loud in relation to all the other sounds. In cutting down your noise problem, you would meet a problem of balance. So, if possible, you should record fairly naturally.

The solution came from the good Dr Dolby. Put as simply as possible, this is how his process works. When the signal is taken from the recording studio and fed into the Dolby unit, the Dolby automatically raises the level of the quieter sounds, keeping the loud ones at their existing level. It also does something else: it splits up the audio range into three chunks.

Imagine the overall sound of an orchestra. The Dolby, listening to it, will say: Tm going to take the bottom end of the audible range, from about 20 cycles up to about 800 cycles.’ That is the really low end of the audible spectrum. ‘That,’ says the Dolby, ‘I will treat as one module of sound. Next, I’ll take the middle bit, from about 800 to 4000 cycles. That I will call the middle sound. Thirdly, I’ll take the range from about 4000 to about 20,000 cycles, and call it the top sound.

‘I will treat these three bands of sound separately. Within each band, I will allow any loud sounds to go through as they are. Any really quiet sounds I will boost a lot - say by about fifteen decibels.’ That, of course, was exactly what you didn’t want in the studio. Don’t worry, that was just the input. The Dolby unit had a second function. After all those signals had gone on to the tape, it reversed the process. Everything that had been boosted up was then taken down by the same degree. The result was that what you heard off the tape was identical to the sound you would have heard in the studio.

In effect, what it meant was that the quiet sounds that went on to the tape weren’t quiet sounds at all. They were amplified, drowning out all the noise. And when they were reduced again, they reverted to pure, quiet sounds, and the noise was reduced that much more.

The critics of the system claimed that it tended to colour the sound, and that something was lost in the process, but frankly, I have rarely been able to hear the difference. What did worry me was the complication, and particularly the expense - especially when we went to twenty-four tracks - because each track needed its own Dolby unit. Dolby equipment also needs a lot of lining up, and constant adjustment and maintenance. In my own studios at AIR I know that this is done. But I found that some American studios were apt not to take the same care, with the result that Dolby units there did colour the sound to a degree that even I could hear. So I stopped using them when I recorded in America.

There is also an important qualification to be made about them. They do not eliminate noise altogether; they don’t claim to. What they do is, as their name implies, to reduce it. So the arrival of Dr Dolby’s invention didn’t mean that you could stick to, say, eight tracks and simply re-dub and re-dub. And the addition of more tracks always aggravates the problem, which is one reason why I am sceptical about the need to go to thirty-two tracks, as many people are advocating. Personally, I don’t see the need for any more. Besides, the change means a whole new set of equipment. We would have to go from two-inch to three-inch tape - and we would certainly need different recording heads on all our machines.

If people really feel that they have to have more tracks, then the simplest way out, which we have already tried and which works beautifully, is to synchronise two twenty-four-track machines together. Allowing for the fact that you have to leave one track on each machine free to carry the pulse which synchronises them, that gives you forty-six tracks. Lord knows, no one should need more than that!

This technique has another advantage over building, say, a single forty-eight-track machine, and that stems *^rom the actual way in which we tend to use multi-track recording. On the ‘layer-cake’ system that I was describing earlier, we usually put down the basic tracks first and then start adding other bits to them. Each time we do that we have to run the tape back and forth, and in the course of a few weeks of recording, with all the various alterations one may wish to make to each track, the tape may have to pass the recording heads thousands of times. Even with modern quality, there is an obvious danger of tape wear, and although we haven’t had great problems so far, we are reaching the point where we could have. So if someone really wants more than twenty-four tracks, I think it is much more sensible to mix the first lot down to a rough twin-track, which you put on to a second machine for monitoring purposes while you start again with a virgin tape.

When you come to mix the whole thing, of course, you can go back to your original tape as well as the new one, since they are linked electronically by the pulse track which you have put on each. In the old days of four and eight tracks that was simply not possible; once you had dubbed your first tape down to two tracks and transferred them to a second tape, you were irrevocably fixing the balance on those two tracks. There was no going back -and even if there had been Dolby units to eliminate all the noise, that would still have been a major disadvantage.

There was another: the inability to do double-tracking, in most cases, because of the sheer lack of available tracks. I have mentioned double-tracking briefly already. It was something we found out ourselves, by experiment. We discovered that the double-tracking of voices or instru­ments gave them a different sound. In other words, if you record Fred Smith singing a song, and then re-record him singing the same song in the same way in time with his own first recording, it will be different from having two Fred Smiths, identical in every way, singing at the same time.

Why this is so is not absolutely clear. It may be partly to do with the cancelling-out of vibratos. It may be some­thing to do with being in and out of tune, since nobody actually sings the same song twice in exactly the same way - a fact which can have odd implications. I was working with the Bee Gees recently, and Barry Gibb in particular was a great stickler for being dead in tune-1 actually found myself telling him that by being so exactly in tune he was tending to spoil the nice parts of the double-tracking. He was so accurate it sounded almost like a single track.

Not all voices, though, are suitable for double-tracking. In fact, there is a kind of inverse law: the better the voice, the worse it sounds when double-tracked, and vice versa. Frank Sinatra, for example, doesn’t sound very good when double-tracked. Billy J. Kramer, on the other hand, had to be double-tracked nearly all the time.

Nowadays it is done a great deal. Elton John quite often double-tracks. Paul McCartney’s ‘Live and Let Die’ was basically double-tracked. Paul does it so accurately that it almost sounds like a single voice, though it still gives a stronger and better sound. Like all effects, it can easily be overdone, but handled properly it is a very useful tool.

Artificial double-tracking, or ADT, was the work of one of the backroom engineers at EMI studios. Ken Townsend, who is now head of Abbey Road studios, was on a lot of my sessions in the early sixties, and he saw how much time and effort was being spent on double-tracking voices and instruments. One night in 1964 he thought about it as he went home after a Beatles session. He had the idea that, if you could take the signal off the recording head of the tape machine as well as off the playback head (rather than off the playback head alone), and delay it until it almost coincided with the signal from the playback head, you might get two sound images instead of one. Moreover, by varying the relative distance of the two images, adjust­ments could be made until it sounded really good.

The original device was quite a lash-up. The process needed a couple of other tape machines as well as the multi-track one, and of course it could only be done at mixing time, since it involved manipulating an existing recording. But we found that varying the speed of the second image gave a bonus. It varied the frequency, thus making the second voice a little different from the first. The variation in distance could take us from a long echo effect, through ADT and into phasing, until the two images become one. ADT is like having two photographic slides overlapping and moving fractionally in relation to each other.

Unwittingly, I coined a new word in our technical lan­guage. When I first tried ADT on John Lennon’s voice, he was knocked out by it. What was it? How was it done? I replied in gobbledygook: ‘Well, John,’ I said very earnestly, ‘it’s a double-bifurcated sploshing flange.’ He knew I was putting him on, but he always referred to it as ‘flanging’ the voice. Many years later I was in an American studio and heard someone using the term. I asked where he had got it from, and was given the explanation that it referred to varying the speed of the tape in the early days of recording by placing a thumb in the flange of the tape spool.

These days there are many sophisticated devices that do this kind of trick - ADT, flangers, phasers and others, besides Harmonisers and so on. And flanging now means something a little different from ADT.

Double-tracking is clearly an extravagant tool if you only have four or eight tracks with which to work. And although it was another reason for wanting - no, needing - multi-track recording facilities, even today’s twenty-four tracks require a fair degree of discipline. Self-indulgence in one area may demand economies in another. For exam­ple, I tend to be quite extravagant in my use of tracks for rhythm. I usually have the bass drum on its own track, then two tracks for the stereo overhead sound of the drums, in order to get an ambient ‘feel’, and a fourth track for the snare drum. That’s four tracks for drums alone. Bass guitar normally takes up just one track; but some­times, if I’m a bit concerned about that sound, I also record it from the sounds of the studio coming out of the amplifier, while another track is recorded by direct injec­tion from the instrument, without going through a loud­speaker. I usually record the piano in stereo, which takes two more tracks. It may not end up as stereo, because I may bring the two parts together later, but at least I know I have the option of spreading the sound of the piano, if I feel I need to fill out my ‘stereo picture’ with a ‘wash’.

So that might be a total of eight tracks used up already, before even thinking about vocals, or brass, or strings, or any backing voices, or any other instruments. But we’ve still come a long way in the decade or so since the Beatles, Geoff Emerick and I made Sergeant Pepper on four-track. Sometimes, looking back, it seems hardly possible. We did though and how we did it is a story in itself.

Date: 2015-02-28; view: 701