20 years of Pink Floyd's Dark Side of the Moon | Alan Parsons
Original engineer on the 'Dark Side Of The Moon' sessions, Alan Parsons, explains the 20th anniversary remastering of this classic album.
It's estimated that one in five British households owns a copy of Dark Side of the Moon. This influential 1973 LP has sold around 25 million copies worldwide and was recently remastered; sitting in on the process was original engineer Alan Parsons. Paul White caught up with him to chat about Time, Money, and technology.
Pink Floyd's Dark Side of the Moon was a rock landmark; released in 1973, it has proved to be one of the classics of popular music and on its 20th anniversary was remastered. As original engineer, Alan Parsons was asked to advise on the project; I talked to him about the original recording and the remastered edition.
What kind of equipment and techniques were available during the original recording of the album?
"There were no digital processors available back in 1973. All effects other than straight reverb were obtained using tape machines, and because of the complexity of Floyd's music in terms of the number of effects required, quite often that would mean using not just every tape machine in the room, but all the machines in all the other rooms too. This meant two control rooms tied up, every machine with a reel of tape on it, and leads trailing down the corridors.
"The recording format at the time was 16-track, 2-inch analogue. Interestingly, the first generation of each recording was non-Dolby; we went to Dolby A to keep the noise down when we went to a second generation 16-track in order to make more tracks available. All the original drum/bass/guitar/keyboard recordings were essentially non-Dolby at 15ips."
Were the songs put down more or less live or built up using overdubs?
"The band had been playing the tunes on the road before the recording, so the four of them would play the songs live with additional instrumental weirdnesses put together afterwards in the studio. We did most of the recording in Abbey Road Studio 2 with the overdubs in Studio 3. I remember using one of those monstrous cream-coloured acoustic screens to screen off the bass amp in one corner. BBC screens had just become available so we were able to screen off the drums and achieve reasonable separation. I seem to remember that the bass was the principal problem, causing spill-over to the drums, as Roger's amp was quite loud. And with that room the way it is, it's a fairly sizeable environment to have bass buzzing around in. It was controllable, but even then I wasn't bothering to stick a mic on the bass amp — I took a DI feed straight out of the instrument. It's something I've done ever since — I've never got a decent bass sound by miking the cabinet, I always got better sounds DI'ing the bass. I probably used some compression, too; I had a fairly standard policy of sticking a Fairchild on the end of a bass, set up to reduce the gain by about 6dB."
How did you approach recording Dave Gilmour's guitar?
"The guitar cabinet was normally close miked, though if it sounded a bit dead, I'd take the mic and wander out into the room a bit. You must remember that the Abbey Road rooms were all of such a size that you could absorb a bit of acoustic by moving back — if you wanted a bit of room sound it was no problem.
"There was always a Binson Echorec [magnetic disk-based echo unit] patched into the gear. The amplification was a Hi-Watt 4x12 with a separate amp top — no big stacks or anything — and Dave had one of those little EMS white guitar processors to provide some of the effects."
Was that the EMS Synthi Hi-Fli, a device that looked something like a spitoon on a stick?
"Yes, that's exactly what it looked like. By modern standards it was pretty poor, but for its time, it was quite revolutionary. I remember seeing the first phasing pedals at about that period too, and I think we used one of the orange MXR models. The only other special effects were either based around tape delays, or for EQ, we'd plug in an external EQ or a graphic. EMI used to have their own names for everything; their parametric EQ was a known as a curve bender, a great green box with ugly fat knobs on it and a 30dB insertion loss which you'd have to make up somewhere else. But it was a nice EQ, and I imagine there are audiophiles out there who would love to get their hands on one. I dare say the units are around, sitting in some garage in Greenford or somewhere."
How was the session mixed? Did you go straight to 2-track with Dolby A?
"That was actually against my wishes at the time, because I knew that there were going to be crossfades and extra generations of recording. I'd always had problems getting the Dolby lined up between different machines, and in this instance, there would be three stereo machines. I would have been in favour of having the final version non-Dolby with Dolby used on the in-between versions, but the band were keen to to use Dolby.
"We used EMT plates for reverb, and of course, live chambers, but most of the reverb used on the album came from just one EMT plate. I feel that there's a considerable plus side in just using one reverb, because you can stick everything through it, and it all sounds as though it's been played in the same environment. It could be argued that you cloud the issue by having a different reverb on every instrument."
"When we created the crossfades between tracks, it was a matter of cueing up the two stereo source machines and then simply getting it right; there was no timecode, no automatic systems, it was somebody with their finger on the button. But it was fun, and that stuff is very creative when you get it right. It's a very important part of the way the record turned out; the way that one track melded into another really worked."
This was long before automated mixing. Was it one of those sessions where the band members helped out, each with their own faders?
"There was a bit of that, but the crossfades were only four-fader operations. You have four things to get right; the level of machine A, the level of machine B, the timing of machine A and the timing of machine B. As long as machine C is in record, that's it. The quality loss caused by the extra generation of copying needed to compile the crossfade is only evident in the crossfade region; the crossfade sections were manually spliced into the original masters. That's why I was wary of using Dolby; if you did mix all the way through, no problem, but if you wanted to cut back into a tape a generation earlier, you had to hope that the Dolby and level alignment was close enough so as to cause no change in sound at the splice point. Something like Sound Tools would have been a Godsend — but then fighting with the technology was part of the creative process."
How did you feel the original recording could be improved?
"I wasn't unhappy with the recording, taking into account the vinyl medium it was destined for. The words Compact Disc hadn't even been thought of in those days, and now that CD provides more dynamic range and a wider frequency response to play with, it seemed a shame not to exploit it. In terms of the overall sound spectrum, I felt it deserved a little top end added, by EQ'ing between 16 and 20kHz. Sounds like that wouldn't have made it to vinyl anyway, so we wouldn't have been concerned that they weren't there when we originally made the album.'
The original stereo master tape of the album, rather than a CD copy, was used during the remastering. Did a tape of that age present any problems — such as the notorious 'sticky shed' syndrome, which has rendered some archive material almost unplayable?
"Sticky shed wasn't a problem, but we had a lot of sticky edits. I drew attention to this and it was subsequently taken care of, either by replacing the splices or by removing the sticky patches. When I was at Hayes with EMI, the standard practice was to sprinkle chalk powder onto the edits!
"Once the tape was playable, it was reassessed for the CD medium with a little EQ applied here and there. There was also the option of doing another crossfade between side one and side two and I haven't actually listened to the CD all the way through to see if that choice was taken up. It was allowed for, but it had to go to the band for approval. It made a very nice crossfade because the piano piece at the ending of side one runs beautifully into the loop sequence in 'Money'. In fact that was always the way it was done live.
"Nearly all the EQ changes were made at the top end of the spectrum, though for some parts we played with two bands. Sometimes we found the overall top end we were adding might make a guitar solo sound too hard, so then we might back off a little during the solos, then ease it back in. It wasn't just a matter of creating an overall blanket EQ, we were making changes throughout.
"I wouldn't have minded booking a couple of hours on a Sonic Solutions No Noise system, had more time been available, just to make sure the recording was as clean as possible. For example, there's a bit of hum on the front end of 'Money' that it would have been nice to get rid of, but that'll have to wait for the next remix or whatever. Which brings me to — yes — why don't we do another mix? I've been suggesting this for years, I can't imagine that the original 16-track tapes aren't still intact. Given today's technology, we could go back to the first generation 16-tracks and sync them up. That might give quite an improvement over the second-generation tapes used for the mix."
On a project of this importance, was there really insufficient time to employ a system like No Noise to ensure the end result was as good as it could be?
"It's one of these things — they wanted it tomorrow, otherwise it wouldn't have been a problem. To James Guthrie's [post-Moon Pink Floyd engineer] credit, he was given the job of getting the whole boxed set into shape, and I wouldn't have known about it if he hadn't given me a call and said 'Help, I'm at Mastering Lab and there's no Dolby tone on the Dark Side of the Moon masters, will you come and help sort it out?' Coincidentally, I was in California at the time, so I was only too happy to do that, and we spent a day tweaking it and had a lot of fun. I'm delighted with the results; I felt that the original CD was a great disappointment, and if I'd been given the chance to do this work 10 years ago when the CD version first came out, I would have loved to. There's nothing that we did on this project that couldn't have been done then, though we do now have better A to D converters. Systems like No Noise weren't around then, but as it turned out, we didn't use them anyway. Apathy on the part of record companies for remastering is a common gripe at RePro meetings. In The Master Tape Book we tried to impress the importance of contacting producers and engineers whenever a remastering takes place. Even the so-called audiophile mastering houses think the original master tape is the definitive statement and don't take into account the changes intended to be made on transfer."
Did Mastering Lab take advantage of super bit-mapping or one of the other sophisticated dithering systems available to increase the dynamic range of 16-bit audio?
"No, though I'm sure they're on top of developments in that area. This opens up the debate as to whether every CD ever released should now be re-released with the benefit of new technology. Sometimes I get impatient with manufacturers and the issue of technology, because it's music we're dealing with, not science. Is it really going to affect the enjoyment of the music whether an analogue recording is remastered using 16 or 20-bit technology? Ultimately, I don't really think it is."
Interview by Paul White
Previous article in this issue:
Next article in this issue:
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!