At the height of his fame, Gabriel toured his classic album, So, throughout the world ending in an open-air theatre overlooking Athens. Recorded and filmed over three nights and painstakingly restored, this is the first time the audio has ever been released.
We are very proud to be able to offer this exclusive audio to Society of Sound members and we asked sound engineer Ben Findlay explains the process:
The original recording was made on two-inch analog tape. Because there were so many channels they had two twenty-four track machines synchronized with timecode, and the timecode was referenced to the film, so that when they were back in the studio it was possible to resynchronize everything.
The first challenge that we had was to locate the material within Peter’s extensive tape archive. The next challenge was to copy (or transcribe) the analog audio recording to a digital format. The fact that the tapes were over 20 years old meant it was necessary to bake each reel in a laboratory oven for three days at 60°Celsius. This stabilizes the glue that binds the oxide particles to the plastic backing and which over time can become gooey and unstable. Once they have been baked this ensures the magnetically oriented particles that represent the audio recording are not rubbed off as they pass over the tape heads.
The original audio recording and the cameras were synchronised together with 24 frame / second time code, so it was essential to use an audio time code synchroniser to keep the playback analogue machine running at exactly the right speed. In essence each second of time code, and it’s corresponding audio, had to last exactly 48000 samples as it played into the digital record machine. If this didn’t happen, when you played the digitized audio back with the 24 frame picture they would run at different speeds and the audio and picture would drift apart.
What’s actually on the tapes are all the individual microphone recordings that were taken from the stage during that show. So the bass drum is on a separate track, the snare drum is on a separate track, keyboards have their own channel, the vocals have their own channel. That’s what was primarily on those reels of tape. But, after the concert had been recorded, the tapes were brought back to the studio and they overdubbed parts where Peter felt they could be improved upon. So there were some keyboard repairs and guitar repairs and the odd vocal repair. Those again were on channels that weren’t utilized for each of the songs, so there was a bit of dancing around getting the right channel to appear in the right place when we started the mix.
After we had transcribed all of the tapes we were able to start laying our guide picture in with the audio, checking the sync and finally start mixing in a modern way. It was all balanced in Pro Tools rather than spreading everything out on an analogue console, as it would have been done in 1987. Mixing in Pro Tools can be a bit like trying to paint a room through a keyhole, but it has a massive advantage from a time point of view in that it allows you to go from one mix to another to another very quickly. This means that production time is reduced and that you are working from a known fixed point every time. One of the problems of mixing in the analogue domain is that if you do a mix and then you want to make adjustments to it, you have to reset all the effects so that they’re right, make sure the console settings are all exactly as they were when you left the mix the previous time. Although the mixing process that we engaged in on this project was not such a pleasant, tactile experience, it does give you a solid working point; every time you go back to the mix you know that’s exactly how it was left because it’s all existing in a computer.
One of the things that was an absolute joy was that the original recording was so beautifully executed, and the analogue characteristics of the recording, i.e. the subtle compression that happens with analogue recording, and actually the background noise that you get as well, has an impact on your listening experience is faithfully transferred into the digital recording. We set about preserving that characteristic as much as we could.
During the mixing process, one of the crucial things is to get your monitoring environment comfortable and understandable, so I would have a surround array that were Mackie loud speakers. Then for the stereos, I would have a small pair of Auratone speakers, which are horrible sounding and very limited in frequency response, but give you a sense of how the production is going to sound on television or on a small system. Then in Peter’s studio, he has Bowers & Wilkins Nautilus speakers, the big ones; they sound huge, in fact they sound like a PA in many ways, like a really beautifully tuned PA system. I would use them for big, loud listening, to get a sense of what was going on truly with the bottom end, and that was very useful.
When Peter Gabriel was finally happy with the way the mixes sounded the audio was then ready to be mastered. In this case we took the project to Metropolis in Chiswick and in collaboration with a mastering engineer Ian Cooper we eq’d and limited the audio to give all the mixes tonal and level continuity across the duration of the film. I have mixed feelings about this process. On the one hand I love it because it allows you to take a step back from the mixing process, and to work collaboratively with great engineers. On the other hand I hate it because implicit in every adjustment that the mastering engineer makes to your mixes is the suggestion that you didn’t get it right in the first place. Having said that Ian was gentle with me and more importantly the mixes sound even better when we left.