|Pink Floyd - Paris, France (Alphi-Neonknight-Jimfisheye) (Master) (24-96)|
|Recorded at the Theatre des Champs-Elysees, Paris, France. Great recording, restored from an audience recording and a radio broadcast.
From the info file:
Théâtre des Champs-Élysées
23 January 1970
This recording is a restoration of this concert from an audience recording along with elements of the radio broadcast mix of the show. This master leans heavily on the audience recording. The sound here is of an audience recording at all times including in the VIP section. Many parts of course are only the audience recording. The restoration and presentation of elements of the broadcast mix is only possible now that this audience recording has surfaced. Special thanks to Gerard for capturing and preserving this most unique show and to Philippe for coordinating this and setting the bar with the initial release! Finally, none of this would be possible without the tireless and extraordinary effort of Neonknight for all his work in coordinating this project as well as archivist and transfer engineer for the best copies of the broadcast recordings.
********** Setlist **********
-- The Main Feature
01. Daybreak pt. 1
03. Tea Time
05. Doing It!
08. A Saucerful Of Secrets
10. Astronomy Domine
11. Green Is The Colour
12. Careful With That Axe, Eugene
13. The Violent Sequence
14. Main Theme (from 'More')
15. Set The Controls For The Heart Of The Sun
16. The Amazing Pudding
-- The VIP Section
B1. The Violent Sequence
B2. The Amazing Pudding
*** Files ***
MASTER: 24bit/96kHz stereo* FLAC
CD (portable master): 16bit/44.1kHz stereo* FLAC (24/96 master > 24/44.1 > 16/44.1)
* The mono program is presented as dual mono in the stereo format. This is to avoid it playing only through the center channel of a 5.1 home theater system or a single channel of a stereo system. This does not increase the file size with FLAC compression.
********** Sources **********
*** Lineage ***
Audience > Philips EL-3302 > Master Cassette > Nakamichi DR-1 (azimuth adj.) > Thomson/Hama Scenium KD4112 cable (unbalanced) > Roland Edirol R-09HR > 24/96 WAV
Used in all tracks. 100% of content presented.
FM radio broadcast 'source 1'
Low Gen Cassette (TDK SA90) > Nakamichi 581 > Edirol UA-1EX > Audacity 1.3 > Wav (24bit/96kHz) ----> Samplitude10 (32bit/96kHz) > Wav (24bit/96kHz) > CDWave (1.98) > FLAC
Used in tracks 1,2,4,11,12,14,16 for mostly vocal reinforcement (plus trombone, vibes, and a few drum hits and other nuance).
AM radio broadcast 'source 2'
R2R > mic in front of R2R speakers > cass[?] > CDR(?) > FLAC
Used in tracks 8 & 15 for mostly vocal reenforcement.
Re-broadcast on Tout Peut Arriver, Europe 1 Radio on 1 May 1982
1st gen cassette (BASF Ferro Extra normal position) > Technics RS-B965-M > Focusrite Saffire Pro 14 > 24bit/96kHz WAV (Audacity 1.3) > FLAC
Used in track 2 to reinforce a few kick drum and tom hits.
Re-broadcast by French radio station Europa 1 on FM 30 April 1995
Cass(1) > CDR(low) > 16/44.1 WAV(EAC) > FLAC
Used in tracks 13,16
Mastering with ProToolsHD v9 at 24 bit 96kHz
*** Broadcast mix ***
Paris radio recorded a mono broadcast mix. This was broadcast in edited form at least twice and then just a song or two and with interviews at least twice. The different broadcasts are each uniquely incomplete and the remaining copies of the 1st 3 are extreme low fidelity.
This mix was heavily compressed (as is common with radio broadcasts). This had a serendipitous effect of preserving a surprising level of nuance during some of the quietest sections. There is fairly heavy delay effect during some parts. This may have been added in post production but it's possible that they had an fx feed that they simply mixed too heavy in the broadcast mix and which was further exaggerated by the compression. This element of the mix is very crudely executed and does not follow the live mix. Hard to tell if it's an echo applied to the entire mix or just the drum/instrument bleed in the vocal mics. You do hear drum bleed getting treated with vocal echo at times during this or any Pink Floyd show but this goes far beyond that to the point that it significantly alters the music.
Broadcast mixes of this era were often made by a radio station engineer mixing the individual inputs from the stage on their own console. Often from a 2nd set of mics (which you can see in many videos from the era) as equipment of that era was not easy to interconnect and make multiple mixes with. Since the mix made for the room can often omit sounds that were plenty loud in the room just coming off the stage, the broadcasters new this and made their own mix in order to better capture the event. The down side is that for bands like the Pink Floyd who had very accomplished mix engineers, the broadcast mix does not reflect the house engineer's work. Further, bands who had high standards for sound employed musicians as mix engineers where most radio engineers were not as accomplished and as such made very crude mixes. One of these explanations would be the reason for the unmusically awkward delay fx that are hot in the broadcast mix.
*** Audience ***
Recorded by Gerard on a portable mono recorder. He was able to successfully capture all the music except the first notes of GITC, AD, & MTFM and two roughly 30 second breaks for tape changes in ASOS and TAP. He paused between songs whenever possible to conserve tape.
This recording sounds like it comes from more or less the center of the room. The stage sound is distant but still fairly well defined. You do not hear any evidence of close proximity to any of the surround speakers which further suggests a centralized location.
This recording suffers from the difficulty of recording a very dynamic performance with very loud and very quiet sections with portable equipment nowhere near up to the task. As such, a great deal of the sound is weak and buried in the noise floor of the recording. This hiss is further exaggerated after the record level was turned down on the master during BMT (no doubt because the loudest peaks were hitting the red). There are also noises from the microphone being bumped and the chair creaking throughout the show but especially during the quiet opening number. Some of these are so severe as to sound like your house crashing down around you when you bring the music level into balance.
*** etc ***
If you study the audience recording you can tell in spots where mix elements were sent to the surrounds vs. the mains either from proximity cues and/or altered room echo. But just barely as the taper really appears centered in the venue. For all intents and purposes, all the surround mixing from this show is lost.
The audience recording is far superior to the remaining original broadcast captures in both fidelity and balance. The 1995 re-broadcast contains elements that are captured in much higher fidelity than the audience recording but the audience recording is much better balanced due to the gross alterations of the broadcast mix.
********** Mastering notes **********
The goal was to restore this show as completely as possible from what remains and leave no reason to listen to the raw recordings for any missing content (which was always a difficult exercise with these sources in raw form). We have the complete show (minus a few missing moments) from the audience with a substantial level of dynamics and detail preserved but clearly distant. We have the 1995 re-broadcast of two songs that tease you with surprising sound quality and nuance but which are also wildly altered dynamically, contain no ambience, and are littered with echo effect artifacts at times. The original broadcast, between the original broadcast quality and the subsequent recording and copying of it is all but destroyed. The high frequencies have been truncated to the point where there are almost no cymbals left in the mix. However, even with this level of destruction, some elements of the sound were so strong in the signal (like vocals, due to them coming from the stage microphones and running directly down the wire) that they still remain. There are elements of the sound preserved this way that were too distant and weak to capture by the audience recorder. With the quietest parts preserved by the broadcast mix I was able to bring some of the poorest preserved moments of the audience recording of the show up to the quality level of some of the better preserved moments.
The mono audience recording was transferred with a stereo cassette deck. The left channel capture is superior to the right. Only the left channel was used as the right adds nothing but increased noise and causes phase issues.
All speed correction work was done with the Serato Pitch 'n Time plugin.
I removed a substantial amount of tape hiss with the iZotope RX2 denoiser plugin. For historical recordings such as this, my goal with noise reduction is to do no harm to the remaining content as opposed to achieving commercial standards for noise reduction. As such, a lot of tape hiss remains in this recording. The level of noise reduction varies with the content with what I could get away with.
Well over 100 mic noises and chair creaks were removed with the iZotope spectral repair plugin. There are a number of tape dropouts (and dropout like artifacts from transfer issues with the old tape - the tape would snag and drift across the play head) that were repaired as well as possible with a combination of using content from the right channel, using nearby content to reconstruct (a couple milliseconds for this kind of thing or less), or severe level boost along with severe noise reduction to balance it with the surrounding content. Where seamless repair was not possible, flaws were kept as opposed to replacing original content with large edits with only fidelity in mind. No alterations to the original performance were made for any reason.
The high frequency attenuation due to the recording quality was corrected as much as possible without increasing the overall level of hiss in the recording. I avoided EQ artifacts with parallel processing by cutting all but the range I wanted to boost with the Waves linear phase EQ and folding this copy back into the original. The recording was extremely saturated at 175Hz and weak at 4.5kHz. I used the UA Precision Multiband to dynamically correct this.
The poorer original broadcast material mostly only offered vocal help. The two songs included in the 1995 re-broadcast were only used sparingly in The Main Feature to strengthen the weak moments of the audience capture without eclipsing the rest of the show with superior fidelity. The awkward delay artifacts (especially in the drum transients) were removed with the iZotope spectral repair plugin as well as possible. The radio whistle artifacts in source 2 were removed with iZotope as well. I synced the remains of the broadcast mix to the audience recording using the Serato Pitch 'n Time plugin. Sync points ranged from 5 to 15 second sections as needed for the variations between analog sources. Useful elements of the broadcast mix were isolated to support the audience recording without adding any noise or artifacts. Restoration and isolation with volume automation and UA Precision Mastering EQ. Noise reduction was used to remove some tape hiss. All 3 sources of the broadcast mix were transferred to a stereo cassette deck in their lineage. The left channel capture is superior to the right for all 3. Only the left channel was used as the right adds nothing but increased noise and causes phase issues. The source 2 broadcast mix transfer was upsampled to 96k.
The missing section in the audience recording of ASOS was complete in the broadcast mix. Elements of the sound that are not present in the broadcast mix were reconstructed under it from nearby parts of the audience recording to restore a more consistent fidelity. The missing section in the audience recording of TAP and the missing opening notes from GITC and MTFM are absent from the broadcast mix (they were in a parts that were edited out of all broadcasts). The arrangements were restored by recycling nearby parts. While not technically authentic, it is far less distracting and closer to authentic sounding than the sound cutting in and out arbitrarily. I recycled less than a second of the keyboard opening of AD for the same reason. The last 5 seconds of TAP are only the 1995 re-broadcast.
********** Welcome to the VIP section **********
The two songs included in the 1995 re-broadcast allowed for a higher level of restoration. These two songs are duplicated, mastered to the highest quality possible, so you get 'VIP seating' for these two tracks and a taste of what could be if someone were to discover the master of the broadcast mix.
********** VIP mastering notes **********
The high hat during the choir section of TAP (almost inaudible in the audience recording) has been recreated from nearby content for the choir section missing from the broadcast mix. The missing main theme section from the 1995 re-broadcast mix before the drum solo has been reinforced by the slight amount possible from the original broadcast as well as reconstructed under that from nearby parts of the 1995 re-broadcast to restore a more consistent fidelity. This to preserve the original performance as opposed to repeating the entire section from a different part of the arrangement. The drum solo was edited shorter in the original broadcast and edited out entirely in the re-broadcast. There was just a little low mid content in a few of the tom hits in the part of the drum solo broadcast on AM that was useful. This helped ever so slightly but the drum solo is still noticeably further away.
********** Philippe's original notes **********
This recording comes from a 40 years old master tape and has never been circulated before. It must have a few glitches from times to times, but it's a miracle that it was not completely gone after all this time. In particular the song Cymbaline has some gaps, but luckily I've been able to play it completely after many attempts since the deck was often stopping the tape during that part. Sadly the very end of The Amazing Pudding is cut because of another concert that was overtaped.
Big thanks to Gerard for digging out his tapes collection, and coming home with some of them, and of course for taping this show at the time!
A soundboard recording also exists, as the show was recorded for French radio Europe 1's Musicorama program. Many recordings based on various FM broadcasts are in circulation, however this is the first time a complete audience tape surfaces and gives us the complete and correct running order of all the songs.
Pink Floyd also played the same venue the day after, 24 January, but no recording of this second performance is to be known. The taper did not attend the second show since he attended (and recorded) The Nice at L'Olympia, Paris, that night. Sadly the very beginning of The Nice recording was taped over the very end of 'The Amazing Pudding'. Some parts between the songs may also be missing, since the recorder was stopped to save the tape.
This audience recording was initially shared in March 2013 only in portable format quality (reduced to 16 bit 44.1k).
Jimfisheye, August 2013