1970-February-11 Town Hall Birmingham, England Disc 1 - 33.30 1 The Embryo 11.50 2 Main Theme From More 11.27 3 Careful With That Axe 10.12 Disc 2 - 72.37 4 Sysyphus 11.39 5 Heart Beat, Pig Meat 5.46 6 Oenone 6.19 7 Moonhead 4.21 8 The Violent Sequence 7.59 9 Set The Controls 13.34 10 The Amazing Pudding 22.55 Lineage: Pink Floyd 1970-02-11 Birmingham, 2nd gen stratcat58 tapes (16bit/44.1kHz) 2nd gen Maxell XLII-S and XLII cassettes (bet.1992-96) > Technics RS-B565 > Focusrite Saffire Pro 14 > Reaper v4.76 > FLAC (24bit/96kHz) > vince666 > header change to 88.2 Khz > WAV 88.2Khz/32bit float > speed correction > WAV 88.2Khz/32bit float > resampled to 44.1Khz and dithered to 16bit > WAV 44.1Khz/16bit > track splits > FLAC 44.1Khz/16bit Notes that came with this recording: *Neonknight's notes* We believe that this is the best copy yet of this great show. Stratcat58's 2nd gen is smoother, closer and more detailed than the 3rd gen we torrented in July. It also has better high end. From the outset it is apparent that the cymbals and hi-hats appear where they were previously buried; always a good sign. In fact you can follow all the instruments more clearly, including during the quieter passages which are less dominated by hiss than our previous copy. Side B's TAP on the 3rd gen ends fractionally later than the 2nd gen so the 3rd gen cannot be a copy of this 2nd gen version. The third gen's lineage was master (probably a cassette) > reel > reel > Nak Dragon > stratcat58's cassettes. Stratcat58 arranged a loan of the 2nd gen cassettes and the lineage for them is master > reel > Nak Dragon > cassettes. Stratcat58's source's source (the owner of the 1st gen) visited with his reel many years ago and our man made two copies, one to cassette and the other to reel. The reel was the source of the 3rd gen cassette that we posted. It's as though two gens have been shaved off, not just one, for this Nak Dragon provenance. The audience chatter before The Embryo is more complete on some other copies e.g. the taper's comment at the very beginning of the recording, "I can't get the hang of his thing", is not present here. Additionally, the opening to side B, Heart Beat Pig Meat, is a bit steadier on lordsnooty's tapes than stratcat58's and his copy has some extra tuning. Stratcat58's tapes are fuller and closer than the lordsnooty version but his copy is cleaner. It's best to keep both! The taper was probably towards the back of the auditorium and possibly had their recording levels set low, at least some of the time. The way the high end falls off indicates that basic equipment was used. The recording levels have probably been subsequently optimised during copying. It's difficult to draw any clear conclusions about the Town Hall's acoustics but they may well have contributed to the overall ambience of this recording. Those not familiar with it might be a bit put off at first but my experience is that it ceases to be distracting when your ear dials into the sound. If you are outside of the UK or have never been to Birmingham it's worth taking a moment to to get a feel for the Town Hall's neo-classical architecture by looking it up on wikipedia. Internal pictures from the time suggest that the hall was quite long and slim with a balcony at the end furthest from the stage. As I mentioned in the 3rd gen text file, here the Floyd are recycling material from the Zabriskie Point sessions and working out what to do next. Little memorabilia from the concert appears to exist. The only thing I could find was a ticket on google image search. I tried contacting the Friends of Birmingham Town Hall in 2012 and they were unable to assist. 11 February 1970 was a Wednesday and the concert started at 19.45. It was promoted by big name promotor Roy Guest, who had joined NEMS Enterprises in the autumn of 1968. Most copies of the concert from the late 1970's onwards started out from a major tape collector in San Diego who sold and traded using low quality unbranded cassettes. His collection was passed to another collector who continued distributing tapes and then another one after that. There is also a guy from Wales, DT, who was probably the source of the San Diego trader in case anybody reading this recognises themselves or can offer a lead. The guy who owned the 1st gen reel that was the source for the tapes used for this release retired from trading and moved away. *Vince's mastering notes* Most of my notes are going to be the same as what i've just described for the previous work on the 3rd gen cassettes, since they have many things in common... so, i start by "recycling" a few points from there, which are perfectly true for these new cassettes, too... especially the point which brought me to choose the final 88,2Khz sample rate for the both of them. This recording did run very fast so, before actually performing the classic speed correction process, I made a header change to 88.2Khz which automatically "forces" the audio to play at a lower samplerate (and then at a lower speed) without actually altering the samples, so, i preferred to perform this header change to simply bring the recording much closer to the correct speed with no true resampling (and digital speed correction is a resampling process). But the choice to switch to 88.2Khz needs that the whole recording will need to stay at this new samplerate (but a good point of the 88.2Khz samplerate is that the conversion to 44.1Khz to make a CD version is more easy/transparent than when you go from 96Khz to 44.1Khz, which would involve much more complex maths to do the trick as 96000 isn't a multiple of 44100). So, with the recording played at 88.2Khz, and then much closer to a correct/reasonable speed, i finally started to think about the classic speed correction process, and it must be noted that the 88.2Khz header change alone did bring the first 9 minutes (or so) of "The Embryo" just to a quite correct speed (within a very small error I will explain later in more detail) so these first minutes weren't subjected to any resampling process as I simply left them alone as they sounded quite right that way... but let's go to the actual work I did... First of all, I must say that this recording isn't just the typical "easy" speed correction work as the speed varies a lot of times throughout the whole recording and, moreover, there are a whole lot of slight speed fluctuations and also a noticeable amount of wow and flutter (and my guess is that the W&F just belongs to the previous generations of this recording, as I seem to remember it's present on any versions of this gig). But, this time, I did choose to work in an even more careful and meticulous way as on the previous one... so, in the first moment, I simply checked the speed virtually second by second by trying the needed corrections "on the fly" without applying them and so I mapped the whole recording with a lot of markers to put in evidence all the segments which needed to be corrected differently (with a lot of markers I mean a few hundred markers!). This way, with this very detailed map of the recording, I could actually correct both the "overall" speed and the small/sudden fluctuations with the first (and only) processing pass, small segment by small segment (at times, tiny segment by tiny segment). On the contrary, on the previous 3rd gen, to make the work a bit easier, I had made a first "overall" correction applied on relatively long segments and then I addressed the short fluctuations within the same "just corrected" segments while applying a second correction process over the first one. And so, the much longer and detailed checking work I had made before brought me to the final result by also processing the recording the least possible... but, actually, after I finished the whole work and was finally making a properly relaxed listening, I caught a handful short things which at that point I did correct in a second processing step... but, hey, it's a matter of a few minor things here and there which are 1 second long each, at max... so not really a big deal, but these few "extra" corrections may have been a bit distracting while listening (or, at least, they were definitely distracting to my ears)... Thus, the result of the processing itself is still more "transparent" than on the previous 3rd gen and, after all, this 2nd gen is a better sounding source just as a starting point. But, since the Birmingham recording is perhaps the most "difficult" in terms of fluctuations and tape related problems of the whole PF recordings story, of course there are still small/light fluctuations here and there and there is still the infamous wow and flutter (more noticeable on some songs, less noticeable on others). I will explain the wow and flutter and the "within a small error" point in more detail... After the initial header change, the first nine minutes of The Embryo came gracefully extremely close to the right speed... I mean that they may still be a hair fast (with "a hair" I mean a fraction of a percentage, say around 0,5%) and it would have been unwise to subject them to processing for such a tiny difference, since they just sounded very nicely in raw shape... and, while trying to find the best correction ratios for this recording I could experiment a few interesting things myself... the wow and flutter may appear to be more noticeable than on some other sources, not certainly because there is more or a deeper wow and flutter oscillation, but simply because the sound is clearer and better defined, so the instruments are more clear and detailed and, of course, the wow and flutter might also be heard more distinctly... if you think about it, the wow and flutter is a continuous and quite periodic speed fluctuation and the clearer you can finally hear the instruments, the clearer you will also detect the W&F, because if you have a confused and muffled sound, you will have the W&F somewhat masked just by the less defined sound itself, since the speed oscillation is detected just on the actual "useful" content and not on the hiss or on the background noise or on any other "non music related" noises. So, in other words, we have better sound and so we can "appreciate" the flaws of this recording in a "better" way and, anyway, I'd prefer to see quite only the "half full" glass of a better sounding tape anyday! The second point I'd like to mention is a sort of an easy way to tame what i've just explained here above about the W&F... in fact, I noticed that, when there is a noticeable continuous W&F speed oscillation, if I keep the overall speed a hair faster then the W&F speed fluctuation oscillates in a speed-range which, for example, goes from the correct speed to a hair fast or, anyway, it ranges mostly on the faster side... and, to my ears, such situation sounds noticeably better than if I try to put the overall speed to the "correct" value and leaving the W&F oscillation ranging also to the slower side as it definitely gives me the impression of a tape machine losing some power and I found it very annoying... so, in general, I tended to stay a hair fast throughout the whole recording because it makes the W&F much less annoying... so, a small error (around 0,5% or so) to the fast side sounded noticeably better than a "just exact" overall speed... and I add that, with the W&F and some minor fluctuations still present, it makes not really sense in talking about the "perfect" speed but it may be much more useful to directly stick to something which makes the recording more enjoyable, which is just what I tried to do. And, believe me, it needed quite a bit more work than the 3rd gen to make the recording how it is now and with the lightest hand I could in applying processing individually on hundreds of small to tiny segments... so, as the other 3rd gen one, at a certain (even longer to arrive) moment, I simply decided to stop working and consider it ready otherwise i would have risked to never finish to work on it as these Birmingham recordings are just those which you could keep improving for years and with countless hours of meticulous manual editing. So, the speed may still be not just perfect at moments and the W&F is still there but, hey, to my ears this time I finally have the impression of being phisically there attending the gig and it's indeed a nice upgrade in sound quality over any other circulating versions, including the recent 3rd gen cassettes which just sounded like a sort of miracle when they finally surfaced only a few months ago. No other kind of processing was done, so no level changes, no normalization, no NR, no EQ, etc... Stratcat58 cassettes / Neonknight tape transfer / Vince666 mastering, October 2017