General discussion: How many alternate takes?

[mmilovan]

Listening to some issues, I've noticed one fact that is not clear to me! For an example, lets have a look in one of my favorite albums: OJCCD-234-2 - Bags Meets Wes! Seven tunes and for some it is displayed clearly from what take chronologically they originated (for "Stairway to the Stars", there are two issued takes on this CD - take 3 and take 2, for "Delilah" Fantasy issued only take 4 and take 3 - in that order, and so on).

What happened with lower take numbers (such as for "Delilah" - no trace of takes 1 and 2)! Were they really takes or breakdowns? Did they erased them from original master or other secondary tape?

The same question occurs for "Jingles" - issued takes have serial numbers eight and nine!!!

So, can we hope for more alternate takes music material in years to come (I am asking this question for any other releases by any other record company), or this is it, and there is no more?

Did musicians way back then recorded so much takes (about nine of them)?

(I know somewhat similar story from 78's era, when they dubbed grooves directly into the wax or lacquered surface, and according to that story, wax was often polished or melted if something went wrong while recording the tune, so wax was reused for another take. Still some alternates survived for us. But did the same happened in era of magnetophone recordings, or tapes were simply dumped in trash or erased)

Of course, not listed here is most extreme example that I find in some Art Peper OJC - 46 takes! Really 46? Well I can not figured it out how they recorded 46 takes. And if they did, it was a bit weird thing going, the whole concept felt after 5th take, lets say...

Edited December 25, 2003 by mmilovan

[king ubu]

Milan, some labels also used one row of take numbers for a whole session - so, for instance, the first tune would be tks 1-3, the second tks 4-11, the third 12, the fourth 13-15 and so on - this practice might explain some incredibly high take numbers.

Where do you get the info on takes from, by the way? Online discographies? Or are you the lucky owner of one of those unaffordable things?

ubu

[mmilovan]

Flurin,

yes, it probably can be the real answer. Honestly, I never thought about such possibility.

For all the takes and CD's I discussed, info about takes was written on back cover. Don't know the situation right now - does Fantasy clearly stamped information about takes, as they used to do back in 1988 or so.

[mikeweil]

A complete lsiting of all issued takes of Wes Montgomery's sessions with Milt Jackson are in the discography of the Complete Riverside Recordings box set.

Wes was particularly self-critical and often asked for another take when everybody else was satisfied - "Wes doesn't like his solo" is one of Keepnews' most frequent notes in the session logs. In the case of Wes, alternate takes always merit a listen as he completey changed his approach to a solo every time, same goes for Charlie Parker.

The accuracy of a discography depends on many factors: the existence of session logs, the way they were written down in the first place; sometimes they used a new take number even if it only was a false start.

Tapes with unused material were stored and kept, as is the case with Miles Davis' Columbia material, were every fart he let go in the studio still exists, but in most cases, and the more so with smaller labels who could not afford storage facilities, they were destroyed, lost, or the tape was re-used.

In the case of Wes' Riverside material, I think everything that merits a release is available - and just as often the original session tapes are lost. The Bags meets Wes, Groove Brothers, Fusion, Live at Tsubos's and all organ trios are those were all session tapes still exist.

Fantasy is not consistent in giving take numbers in liner notes or on the CD inlay card, sometimes they do, sometimes not. It took me quite a few hours to figure out which takes of Wes' last Riverside organ trio sessions are on what specific release.

[mikeweil]

In the case of Wes, the discography on jazzdisco.org gives a pretty accurate listing of all takes issued.

Edited December 26, 2003 by mikeweil

[Christiern]

• As has been pointed out, some producers give a new take number to every scrap--false starts, incomplete takes, etc. In such cases, it is possible to reach take 36 without, in fact, having much usable material. In the case of Miles Davis at Columbia, I never understood how Teo figured it out and came up with something that sounded seamless. I recall seeing him with Miles session tape piled high, getting ready to sort it out into something coherent. I really can't see how these bits and pieces could have been assigned take numbers.
  
  Some producers did not give takes consecutive numbers, so what eventually was released as the first take might have been the seventh. This was common practice at Columbia during the period of transition from 78 rpms to LPs. The practice of recording directly onto wax-like discs gave way to the use of 16" acetate discs that could hold several takes. Before that, the 78 rpm discs could only be played back once in the studio to check the balance. Having been played, were useless for mastering, so any balance problems were corrected, the disc was recycled and the session proceeded. Thus false starts were automatically discarded in the early days. I guess some producers may have re-used tape, but I never saw that happen, not at Riverside nor at Prestige. We hear a lot about tape vaults, and it is easy to imagine some kind of steel room with a combination lock, but the fact is that master tapes often were handled very casually. Columbia stored (and probably still does) its tapes at Iron Mountain, a actual mountain storage that also houses important corporate documents for banks, insurance companies, etc. At Riverside, we simply had a large closet with shelves of numbered tape boxes and an air conditioner that kept the temperature right.
  
  While I don't recall master tapes being thrown away, I seem to remember that we reused our safeties. I also recall that we physically cut out of the original tape the selections used, and assembled them for the master tape. This left reels of tape containing everything but that which had been released. Remember, I am talking about methods I observed and used prior to the introduction of digital recording. Today there is no problem making exact copies, but an acoustic dub was always a degenerating process.
  
  BTW, except for very rare instances, studio talk and extraneous activity was not preserved until acetates came into use. I recall finding a perfectly good, yet previously unissued Billie Holiday selection on an acetate--it turned out that it was not issued because it exceeded the time limit of a 78 rpm disc (Lester Young played one chorus too many). I included it on a reissue album.
  

[brownie]

[bTW, except for very rare instances, studio talk and extraneous activity was not preserved until acetates came into use. I recall finding a perfectly good, yet previously unissued Billie Holiday selection on an acetate--it turned out that it was not issued because it exceeded the time limit of a 78 rpm disc (Lester Young played one chorus too many). I included it on a reissue album.

Chris, thanks for your contribution. Care to tell us which Billie Holiday/Lester Young unissued alternate you unearthed and on which album it was finally released?

[Alexander]

Yes, thanks for the "inside" info, Chris! Very interesting!

[Christiern]

Chris, thanks for your contribution. Care to tell us which Billie Holiday/Lester Young unissued alternate you unearthed and on which album it was finally released?

• It was "All of Me," from a March 21, 1941 session with Eddie Heywood and His Orchestra (Shad Collins, tp; Leslie Jonakins and Eddie Barefield, as; Lester Young, ts; Heywood, p; John Collins, gt; Ted Sturgis, bs; Kenny Clarke, dm.
  
  The matrix number for that selection is 29990, and the take that was originally issued bears the designation -1, but the longer performance was, in fact the first take. It appears in discographies as take 2, but things are often not what they see to be.
  
  I don't recall which album I put it on, but I'll check my LP shelves.
  

Edited December 26, 2003 by Christiern

[Shrdlu]

MMilovan, I am pretty sure that Riverside started at take # 1 for each tune, rather than using the Blue Note system of using the next number after the last take of the previous tune.

Anyway, it needs to be realized that, not every take is a complete performance of the tune. We have "Jingles", takes 8 and 9 on the CD. These are probably the only two complete takes of that selection. The other 7 are probably just short snatches that were stopped because someone made a mistake, or because the feel wasn't right. Guys like Wes and Milt (and the rhythm section) normally wouldn't need to make many complete takes. Just think about a live performance. You only get one chance with each tune, and if you're a reasonable musician, you can play a good version. I'm speaking as one who has played for many years.

If a group DID play a piece 8 times, they would probably get very stale after two or three goes. How many fresh solos on one tune can a guy come up with at the one session?

Of course, if you have a large band (say the Mulligan Concert Jazz Band) then you are going to have to make several takes of each piece in order to smooth out the ensemble parts. I recently saw the discography for the famous "Peter Gunn" album, and my eyebrows went up when I saw that the title track was something like take 16! That tells us that our Henry was a perfectionist - the end result sure was a masterpiece of an album.

[mmilovan]

• Some producers did not give takes consecutive numbers, so what eventually was released as the first take might have been the seventh. This was common practice at Columbia during the period of transition from 78 rpms to LPs. The practice of recording directly onto wax-like discs gave way to the use of 16" acetate discs that could hold several takes.
  
  I recall finding a perfectly good, yet previously unissued Billie Holiday selection on an acetate--it turned out that it was not issued because it exceeded the time limit of a 78 rpm disc (Lester Young played one chorus too many). I included it on a reissue album.
  
  

Chris,

many thanks for those enlighten rememberings about new material you found then, back in 1970's. That take was real great one!

If I can understand whole thing correctly, Columbia did, during 78 rpm era, safety 16" acetates for everything that went on studio, right? So, that explains how that particular take of "All Of Me" turned out in today's Columbia's release of BH material - we can hear studio talk (they talked about execution of tune - that exceeded particular 3 1/2 minute), if I understood a somewhat mouldy sound between Lady and studio guy.

Did I heard correctly strange noise artifacts bouncing across that take in 33,33 rpm like style - was that acetates 33,33 rpm?

Can we expect more acetates of old recordings to became available in future?

Edited December 28, 2003 by mmilovan

[Christiern]

If I can understand whole thing correctly, Columbia did, during 78 rpm era, safety 16" acetates for everything that went on studio, right?

• Not quite. Through most of the 78 rpm era, acetates were not used. Recording alternate takes was, in a sense, done in order to have a "safety." In other words, some alternate takes were gratuitous, and the main reason for this approach was that--as I said--one could not listen to an original wax-like disc without destroying it--thus it was not until a test pressing was made that one actually heard an releasable result.
  
  Acetate discs changed that, and they replaced the old 3 to 5-minute discs until magnetic tape became the standard. The reason you hear studio talk and take announcements is simply that the disc's capacity allowed it, and the performance could be lifted from the acetate disc for mastering.
  
  BTW, acetate discs are normally aluminum-based, but the war effort saw aluminum replaced by glass, so the wartime discs were quite breakable. To show you the frugality war restrictions called for, I found a Billie Holiday session on a disc that had, on the other side, Ed Murrow reporting from London.
  

So, that explains how that particular take of "All Of Me" turned out in today's Columbia's release of her material - we can hear studio talk that execution of that tune exceeds particular 3 1/2 minute, if I understood a somewhat mouldy sound between her and studio guy

.

• Yes, and I gather that "mouldy" translates into off-mike.
  

Did I hear correctly strange noise artifacts bouncing across that take on my new Columbia set in 33,33 rpm like style - was that acetates 33,33 rpm?

• The acetates I worked with were recorded at 33 1/3 rpm
  

Can we expect more acetates of old recordings to became available in future?

• It would not surprise me if there still are some acetates lying around, undiscovered.
  

[Christiern]

MMilovan, I am pretty sure that Riverside started at take # 1 for each tune, rather than using the Blue Note system of using the next number after the last take of the previous tune.

You forget one thing. Riverside existed only in the tape era. Tape introduced the capability to make physical edits. Quite often a performance was good, but the intro (or some other segment) was flawed. In such cases the flawed part was re-performed; in other cases, parts of two or more complete takes were combined to make a satisfactory whole. Some producers gave these "partials"(inserts) a separate take number, others tagged them as intros, codas, or whatever.

In other words, what you hear on a release might be the result of a combination of takes--not so in the 78 rpm era.

The following scan is an excerpt from editing notes made by George Avakian during preparation of the Louis Armstrong live in Milan album. I hope it comes out in a readable form and that it will demonstrate what I am talking about:

[Christiern]

Here's another Columbia/Armstrong example. Note how choruses were extended by combining takes, how ending was inserted, and how solo were "reorganized."

[mikeweil]

in other cases, parts of two or more complete takes were combined to make a satisfactory whole.

Orrin Keepnews revealed some of these "combined" takes in the comments to his box set reissues.

Edited December 28, 2003 by mikeweil

[mikeweil]

In the case of Miles Davis at Columbia, I never understood how Teo figured it out and came up with something that sounded seamless. I recall seeing him with Miles session tape piled high, getting ready to sort it out into something coherent. I really can't see how these bits and pieces could have been assigned take numbers.

Thanks a lot for sharing these precious recollections!

I guess we would not hold Miles' music in such high esteem were it not for Teo's edits, or at least we would view in a completely different way - the reactions to the unedited material in the recent box set reissues are telling. A lot of it sounds like jam sessions searching for some direction - and that's what it was, basically. If Teo wouldn't have done this job, maybe Miles would have focussed some more?

And we tend to overlook Macero was a trained musician and maybe live out his ambitions as a composer that way? It seems like he was never really satisfied with his own works?

Edited December 28, 2003 by mikeweil

[Christiern]

I'm glad that I was able to illustrate my point, I had forgotten that I have a file of such documents. Yes, I think some of the Miles Davis releases from that period reflected as much Teo's creativity as they did Miles'.

[Christiern]

I don't have the soundtrack album or film handy, but from the looks of the editing transcript (which I do have a carbon copy of) it looks like someone wanted to protect bop!

Edited December 28, 2003 by Christiern

[Big Al]

So this may sound like a naive question, but which was more common (once magnegtic tape arrived): issuing complete unedited takes, or issuing edited takes with all the fluffs and errors removed and replaced?

FWIW, if it's the latter, I couldn't ever tell; the edits all seem very smooth to me. F'rinstance, in the mammoth Gerry Mulligan/Ben Webster reissue, there's the released "master" take of "Go Home," which is not a complete unedited take: the ending is spliced in to replace a rejected ending. To these ears, the edit is flawless and makes the whole piece sound like one continuous take. So, in other words, just how common was this practice?

[Christiern]

So this may sound like a naive question, but which was more common (once magnegtic tape arrived): issuing complete unedited takes, or issuing edited takes with all the fluffs and errors removed and replaced?...So, in other words, just how common was this practice?

• Based on my experiences, using an unedited performance was more common, and when edits were made, they usually consisted of replacing a flawed intro. I have seen a bad solo chorus removed or replaced (not always possible) in cases where the overall performance was simply too good to reject.
  
  Remember that the early years of tape allowed only two tracks, which meant that edits were physical (cut and splice), so complex manipulation was not possible. When multiple tracks became the norm, it was possible to make some rather amazing electronic modifications, seamlessly. I recall a Laura Nyro session where about 50% of her vocal was re-done without any changes to the instrumental tracks.
  

[king ubu]

Chris, thanks a lot for sharing your "insider" knowledge! Highly appreciated, and very interesting to read!

ubu

[Big Al]

• Based on my experiences, using an unedited performance was more common,
  

Thanks. I always wondered about that.