Jump to content

Recording in more than one take in the 78 era


Larry Kart

Recommended Posts

Our friend Terry Teachout in his blog today writes about Golden Era piano recordings:

and says of Benno Moiseiwitsch’s 1939 recording of Sergei Rachmaninoff’s transcription of the scherzo from Mendelsson’s incidental music for A Midsummer Night’s Dream: "This performance was made in a single take, as though this were remarkable.
Please correct me if I don't know what I'm talking about, but In the pre-tape era were not virtually all recordings of pieces that would fit onto one side of a 78 made in single takes, with each take of a longer work coinciding in length with the amount of music that would fit onto one side of a 78? If so, how common would it have been for two or more takes of a performance to be joined together to make up what was heard on one side of a 78?
I recall reading somewhere that while that was possible, joining together several partial or whole phonograph-recorded takes and transferring them to a master disc to make up a performance that seemed continuous but had not been continuously performed was not only tricky but also led to a rather noisy final product.
Link to comment
Share on other sites

Our friend Terry Teachout in his blog today writes about Golden Era piano recordings:

and says of Benno Moiseiwitsch’s 1939 recording of Sergei Rachmaninoff’s transcription of the scherzo from Mendelsson’s incidental music for A Midsummer Night’s Dream: "This performance was made in a single take, as though this were remarkable.
Please correct me if I don't know what I'm talking about, but In the pre-tape era were not virtually all recordings of pieces that would fit onto one side of a 78 made in single takes, with each take of a longer work coinciding in length with the amount of music that would fit onto one side of a 78? If so, how common would it have been for two or more takes of a performance to be joined together to make up what was heard on one side of a 78?
I recall reading somewhere that while that was possible, joining together several partial or whole phonograph-recorded takes and transferring them to a master disc to make up a performance that seemed continuous but had not been continuously performed was not only tricky but also led to a rather noisy final product.

That's my understanding as well. I don't see why a 4:11 performance should be considered remarkable.

I don't think it is about editing but about consecutive recordings of the same composition before a satisfactory take was achieved.

Link to comment
Share on other sites

On the other hand, it would be very unusual for a record company to record only a single take of a selection, even if the first take was musically superb. Multiple takes were generally recorded in case the first-choice master was damaged.

I don't claim to have absolute knowlege about this subject but here are my observations:

In the early years of the recording industry Victor would re-record a performance by the same artist when a master began to wear. Even when multiple takes were originaly recorded. Presumably at that time they didn't keep takes that were not issued. See Allen Suttons' Cakewalks, rags and novelties.

Later Victor recorded 4 takes of every tune one was used for issue. A second one kept, two were scrapped before test pressings were made. Those second ones were issued later in the 1930's or the LP era. See the discographical section in Don Rayno's biography of Paul Whiteman. There a more 1920's/1930's Victor alternate takes by Duke Ellington than from the 1940's.

Columbia recorded and kept many Benny Goodman takes in the 1940's. A lot are still around. See the D. Russel Connor/David Jessup Benny Goodman disco's.

Though take letters indicate offten more than one take there are very few alternate takes by Bix, Tram or Louis on Okeh.

There are a few Brunswick alternate Red Nichols takes that were issued in Canada only. See the Jazz Oracle booklet with the 3 Red Nichols 3CD sets.

I don't remember were I read it but sometimes diffrent takes were issued on East and West of the Rockies.

The best I can say after a short survey of multiple takes is that there is no general rule for the whole of the recording industry. It differs from label to label and from period to period.

Link to comment
Share on other sites

Good points all, but the implication of Teachout's remark, it seemed to me, was that it was remarkable for a pianist to need only one take to record an admittedly difficult piece like the Rachmaninoff Mendelssohn transcription. Seems to me that a very difficult piece, if one had to play it through without flaw, would be no less difficult to get through on, say, the fourth take or the tenth, perhaps even more so the more flawed versions had accumulated.
I'm reminded of the incident that ended the relationship between Arthur Rubinstein and Fritz Reiner, this by way of CSO oboist Ray Still.

'Reiner was conducting the Chicago Symphony in a recording of the Rachmaninoff Second Piano Concerto with pianist Arthur Rubenstein. We had been recording all day, and it had been a very long day. Reiner was annoyed that he was accompanying anybody, since he had a recording contract with RCA at the time, and only wanted to record the big orchestral pieces. Everything was finally finished when the first horn player, Phil Farkas, told Reiner that he had made a mistake, and he asked if we could redo that passage. Reiner said, ‘Oh, Phil, I didn’t hear it.’ But Farkas said the mistake was there and that he would not want other horn players to hear it. Then Rubenstein jumped up, too, with some things he wanted to rerecord. Rubenstein was famous for a lot of mistakes—he didn’t really start practicing until he was about sixty years old! Reiner shot back at Rubenstein with: ‘If we corrected everything you did wrong, we would be here all night!’ Reiner and Rubenstein never spoke again after that.'

Link to comment
Share on other sites

On the other hand, it would be very unusual for a record company to record only a single take of a selection, even if the first take was musically superb. Multiple takes were generally recorded in case the first-choice master was damaged.

Russell Connor, in his "bio-discography" of Benny Goodman, _BG: On the Record_, seems to indicate that in the 1930s Victor used two turntables to simultaneously record one tune (p. 140), presumably to circumvent this sort of problem. I got the feeling that BG only recorded an alternate take if he was dissatisfied with the previous one.

But them BG was a little 'fussy' on issues like this, wasn't he?

Link to comment
Share on other sites

  • 3 weeks later...

And in what Connor says is "so far as can be determined, the single most productive recording session in all of Jazz", BG cut 51 tunes for broadcast on NBC.

Most productive recording session? How would it have been measured? Number of tunes or the total length of the music?

Willis Jackson recorded four albums in one night - total length 2:24:09 - 'Jackson's action'; 'Live action'; 'Soul night live', 'Tell it' - 22 tunes, though. How long was that BG session?

MG

Link to comment
Share on other sites

What came immediately to mind when I began reading this thread were the Charlie Parker Dial and Savoy recordings. Numerous takes on many tunes.

But what I meant/asked originally was different -- how common was it in pre-tape era for two or more takes of a classical work to be stitched together to make up a 78 master? Also, per a later comment about the Terry Teachout item that inspired my question, how uncommon was it in the pre-tape era for a work that took up one side of a 78 to be satisfactorily recorded in one take?

Teachout, writing of Benno Moiseiwitsch’s 1939 recording of Sergei Rachmaninoff’s transcription of the scherzo from Mendelsson’s incidental music for A Midsummer Night’s Dream, said "This performance was made in a single take, as though that were remarkable. Yes, it's a difficult piece, but Moiseiwitsch would have been expected to play the piece very well if he were performing it at a concert, where one doesn't get second chances, so why should a good enough to release first-take performance in a recording studio be remarkable? And if that's not what Teachout meant, we're back to my first question about how common it was in the 78 era for a released 78 to made up of two or more stitched together takes.

Link to comment
Share on other sites

But what I meant/asked originally was different -- how common was it in pre-tape era for two or more takes of a classical work to be stitched together to make up a 78 master? Also, per a later comment about the Terry Teachout item that inspired my question, how uncommon was it in the pre-tape era for a work that took up one side of a 78 to be satisfactorily recorded in one take?

Teachout, writing of Benno Moiseiwitsch’s 1939 recording of Sergei Rachmaninoff’s transcription of the scherzo from Mendelsson’s incidental music for A Midsummer Night’s Dream, said "This performance was made in a single take, as though that were remarkable. Yes, it's a difficult piece, but Moiseiwitsch would have been expected to play the piece very well if he were performing it at a concert, where one doesn't get second chances, so why should a good enough to release first-take performance in a recording studio be remarkable? And if that's not what Teachout meant, we're back to my first question about how common it was in the 78 era for a released 78 to made up of two or more stitched together takes.

Soon after this thread started, I wrote a long post that touched on Larry's original question, restated above. As I was finishing the post, my internet connection went dark, so I couldn't post it, and it didn't save. I didn't have the heart to rewrite it at the time. But here it is - take two:

Combining takes or overdubbing in pre-tape days would have involved dubbing from one disc to another, which would degrade the sound quality. Dubbing of this kind did take place - for instance, until 1926 the Pathe label recorded on a paper-towel-roll sized wax master cylinder, which revolved at something like 200 RPM. They then dubbed the takes they wanted onto disc. Not only was there the usual loss of quality from dubbing, but the motor that turned the cylinder was noisy, so rumbles from that were transferred along with the music. To quote 78 expert Tim Grayck, "Much fidelity was lost in the dubbing process, which is why no other record company considered the process, even for a minute." However, many early jazz reissues in the 1930s were dubbed from discs if the masters were lost or unavailable.

The most famous example of overdubbing in the days of disc recording is probably the Sidney Bechet "one-man-band" record. Bechet said that he got the idea after his manager, John Reid, witnessed an orchestral recording session during which the oboe player missed an entrance. The engineer then dubbed the just-recorded orchestra and the oboe player, playing "live," onto a fresh disc. So such things did apparently happen in the world of classical recording, but I can't imagine this technique was used very often, especially to "marry" parts into a whole. It would have been pretty tricky to get right.

And back to Bechet's record - it's a good example of how dubbing from one disc to another degrades the sound. For "The Sheik of Araby," Bechet first recorded the tenor sax, then overdubbed bass, drums, piano, soprano sax, and clarinet. (It seems counter-intuitive that he started with tenor, but it's a very rhythmic, in-the-pocket part.) By the time the final pass was finished the tenor sax sounded like a bassoon played in an adjoining room, and the bass, drums, and piano didn't sound much better. "Blues of Bechet," on the other hand, only has two overdubs. He put down a piano part, overdubbed a part on which he played clarinet in the first half and soprano sax in the second half, then dubbed tenor over that. The tenor is the final pass, and it sounds good - like a real tenor sax.

Lots of digression here, but the bottom line is that I've never read that dubbing partial performances together onto a final master was a common technique in 78 recording, and I seriously doubt that it was used more than occasionally, if ever.

Edited by jeffcrom
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...