Increasing speed of records half a tone

[The Magnificent Goldberg]

The sleeve note to vol 5 of the Ace compilation of the complete Fats Domino Imperial singles contains the following addendum:

According to owner/engineer Bunny Robyn at Master Recorders in Los Angeles everything from Fats' first analogue tape session (the fourth) was sped up half a tone for release. Even the earlier lacquer recordings were sped up when issued on album. According to Robyn it was his idea, and he convinced Lew Chudd to issue the sides this way. Robyn told Rick Coleman, 'I had a special capstan made for Fats' work, which automatically moved his key up a half-tone, which gives it just a little more movement and, of course, a little edge to his voice.' He describes the original recordings as 'loagy'. The Domino sessions were taped in New Orleans at Cosimo Matassa's studio and mastered in Los Angeles at Master Recorders.

Now, mathematically this doesn't seem to me to work, because the larger capstan is going to increase the frequency by exactly the same percentage all the time. This table gives the frequencies of 8 octaves of equal tempered notes.

http://www.phy.mtu.edu/~suits/notefreqs.html

The differences between one note and the next one up or down aren't the same. They aren't even the same percentage different. And the differences don't all run in the same direction.

ie middle C is 261.63 Hz

C# is 277.18 which is 5.9435% bigger

D is 293.66 which is 5.9456% bigger

D# is 311.13 which is 5.9491% bigger

E is 329.63 which is 5.5,9461% bigger

That must mean that some notes are slightly not what they're supposed to be - they're out of tune, in other words.

So, is the inaccuracy undetectable? Or can some of you musicians tell the difference?

MG

[Scott Dolan]

I have no idea what you just said, but I will admit that I enjoyed reading it.

[The Magnificent Goldberg]

Gawd blesher, Guv!

MG

[JSngry]

Yes, it makes sense, and yes, you can tell the difference, although not necessarily automatically. Anything "artificially" raised by that amount (a half-step is a pretty significant degree, actually) is going to sound and feel different just because you're compressing real-time events into a smaller actual space.

Think about it in visual terms, like if a 30 minute block of TV programming was compressed into 25 minutes (or whatever the mathematical equivalent would be. Everything would move faster, the voices would sound higher and speak faster, everything. Same thing with increasing tape speed. It's not just the tempo and the that gets changed, it's everything.

Nowadays, with digital, I don't know what all you can do. Digital manipulation is a whole 'nother world. Digital pretty much rules out (or is in the process of ruling out) "impossible". But analog, like we're talking about here, yeah.

And it's not really an "inaccuracy". Everything still remains as accurate (or inaccurate) to everything else within/relative to itself, at least as long as you're dealing with the master. When you get into Beatles type stuff, like on "Strawberry Fields Forever" where they combined takes that were recorded in different keys and then altered the speed/pitch to combine them, then, yeah, there might well be clashes (at least from a "technical" sense. Esthetically is something else again, right?). But the Domino stuff, they didn't make the records like that, to put it mildly.

An interesting sidelight, perhaps (and a definite coincidence of New Orleans-isity). When I was a kid, I used to be able to pick up WNOE AM, a big Top 40 station out of New Orleans, after the sun went down and the airwaves opened up. That station would quite often speed their turntable up by at least a half-step, and when you lived on Top 40 24/7 like I did back in the day, hell yeah you could notice the difference. Their shit sounded FUNNY! I could only take it for so long, ya' know? I figured they did this so they could play more records and sell more commercial time, like yeah ok, we can give you five records between commercials instead of four and do it in such a way that we can sell an extra 30 or 60 second slot per 15 minute block, kinda like a win-win for them, and if nobody was paying close enough attention to tell that it sounded like their radio had just inhaled a little bit of helium, hey, no problem, right?

Equally interesting to me in terms of "studio as instrument" is how Brain Wilson put together "Good Vibrations" from sessions in three (maybe four?) different studio because each different room and each different board had it's own sound, and by god, he wanted those differences, it was a form of orchestration. And by god, yes, you CAN hear it! Again, with digital, hey, no need for all that, but I dig it when people think about stuff like that, because, no, it doesn't always matter when you don't, but it always does when you do. Kinda one of those you-didn't-have-to-do-it-like-you-did-but-you-did-yes-you-did-and-I-thank-you kind of things. I love it when that happens, especially in pop music, because sometimes that's all that pop music has to offer you, it being pop music an all. But it's enough of an enough when it does. Just anybody thinking about anything is nice, if you know what I mean.

[robertoart]

Kind Of Blue (only the stereo I think)? was infamous for being a half step sped up for years until someone in the cd era decided it should be converted to concert pitch.

[JSngry]

The lower the fidelity, the easier it is to pick up that the pitch is off. The Bird Rockland Palace stuff, you could tell right away that that was off. Lots of bootlegs are like that. Especially if you're a player and you hear a guy using some kind of false fingering on a note and you KNOW that, no, that violates all laws of instrumental physics, UH-oh. But sometimes, as with KOB, you don't really notice until you get into transcribing and then oh, REALLY?

Funny how everything's come full half circle now in terms of what is better, sped up or slowed down...

https://www.youtube.com/watch?v=ppZc_Lh37Vw

Syrup - it's not just for pancakes any more!

Edited April 6, 2013 by JSngry

[GA Russell]

Jim, were you listening to CC and Lew Kirby on WNOE?

Edited April 6, 2013 by GA Russell

[JSngry]

Don't remember any names, honestly, but it was nights from about 1968-1971.

[The Magnificent Goldberg]

Yes, it makes sense, and yes, you can tell the difference, although not necessarily automatically. Anything "artificially" raised by that amount (a half-step is a pretty significant degree, actually) is going to sound and feel different just because you're compressing real-time events into a smaller actual space.

Think about it in visual terms, like if a 30 minute block of TV programming was compressed into 25 minutes (or whatever the mathematical equivalent would be. Everything would move faster, the voices would sound higher and speak faster, everything. Same thing with increasing tape speed. It's not just the tempo and the that gets changed, it's everything.

Nowadays, with digital, I don't know what all you can do. Digital manipulation is a whole 'nother world. Digital pretty much rules out (or is in the process of ruling out) "impossible". But analog, like we're talking about here, yeah.

And it's not really an "inaccuracy". Everything still remains as accurate (or inaccurate) to everything else within/relative to itself, at least as long as you're dealing with the master. When you get into Beatles type stuff, like on "Strawberry Fields Forever" where they combined takes that were recorded in different keys and then altered the speed/pitch to combine them, then, yeah, there might well be clashes (at least from a "technical" sense. Esthetically is something else again, right?). But the Domino stuff, they didn't make the records like that, to put it mildly.

An interesting sidelight, perhaps (and a definite coincidence of New Orleans-isity). When I was a kid, I used to be able to pick up WNOE AM, a big Top 40 station out of New Orleans, after the sun went down and the airwaves opened up. That station would quite often speed their turntable up by at least a half-step, and when you lived on Top 40 24/7 like I did back in the day, hell yeah you could notice the difference. Their shit sounded FUNNY! I could only take it for so long, ya' know? I figured they did this so they could play more records and sell more commercial time, like yeah ok, we can give you five records between commercials instead of four and do it in such a way that we can sell an extra 30 or 60 second slot per 15 minute block, kinda like a win-win for them, and if nobody was paying close enough attention to tell that it sounded like their radio had just inhaled a little bit of helium, hey, no problem, right?

Equally interesting to me in terms of "studio as instrument" is how Brain Wilson put together "Good Vibrations" from sessions in three (maybe four?) different studio because each different room and each different board had it's own sound, and by god, he wanted those differences, it was a form of orchestration. And by god, yes, you CAN hear it! Again, with digital, hey, no need for all that, but I dig it when people think about stuff like that, because, no, it doesn't always matter when you don't, but it always does when you do. Kinda one of those you-didn't-have-to-do-it-like-you-did-but-you-did-yes-you-did-and-I-thank-you kind of things. I love it when that happens, especially in pop music, because sometimes that's all that pop music has to offer you, it being pop music an all. But it's enough of an enough when it does. Just anybody thinking about anything is nice, if you know what I mean.

That's not what I meant, Jim. Sure speeding the thing up half a tone will raise the pitch half a tone - but only approximately. If the machine raises middle C by 5.9435%, that note becomes C# exactly. But if someone's playing D, that 5.9435% invrease is going to change the D to a little bit flat of D#; and other notes are going to be a little bit sharp of the note that's half a tone up from them. But maybe these differences are so small as not to be noticeable.

MG

[JSngry]

Are you thinking in terms of natural temperament or even temperament?

String players routinely differentiate between C# & Db, G# & Ab, etc., especially when playing chamber works, Has to do with overtones relative to key center and/or some such. Nest time you see a string quartet on a smoke break, stop by, say hello, and ask them about that. You'll be surprised how chatty the can get when they start talking shop!

But if you're asking if it's possible to, by altering the speed of a master tape, to in some way find a % where the top-most & bottom-most pitches end up in different or almost-different keys relative to each other, I can't say that I've eve heard of such a thing, and I know about a lot of people playing a lot of games with tape speeds. But that doesn't necessarily mean that it couldn't happen. But I really don't think it works that way.

Bottom line - if you can find a pop record where everything is perfectly in tune to even temperament for every note played and sung, you've found a true rarity! So you're essentially starting with flawed data anyway, to be honest.

[robertoart]

Re-the nuisance that was transcribing or playing along with KOB at up a half step...

Couldn't all you Saxophone players just wiggle your mouthpieces a bit and be in Aflat

[JSngry]

Coming from the player of an instrument where changing keys is just a matter of starting from a different fret, that's funny!

The more I think about MG's query, the more interesting it becomes, because he's really dealing with the acoustics of microtonality. If 7/4 is reading this, he could no doubt get to the gist much better than me, because I'm not sure but that MG doesn't have a mathematical point.

This would be an interesting experiment, perhaps - record a string/horn/whatever section playing, say a C# Maj 7 chord, sustaining it for, say, 10-15 seconds with no vibrato (because vibrato changes everything). Then have the same section record a C Maj 7 chord for the same duration, again without vibrato. Then digitally bring the C Maj & up to a C# Maj 7, but keep the duration at 15 seconds (easily done with digital). Feed both tracks into a single channel, and use a digital mixer to feed both into a single track, gradually transitioning from the natural C# chord to the sped up C# chord in a mathematically perfectly symmetrical 15 second arc of phasing one in and the other out. At the 7.5 second mark, you would have equal parts each, and double density, but you'd also have equal parts of whatever micro-tonal adjustments the math would have made to the "trueness" of the pitch. By the end of it, you should be able to tell how much has really changed to you ear, and then figure out if it was a timbral thing, a micro-pitch thing, or if these two are really the same thing. Or even better, run both into two separate tracks and A/B them at random intervals. You would likely notice a difference, but what the nature of that difference would be, hmmmm....

You could do the same thing with analog, just make two loops and use a stopwatch. But that's old-fashioned.

[Rooster_Ties]

Kind Of Blue (only the stereo I think)? was infamous for being a half step sped up for years until someone in the cd era decided it should be converted to concert pitch.

Don't think it was just the stereo -- and (iirc), it wasn't a full half-step, but rather something closer to a quarter-step (or something in the cracks). Best as I recall.

[Teasing the Korean]

What key are the tunes in when you spin them? If they are in sharp keys, they are probably sped up a half tone. (Pianists tend to favor flat keys.) If they are in flat keys, they are probably correct.

[The Magnificent Goldberg]

What key are the tunes in when you spin them? If they are in sharp keys, they are probably sped up a half tone. (Pianists tend to favor flat keys.) If they are in flat keys, they are probably correct.

Goodness knows! Buy a Fats Domino record and see

MG

[The Magnificent Goldberg]

Coming from the player of an instrument where changing keys is just a matter of starting from a different fret, that's funny!

The more I think about MG's query, the more interesting it becomes, because he's really dealing with the acoustics of microtonality. If 7/4 is reading this, he could no doubt get to the gist much better than me, because I'm not sure but that MG doesn't have a mathematical point.

This would be an interesting experiment, perhaps - record a string/horn/whatever section playing, say a C# Maj 7 chord, sustaining it for, say, 10-15 seconds with no vibrato (because vibrato changes everything). Then have the same section record a C Maj 7 chord for the same duration, again without vibrato. Then digitally bring the C Maj & up to a C# Maj 7, but keep the duration at 15 seconds (easily done with digital). Feed both tracks into a single channel, and use a digital mixer to feed both into a single track, gradually transitioning from the natural C# chord to the sped up C# chord in a mathematically perfectly symmetrical 15 second arc of phasing one in and the other out. At the 7.5 second mark, you would have equal parts each, and double density, but you'd also have equal parts of whatever micro-tonal adjustments the math would have made to the "trueness" of the pitch. By the end of it, you should be able to tell how much has really changed to you ear, and then figure out if it was a timbral thing, a micro-pitch thing, or if these two are really the same thing. Or even better, run both into two separate tracks and A/B them at random intervals. You would likely notice a difference, but what the nature of that difference would be, hmmmm....

You could do the same thing with analog, just make two loops and use a stopwatch. But that's old-fashioned.

Yes, that's what I'm on about. 7/4 hasn't dropped in, yet...

Anyway, it's something like what you say about Johnny Hodges; that he (and Teddy Edwards, Sidney Bechet and some others) played an infinite number of notes. I know you don't really mean infinite, because it would take an infinitely long period of time to play that, but I do understand very large indeed So, if not infinite, the notes they play between C and C# must be more like a flight of very tiny stairs than a lift (elevator). I can't hear those steps as separate steps but as a continuous smooth movement (as in a lift, which passes through all possible points). But I expect you can. Or can you? Perhaps you hear it the same way I do, which is kind of an answer, because if you also can't hear these steps, maybe you couldn't hear the difference between a frequency of 293.66 Hz (D nat above middle C) and 293.6542 (C# raised by the same % as C# is above C nat). Or if you can hear those steps, you'd be conscious of the 'oddnesses' in a Fats Domino record.

Alternatively, get a 78 of one of Fats' early recordings and an LP or CD version and play them back to back

Hey, I bet Jeffcrom has both versions of 'The fat man'!

MG

[mjzee]

Other records that were sped up for 45 release:

I Got You (I Feel Good) - James Brown

Gimme Some Lovin' - Spencer Davis (Stevie Winwood)

[The Magnificent Goldberg]

Other records that were sped up for 45 release:

I Got You (I Feel Good) - James Brown

Gimme Some Lovin' - Spencer Davis (Stevie Winwood)

Do you know if there are releases of those recordings at the correct speed?

MG

[Teasing the Korean]

I'm not sure about "I Got You," but "Papa's Got a Brand New Bag" was sped up significantly, I believe a whole tone rather than half tone.

I just played "Blueberry Hill" on youtube and it is in B natural. I would bet the farm that the band played it in Bb at the session.

Edited April 7, 2013 by Teasing the Korean

[king ubu]

The mono take of "Papa's Got a Brand New Bag" at the very end of "Star Time", the great 4CD JB compilation/best-of, seems to be running in the originally recorded speed ("Previously unreleased complete take. This version was later sped up and edited to create the two-part King single master." - the duration of the track on Star Time is given as 6:56)

"I Can't Stand Myself (When You Touch Me)" was btw. issued in slower speed than originally recorded. "Star Time" has pt. 1 (stereo, 3:29) on disc two:" Released December 1967 at a slower speed, King single. It is presented here in its original recorded tempo.")

I'm not sure about "I Got You" though. "Star Time" has the withdrawn stereo take (2:27, rec. NYC 1964-09-09) on disc one, "Intended to be released Fall 1964, as a Smash single and on Smash album OUT OF SIGHT; withdrawn due to court order". It also has "I Got You (I Feel Good)" (mono, 2:45, rec. Miami, 1965-05-06) issued as King single but there's no note about change of speed there.

Edited April 7, 2013 by king ubu

[The Magnificent Goldberg]

I'm not sure about "I Got You," but "Papa's Got a Brand New Bag" was sped up significantly, I believe a whole tone rather than half tone.

I just played "Blueberry Hill" on youtube and it is in B natural. I would bet the farm that the band played it in Bb at the session.

So, what you're saying is that, when a record's sped up (or slowed down, as someone noted), the inaccuracies that are inherent in speeding up by a constant amount are too small to be noticed.

MG

[Teasing the Korean]

I'm not sure about "I Got You," but "Papa's Got a Brand New Bag" was sped up significantly, I believe a whole tone rather than half tone.

I just played "Blueberry Hill" on youtube and it is in B natural. I would bet the farm that the band played it in Bb at the session.

So, what you're saying is that, when a record's sped up (or slowed down, as someone noted), the inaccuracies that are inherent in speeding up by a constant amount are too small to be noticed.

MG

If you're thinking in terms of western divisions of pitch, I would say raising something a half step, e.g., C to C#, does not radically change the inherent tonal qualities. It gives it a little tempo bump but not enough to be readily noticed. It can add a bit of urgency to the sound, and make the vocals brighter.

Now, I would saying lowering something by that same amount is much more noticeable. It makes the vocal sound lower/flabbier, and it reduces the tempo more noticeably. Technically, it varies the speed by the same amount as the other direction, but in terms of our perception, we hear it more. That is subjective but it has been my experience.

When you get to raising/lowering something by a whole-step, you change the tonal qualities and tempo much more noticeably.

it is interesting that the Beatles, being the great singers that they were (at least John and Paul), messed considerably with the speed of their voices between Revolver and Magical Mystery Tour. "Here There and Everywhere," "Strawberry Fields," "When I'm 64" and MMT (title song) are some of the obvious examples.

Edited April 7, 2013 by Teasing the Korean

[The Magnificent Goldberg]

Now, I would saying lowering something by that same amount is much more noticeable. It makes the vocal sound lower/flabbier, and it reduces the tempo more noticeably. Technically, it varies the speed by the same amount as the other direction, but in terms of our perception, we hear it more. That is subjective but it has been my experience.

Strange, it shouldn't be. B is 5.9465% higher than B flat. But B flat is 5.6127% lower than B. (105 is 5% more than 100, but 100 is only 4.7619 less than 105.) So you don't slow down as much as you speed up.

MG

[Teasing the Korean]

I'm not sure how you're calculating those numbers. It should be the same in either direction.

Still, it's a perception thing. We're more likely to notice that something has been slowed down. Again, this is subjective, and from my experience.

[The Magnificent Goldberg]

I'm not sure how you're calculating those numbers. It should be the same in either direction.

No - try this on a calculator: Work out what five percent of 100 is - OK, do that in your head 105. Now work out (on a calculator) what 95% of 105 is - it's NOT 100.

Still, it's a perception thing. We're more likely to notice that something has been slowed down. Again, this is subjective, and from my experience.

I believe you there. It's the audio equivalent of an optical illusion.

MG