Stars and Stripes Forever - Sousa's Band on Blue Amberol

[Henry]

Playback on the first recording is too fast: the march ends in Ab, but here it's a quarter-tone sharp (i.e., somewhere in between Ab and A natural). The second version is spot on Ab, notwithstanding the annoying speed/pitch variation, especially noticed in the trio da capo. Also, obviously, version two is considerably more distorted than version one. Of course, these differences can be accounted for by purely mechanical explanations. I'm basing my assertion on musical performance differences that I hear between the versions; they're close, but not identical. Take it from one who has played the first trombone part of this march many times in bands and orchestras, and has heard it many more times played by others. I'll cite two examples: (1) the famous piccolo solo in the last strain of the trio; both are well played, but version one has the edge for crispness, accuracy, and beauty of tone production, and (2) the descending chromatic runs in the "break" strain (between the two strains of the trio) are much cleaner in version one; in version two they're muddled, and there are wrong notes (yes, they DO happen, even in the best of groups!).

"The Stars and Stripes Forever" was Sousa's most popular march, and his band recorded it several times, both early and late in the band's history. I highly recommend Crystal Records CD461-3, "Sousa Marches Played by the Sousa Band: The Complete Commercial Recordings," for its historical discussion as well as its four different recordings of "Stars and Stripes," only one of which was actually conducted by Sousa himself. (But, alas, it doesn't include the blue amberols in this thread, so I guess it's not as "Complete" as it could/should be!)

[Valecnik]

Playback on the first recording is too fast: the march ends in Ab, but here it's a quarter-tone sharp (i.e., somewhere in between Ab and A natural). The second version is spot on Ab, notwithstanding the annoying speed/pitch variation, especially noticed in the trio da capo. Also, obviously, version two is considerably more distorted than version one. Of course, these differences can be accounted for by purely mechanical explanations. I'm basing my assertion on musical performance differences that I hear between the versions; they're close, but not identical. Take it from one who has played the first trombone part of this march many times in bands and orchestras, and has heard it many more times played by others. I'll cite two examples: (1) the famous piccolo solo in the last strain of the trio; both are well played, but version one has the edge for crispness, accuracy, and beauty of tone production, and (2) the descending chromatic runs in the "break" strain (between the two strains of the trio) are much cleaner in version one; in version two they're muddled, and there are wrong notes (yes, they DO happen, even in the best of groups!).

Henry, sounds like you definitely know you stuff, certainly you have a better ear than I do. Regarding it being too fast, the first version, (on the Amberola 1B) is played at 160rpm. I checked it with a new electronic stroboscope I just got. Do you think it was recorded at a different speed or?

[Henry]

Henry, sounds like you definitely know you stuff, certainly you have a better ear than I do. Regarding it being too fast, the first version, (on the Amberola 1B) is played at 160rpm. I checked it with a new electronic stroboscope I just got. Do you think it was recorded at a different speed or?

Thanks for the compliment, Valecnik. I like to give the impression that I know my stuff, even when I sometimes don't! (I couldn't have survived 29 years of college teaching otherwise

As for the sharp pitch in version 1, it could either be due to a too fast playback or a too-sharp tuning of the ensemble. The latter I consider unlikely, because a quarter-tone (half of a half step) sharp is a larger adjustment than most wind instruments can make without throwing off the tuning in other ways. The faster tempo of version 1 could be the result of too-fast playback, which of course would also account for the sharp pitch. It's certainly faster than I have ever heard this march performed!

P.S. [later addition]Since you've determined that the playback speed is accurate, then the recording speed could have been slow, which would account for the faster-than-standard tempo as well as the sharp pitch on playback.

[Valecnik]

Henry,

I double checked the speed and it played at around 159-160 the whole way through so that's right. I wonder if there's any way to determine the original speed at which it was recorded?

Also, the thought crossed my mind that they played it extra fast just to get the whole march on the cylinder. This is a very long recording, taking up virtually every bit of space on the cylinder.

[Henry]

Henry,

I double checked the speed and it played at around 159-160 the whole way through so that's right. I wonder if there's any way to determine the original speed at which it was recorded?

Also, the thought crossed my mind that they played it extra fast just to get the whole march on the cylinder. This is a very long recording, taking up virtually every bit of space on the cylinder.

As to your first question, I'll leave that to the real specialists/experts in cylindrical things Edison. I'm a Vic guy myself.

In re: the time-length-tempo matter, versions one and two have exactly the same material, yet the v.2 isn't taken at the lickety-split tempo of v.1. Aren't these both four-minute cylinders?

Just as an aside, in all the printed versions of this (and every other) march that I've ever played, each strain of the march is repeated directly, and the march ends at the conclusion of the trio. But in the present recordings, there are no interior repeats; the piece is played all the way through twice. The latter way makes the piece just a tad longer to perform, because the intro. is not repeated in the printed versions. Playing the whole piece twice through adds about 4 seconds to the performance when the extra time for the intro. is added in (four measures of music with two beats to the measure at 120 beats per minute, which is typical march tempo).

[Valecnik]

Now I'm more curious. I have at least one more copy of this same record no. on Blue Amberol. I'll try to find it. If it's a different take, it would be interesting to compare...

[gramophoneshane]

It'll be interesting to hear what you find. I've been wondering about the speed of my 30 ever since jnorman111 posted his video of Golden Wedding, which sounded too fast to me. Then when I heard Bruces video here, I was thinking my machine must be too slow, until Henry mentioned the note being spot on.
Now I'm not sure what to think

[Valecnik]

Well actually I found two more copies. So I have;
- Take 1 mould 1
- Take 1 mould 2
- Take 1 mould 60

So if all of these are Take 1, including Shanes,are Take 1, that Take could have only been recorded at one speed correct?

[recordo]

I'll find my copy of it on HMV EA87 - one of the best sounding early electric disks I've ever heard. Sousa's band on an electric recording really is something else!

[Henry]

Well actually I found two more copies. So I have;
- Take 1 mould 1
- Take 1 mould 2
- Take 1 mould 60

So if all of these are Take 1, including Shanes,are Take 1, that Take could have only been recorded at one speed correct?

Well, the two recording machines used for the molds in question could have been running at different speeds, one of them presumably correct and the other a tad slow. But that would not account for the performance differences I hear. If both examples posted in this thread are indeed from the same performance, then I can only attribute the perceived differences to the extremely degraded sound quality on version 2. The sound quality of version 1 is really terrific IMO!