Real bass trombones

[harrisonreed]

Guess I'll have to record some notes and post some spectograms if you don't believe this.

Don't tease us now! I bet you won't!

[robcat2075]

That is not correct.

You can map out the harmonic series of any note you are playing by playing the notes that are available in that position.

So, the middle A being referenced in second position has a harmonic series of A, starting 2octaves above the pedal A fundamental, of A (root), C#, E, A, etc.
. That same pitch played in 6th position has the harmonic structure of F available in 6th position. A (root) C, Eb, F, etc. ie the F harmonic structure

I don't believe this to be true and I've never heard anyone in physics assert such a thing. Consider that the first harmonic over any fundamental tone is an octave above it, the next is an an octave plus a fifth, the next is two octaves...

The existence of a playable note a mere third above the played one is not a factor in this in the overtone series produced by the played note.

The harmonics of any performed pitch are so far above the reliable "slotting" of the tubing that I am very doubtful that the resonances of the tube bend the harmonics as hornbuilder is describing.

Possibly... they have some small role in causing some natural harmonics to predominate and others to be dampened?

A certain length of tube can only produce one set of harmonics. If that wasn't the case, we could play any note in any position. That is why we can play alternate positions. The "alternate" note has to be a note available on the harmonic series of that particular slide position.

The conventional statement is that a certain length of tube can produce only one base pitch (sometimes called a "fundamental") and several overblown notes (sometimes called "partials"). In classic physics these overblown notes are always whole number multiple frequencies of the base note. (Real musical instruments are contrived to approximate this as closely as possible.)

This is different from the harmonics that a played tone contains.

The harmonics produced by by a classic resonator, string or air column, will always be whole number multiples of the fundamental frequency.

[Kbiggs]

Back to the OP’s question: Several people have answered this very well, but I’ll add something something to the mix. The modern bass trombone is a compromise arrived at through experimentation (history) and efficiency (the desire to play things more easily and allow more people to play the instrument). It is designed to work as a bass trombone. Being pitched in Bb, the slide has 7 positions, so it can produce a chromatic scale and it’s about an arm’s length. It’s easier to play than a bass trombone with a handle, whether that instrument is pitched in F, Eb, D, or any other pitch. The large bore helps to produce a “darker” sound, a compromised “mimic” of the sound of a bass in a longer length.

The more important thing, however, is the role and function of the part and player. The person playing the part must assume play the instrument in the role and function served by the music regardless of the instrument. An extreme example to prove the point: If you can play a bass trombone part on a tenor sarrusophone and still produce a characteristic sound, articulation, phrasing, blend, etc., then what would it matter? It’s the sound, the function (in the chord), and the role (what part does it play in the group) that’s important.

Several people have mentioned playing different pieces on smaller-bore trombones. I’ve played bass trombone parts on a large-bore tenor when the first played alto and the second played a 36 or a 78H. (I would have preferred an F bass, but I don’t have the money or availability for one.) If it works—if it makes the sound, and serves the role and function—why would it matter what instrument is played?

Yes, the different lengths produce different colors of sound, and in the hands of a knowledgeable and accomplished player, an F or Eb or G bass will sound “better.” But for the rest of us mortals, we’ll have to make do with a Bb bass.

[hornbuilder]

Robert.
Your assumption is incorrect.

We can only play notes that are available to us on any particular length of tube. The harmonic frequencies and the alignment of those frequencies of that tube do not change.

So. When we play a pedal Bb in 1st position, sound we hear is the fundamental pitch of Bb, with the harmonic structure available on that length of tube. ie Bb one octave above the fundamental,then F, then Bb, D, Ab, Bb, C etc..

Playing the low Bb, in first position, resonates the same harmonic series as the pedal Bb, but minus the fundamental. So the harmonic partials are the same F, Bb, D, Ab, Bb, C etc.

Playing first position F gives the harmonic series above it of Bb, D, Ab, Bb etc.

[harrisonreed]

I thought this was common knowledge. This is why your Bb in 5th will never sound the same as Bb in first. They have different harmonic series forming their tone, on different lengths of pipe.

[brassmedic]

Guess I'll have to record some notes and post some spectograms if you don't believe this.

Don't tease us now! I bet you won't!

Don't even need to. This is very easily done. I recorded the opening of Tuba Mirum from YouTube. I isolated the second note, F and plotted the spectrum. The peaks are 351 hz - F4, 529 hz - C5, 703 hz - F5, 877 hz - A5, 1052 hz - C6, etc. The overtones are of the pitch that we hear, F, not from the fundamental of that slide position, which would be Bb. Which is what I would expect. If it didn't have the overtones of the note F, I don't think we would even hear it as an F. In fact, it wouldn't be an F.

[2bobone]

NO ! Not the dreaded charts ! This thread is heating up too much and should be shut down !

[brassmedic]

NO ! Not the dreaded charts ! This thread is heating up too much and should be shut down !

Get out the popcorn!

[ithinknot]

The overtones are of the pitch that we hear, F, not from the fundamental of that slide position, which would be Bb. Which is what I would expect. If it didn't have the overtones of the note F, I don't think we would even hear it as an F. In fact, it wouldn't be an F.

Agreed. That's also why the brain readily and often infers fundamental pitch from the harmonic content of sounds where the actual 1st partial is absent or weak (either 'actually' weak, or because the 1st partial lies towards the bottom of the average hearing range for sine waves).

For 'same note played in different positions' purposes, the different tube length must modify standing wave behavior, so the relative strengths presented in the standard 'within note' harmonic structure will be somewhat different - hence the somewhat different sound. (And yes, the conical/cylindrical issue of slide extension must have some effect on perceived tone, but that's probably where the modelling gets complicated.)

WRT Matthew's contention, playing a given note on a given partial is not merely extracting a sine wave component from the 'complete spectrum' of the pedal note...

[brassmedic]

The overtones are of the pitch that we hear, F, not from the fundamental of that slide position, which would be Bb. Which is what I would expect. If it didn't have the overtones of the note F, I don't think we would even hear it as an F. In fact, it wouldn't be an F.

Agreed. That's also why the brain readily and often infers fundamental pitch from the harmonic content of sounds where the actual 1st partial is absent or weak (either 'actually' weak, or because the 1st partial lies towards the bottom of the average hearing range for sine waves).

For 'same note played in different positions' purposes, the different tube length must modify standing wave behavior, so the relative strengths presented in the standard 'within note' harmonic structure will be somewhat different - hence the somewhat different sound. (And yes, the conical/cylindrical issue of slide extension must have some effect on perceived tone, but that's probably where the modelling gets complicated.)

WRT Matthew's contention, playing a given note on a given partial is not merely extracting a sine wave component from the 'complete spectrum' of the pedal note...

Absolutely. I would expect to see the same overtones, but different ones emphasized. That's what timbre is, and I think its patently obvious that the same note played in a different position has a different timbre.

[Chatname]

First , I don’t play bass trombone so any experimentation would have to be done by other players. Otherwise, I would certainly already be doing it.
Second, I regret using the phrase “real bass trombone”. It was meant to be tongue in cheek but was unserious on my behalf, and probably provocative. I should not be so caught up with the words and terms. I apologize.
Third, on the first chair in the symphony orchestras I know of and also in mine, three or four instruments are used regularly: a large tenor (with F attachment), a small and maybe even a medium tenor (possibly straight) and an Eb/alto. On the third chair one is expected to play a large double valved Bb bass, a little smaller single valved Bb bass and at least in opera or large orchestras a contra bass in F. My thought was that in addition to this, a small F trombone would be very interesting to experiment with. Many instruments to handle, obviously. Maybe too many. However, if one learns how to play the F contra, a smaller bore F bass should not be impossible to add to the mixture and might keep life in the orchestra more interesting. Might add some exciting colors and expressions to the section playing as well, like the alto.
Fourth, the physics! Very interesting discussion by you people who know a lot more than me! I got the popcorn out and am learning a lot!
Thank you

[LeTromboniste]

Agreed. That's also why the brain readily and often infers fundamental pitch from the harmonic content of sounds where the actual 1st partial is absent or weak (either 'actually' weak, or because the 1st partial lies towards the bottom of the average hearing range for sine waves).

For 'same note played in different positions' purposes, the different tube length must modify standing wave behavior, so the relative strengths presented in the standard 'within note' harmonic structure will be somewhat different - hence the somewhat different sound. (And yes, the conical/cylindrical issue of slide extension must have some effect on perceived tone, but that's probably where the modelling gets complicated.)

WRT Matthew's contention, playing a given note on a given partial is not merely extracting a sine wave component from the 'complete spectrum' of the pedal note...

Absolutely. I would expect to see the same overtones, but different ones emphasized. That's what timbre is, and I think its patently obvious that the same note played in a different position has a different timbre.

Yes. If playing an F in 1st position significantly resonated the system at Bb and D, then we'd perceive the note as a Bb below that. That's how they fake 32' and 64' stops on organs, and also a lot of what happens when we play pedal tones (which are not a mode of resonance of the system, and have a very weak fundamental, but the brains hears them because the right pattern of overtones is there)

The instrument is a resonator, and it essentially filters the input frequencies and resonates only at or near its modes of resonance, which in first position is roughly a Bb harmonic series.

These are incredibly complex phenomena that are still not fully understood, and I'm not an expert, so I might be very wrong on parts of this, but as far as I understand, when you play an F in 1st, the system will not resonate on the overtones of F that aren't also modes of resonance for that particular length of trombone tubing. However, it will also not resonate at modes that are not being excited by the input.
And while the instrument resonates only those Bb harmonic series frequencies, the same is not true of our lips or vocal tract. When we play an F, we are not just generating random noise and letting the instrument filter it. We are producing an F harmonic series (maybe a bit rough, with some amount of noise). That the length of tubing is part of what makes our lips want to vibrate at that pitch in the first place doesn't change the fact that our body is not a fixed resonator like the instrument is, and doesn't have a preference for Bb.
So, the input is an F harmonic series, the instrument acts as a resonator for the all the F's and some of the higher C's within our note (i.e. the frequencies that are both harmonics of F and near modes of resonance of the pipe), and as a megaphone for the rest of the harmonic spectrum that we produce but that doesn't line up with the instrument's modes. So you get an F harmonic series with some but not all harmonics reinforced by the resonator. If you play that F in 6th position, all of the harmonics correspond to modes of resonance, and so the harmonics are reinforced in a different way. Of course then you have to deal with a different resistance and different problems with the altered conicity (which probably introduces a fair bit of extra inharmonicity too).

[hyperbolica]

Back on the original question, I've also often wondered about the singular nomenclature applied or possibly misapplied to the "bass trombone". I keep searching for a better playing alternative, and I keep coming back to a valved F instrument. The slide kind of complicates or even hinders the bass voice, in my view. As long as there is some sort of effective tuning mechanism, a small cimbasso / bass trombonium 4/5 valve thing in F would be better for real bass parts. Orchestral 3rd parts are often just on-the-staff 3rd parts, and usually don't require even a second valve. Real bass parts intrude on the tuba range, of course with a brighter sound.

Current 2valve bass trombones are problematic on a number of fronts, and it wouldn't be much of a chore to improve on it. Ergonomics, blatty sound, playability, confusion amongst composers about its role. Did I mention ergonomics?

[Digidog]

I don't give a rat's a$$ if a bass instrument is deemed real or not; I care for its use and its sonority. If an instrument is used for a bass function, it is a bass to me; if it sounds like a bass in the setting it's used, then it is a bass to me.

That said, I think that modern trombones are getting too large, and that orchestras and bands use too large trombones; too large bores, too wide bells. My preferred trombone section sound, regardless of setting, is a sonor vocal sound, with a distinct underlying hint of the ability to become piercing and brutally explosive without previous warning. A group of instruments that sound sweet and nice, but hints at totally different underlying characteristics.

Leaving the diffused attacks, the round, mellow and saturated brass sounds, to the horns and the tubas.

[tbonesullivan]

Chatname, I believe you are ignoring the reason we went to the current style of bass trombone.

Instruments in G, F, or Eb have VERY long slides. You can't reach all the positions with just an arm -- you need some kind of extension. Usually it's a handle. The handle is cumbersome to manipulate. When it became possible to just add the additional tube length using a "quartventil" the new technology effectively replaced the old nearly overnight. You still use the wide bore for response in the lower range, but you get effectively 10 positions between the Bb and F sides of the instrument.

I believe back then even the these bass trombones had smaller bores than a "modern" bass trombone. I know the "classic" bass or contrabass trombone is pictured with the handle on the slide, but that really was mooted with the introduction with the F attachment. Combined with using larger bells and bores, the space that the original bass trombones had in the orchestra was filled.

Once the second valve was added, that was pretty much the end.

I know that there are modern F / G bass and contrabass trombones, but those also tend to be larger in bore than those used "back in the day", mainly because the sound desired by orchestras and conductors changed. You can see a similar change with Tubas and Bass Tubas. Bores, Bells, and instruments got bigger as a more "organ-like" sound was desired. It's especially noticeable when you compare F and Eb tubas from the late 19th / early 20th century to what they became by the mid 20th century.

[BGuttman]

I think Chatname was looking for a large bore Eb or F trombone; the length of the old instruments but a modern bore size.

I stand by my feeling that it is obsolete. If you want the sound of an F or Eb bass trombone, play the Bb/[whatever] one with the valve depressed. I have managed to play my Bb/F trombone on the F side out to about 5th position (Db/Ab/Db/...). I've found that playing the notes on the F side don't really make enough of a difference to matter; at least in the ensembles where I play. I can use these trigger notes to make difficult passages within and below the bass staff play better, but sound-wise they don't really provide an advantage. In fact, trying to play this extended tubing sometimes makes the notes more difficult to play with a resulting awful sound.

A large bore instrument favors the bass register, regardless of its length.

[StefanHaller]

Here's the spectogram that I promised earlier. Not because I think it's still needed to prove my point (after a few other forum members have chimed in), but because Harrison challenged me to it.

In this recording I played three A's (top of the base staff), the first one in 2nd position, the second in 6th position, and the third in trigger-5 (or #6 or whatever you want to call that). As you can see, the harmonic spectrum is the same for all of them, and on the keyboard on the left you can easily see that the harmonics are A, A, E, A, C# and so on.

The forth note I played was a low F in 6th position, just to show that its harmonic series does not line up with the A in 6th position, as was claimed before. (I now realize that I should have played a pedal F to make that even clearer; sorry.)

Screenshot 2022-07-04 at 20.48.27.png

[StefanHaller]

I thought this was common knowledge. This is why your Bb in 5th will never sound the same as Bb in first. They have different harmonic series forming their tone, on different lengths of pipe.

So this just shows how wide-spread this misconception is, then.

[boneagain]

Stefan,
Looks like pretty good control to get the A's as close in volume as you did.
Does your tool have a simple way to normalize the volume, so the side-by-sides are more directly comparable?
These three A's always sound different to my ear. Since they are, for me, qualitatively different, I wonder if your tool can help wtih ANY quantitative differences?
Regardless of whether timbre changes come from the ratios of tubing diameter to length to frequency, or materials, or acoustic path discontinuities, our ears DO pick them up. Kinda hard to get good numbers to back that up, though.

Also, I wonder how much contribution goes in here from the "every other partial" nature of a pipe closed at one end? The way the bell and mouthpiece cup squash the partials so we get the "natural" series we expect from string based theory is pretty wild. The squashing does NOT have to match the natural series very well to work. But would one not expect that having a shape that DOES match the natural series well would mean better reinforcement for the harmonics that line up with one of the resonant frequencies of the horn? Can't say that I could see that very clearly in the spectogram, yet I still wonder.

Thanks for posting it!!

[boneagain]

Hmmm... I'm looking at Brad's post again. I realized I know how to get that data in Audacity, and export it.
I wonder if I could normalize the volume with Audacity, then overlay the spectrum data in one plot, so the differences would be clearer side-by-side. Maybe even python, so I could zoom in on the kinkiest areas... just what I need... another project... must... resist...

[tbonesullivan]

Now stuff like this always interests me, like how when you play a Pedal note on a trombone, the actual note itself isn't really ringing, it's just the overtones.

[harrisonreed]

I thought this was common knowledge. This is why your Bb in 5th will never sound the same as Bb in first. They have different harmonic series forming their tone, on different lengths of pipe.

So this just shows how wide-spread this misconception is, then.

Yeah, let me stand corrected. Your data and the previous spectograph data is even more evidence showing just how useless a "real bass trombone" would be vs. a modern one in most playing situations. Not saying there isn't a use, but overtone series dictating tone as one of them is disproven. I'm glad you did it, the image is clear as day. It looks like, if anything, the upper overtones are stronger farther out on the slide. But like notes are all made out the same parts.

[brassmedic]

So this just shows how wide-spread this misconception is, then.

Yeah, let me stand corrected. Your data and the previous spectograph data is even more evidence showing just how useless a "real bass trombone" would be vs. a modern one in most playing situations. Not saying there isn't a use, but overtone series dictating tone as one of them is disproven. I'm glad you did it, the image is clear as day. It looks like, if anything, the upper overtones are stronger farther out on the slide. But like notes are all made out the same parts.

Huh?

[harrisonreed]

It looks like the notes are all the same, regardless of position, aka horn tuning

[LeTromboniste]

It looks like the notes are all the same, regardless of position, aka horn tuning

Sure, then let's all play euphoniums or bass trumpets. Those will have similar if not identical graphs. Even a cello or bari sax might have a similar looking graph... The purpose was to show that no, the harmonic series doesn't suddenly include other notes because you play in a different position. This graph is way too low resolution and doesn't present data to show anything beyond the fact that an F is an F, and has an F harmonic series. Timbre is way, way more complex (and still far from well understood even by acousticians) than just the rough harmonic spectrum (which also doesn't account for playing characteristics). The sustain of most instrument's tone is so close that when isolated from the attack and decay, one often can't correctly guess what instrument produced it.


I actually play long basses for a living. No, an instrument pitched a fourth lower doesn't sound or play the same as when playing in 6th position, or engaging the valve. It just doesn't.

I've read you writing about the Edwards harmonic bridge and strongly supporting the fact that inserting a mere screw in there does changes your instrument. Are you really going to say that you believe an instrument that is actually built completely different, with different proportions, different balance, different conicity, etc, would somehow sound the same?

[Digidog]

I actually play long basses for a living. No, an instrument pitched a fourth lower doesn't sound or play the same as when playing in 6th position, or engaging the valve. It just doesn't.

I've read you writing about the Edwards harmonic bridge and strongly supporting the fact that inserting a mere screw in there does changes your instrument. Are you really going to say that you believe an instrument that is actually built completely different, with different proportions, different balance, different conicity, etc, would somehow sound the same?

Most people here seem to miss the fact that frequencies can be the same, but the sine waves can look different - this producing the same note with a different sound.

Instrument builders know this, and that's why seemingly similar instruments - like fluegelhorns and trumpets - sound very different when played with the proper fittings.

[harrisonreed]

Oh ok. I'm back to my original belief then. I knew D in 4th sounded different.

[imsevimse]

It looks like the notes are all the same, regardless of position, aka horn tuning

Sure, then let's all play euphoniums or bass trumpets. Those will have similar if not identical graphs. Even a cello or bari sax might have a similar looking graph... The purpose was to show that no, the harmonic series doesn't suddenly include other notes because you play in a different position. This graph is way too low resolution and doesn't present data to show anything beyond the fact that an F is an F, and has an F harmonic series. Timbre is way, way more complex (and still far from well understood even by acousticians) than just the rough harmonic spectrum (which also doesn't account for playing characteristics). The sustain of most instrument's tone is so close that when isolated from the attack and decay, one often can't correctly guess what instrument produced it.


I actually play long basses for a living. No, an instrument pitched a fourth lower doesn't sound or play the same as when playing in 6th position, or engaging the valve. It just doesn't.

I've read you writing about the Edwards harmonic bridge and strongly supporting the fact that inserting a mere screw in there does changes your instrument. Are you really going to say that you believe an instrument that is actually built completely different, with different proportions, different balance, different conicity, etc, would somehow sound the same?

Good post!

I think we do our best to make an egal sound on the instrument we play. That is an important part of learning the instrument to make an egal sound in all registers. I know it is not possible because the instrument has different timbre in high and low register no matter what but it is still a goal to connect registers sound wise. I think best is just to listen and play.

/Tom

[brassmedic]

You can definitely see timbre in the data when you plot a spectrogram. Here's the first note (D) of the Bach Sarabande played by James Markey (on top) and then by Yo Yo Ma.

sarabande2.jpg

The difference is actually rather striking. To me it looks like the trombone sound has more "core". But this is a crude representation, so we don't know how much is due to the instrument and how much is due to the sheer volume of the note, the acoustics of the room, and how they were recorded. To be really scientific about this, you would have a lot of variables to control.

[2bobone]

NO ! Not the dreaded Spectrograms ! This thread is heating up too much and should be shut down !

[boneagain]

NO ! Not the dreaded Spectrograms ! This thread is heating up too much and should be shut down !

YES INDEED! SPECTROGRAMS!

They are not just for breakfast, er, single notes. They work well for any waveform, such as a musical excerpt.
This is by a player who, to my ears, is VERY consistent in his playing. He played the same excerpt on his Bach 36, then on his Bach 42.

The "full" attachment is for spectral values across the whole range sampled from the WAV file.
The "upper" attachment is for the final octaves.

As in Brad's example, you can see definite differences in how the same basic spectral "steps" combine for different timbres.

And now I'm curious enough to try a single note or two in as many positions as I can find to go with the previous single note comparisons. I find it easier to visualize when they are on one chart like this rather than in two.

Which means... even MORE spectrograms, Bob!

Dave

PS: for the technically minded: I switched from the traditional "frequency in Hz" on the x-axis to "octaves over lowest note" because many folks relate more easily to octaves.

PPS: On these it is worth keeping in mind that the visually greater deviations in the upper octaves are subject to the psycho-acoustic audio curve that require more power at extreme frequencies to be audible.

PPPS: sorry if I got duplicates in the html. Struggled mightily but lost to the attachment demons!

36v42_full.jpg

36v42_upper.jpg

[BGuttman]

We seem to be straying somewhat from the original question, which was why when other instrument families go from soprano to alto to tenor to bass to contrabass they get longer while some trombone models do not. In fact, if we were to accept the original poster's plaint what we call a contra in F would be a "true" bass, and something like the Mirafone would be a "true" contrabass. It is a definition that goes back to Praetorius, who illustrated a whole family of trombones of increasing lengths. The problem is that as we descend to the larger instruments they become unwieldy to manipulate. Also, we have the examples of saxhorns and other valved brass from the mid 19th century where the tenor, baritone, and bass instruments were all the same length but different bores.

It might be interesting to see the spectrograms of, for example, F3 in 1st, F3 in 6th, and F3 in trigger 5.

[LeTromboniste]

We seem to be straying somewhat from the original question, which was why when other instrument families go from soprano to alto to tenor to bass to contrabass they get longer while some trombone models do not. In fact, if we were to accept the original poster's plaint what we call a contra in F would be a "true" bass, and something like the Mirafone would be a "true" contrabass. It is a definition that goes back to Praetorius, who illustrated a whole family of trombones of increasing lengths. The problem is that as we descend to the larger instruments they become unwieldy to manipulate. Also, we have the examples of saxhorns and other valved brass from the mid 19th century where the tenor, baritone, and bass instruments were all the same length but different bores.

It might be interesting to see the spectrograms of, for example, F3 in 1st, F3 in 6th, and F3 in trigger 5.

And Praetorius doesn't name a "bass trombone", but rather uses the term "Quartposaune" for the long bass in D.

Nomenclature has never, ever been consistent. A bass trumpet is not a bass instrument. A cello is, literally, a "small large viol". You can't fit a nearly 600 year-old instrument into neat, consistent, late-19th-century boxes (especially when instrument families actually invented in that time often can't fit the boxes themselves!). We just need to accept that and move on.

[brassmedic]

All the brass families got shorter with the advent of valves, except the cornet, which originally had finger holes to allow it to play chromatically without extra length. Some will probably say, "but the French Horn is in F". It's really no different than modern trombone, though. Higher notes are played on the Bb side and lower notes are played on the F side. Modern bass trombones are mostly in the key of D with both valves engaged, the same as the old quint bass with the tuning slide pulled. The difference now is that instruments can be shortened on the fly, so higher notes don't have to be played on the extreme upper partials of the instrument. The tuba is the only one that has actually evolved into a longer instrument, but tuba didn't exist before valves. Brass instruments have a different quality of sound when played on higher partials, which is why we have ensembles that recreate period instruments to get that more interesting timbre. But why would we want to do this only for the bass trombone if we aren't doing it for all the other instruments in the ensemble?

[WilliamLang]

it's an interesting assumption to say that microphones will always capture 100% of the audio information being fed into them. a spectral analysis is also dependent on the information being captured, and microphones are imperfect receivers, even at the higher end of the available spectrum.

as far as having preferences for other instruments, go ahead and bring them to gigs and auditions. if you can win over people, then that's great

[elmsandr]

YES INDEED! SPECTROGRAMS!

They are not just for breakfast, er, single notes. They work well for any waveform, such as a musical excerpt.
This is by a player who, to my ears, is VERY consistent in his playing. He played the same excerpt on his Bach 36, then on his Bach 42.
....

This is one thing that always gets me on these types of studies, to your ears they are consistent. However, with this tool (the gage being the microphone and analysis), we can actually quantify exactly how repeatable they play as measured by the graph. Do we know that the difference in the 36 and 42 graphs shown is greater or lesser than the random variation of just playing the same thing over again on the 42? I don't know that and if the player was me I would bet against it. Quantifying any difference here and the scale of such a difference on average is fascinating to me. Actually define how much is different horn to horn, person to person, performance to performance. Heck, right now we do not even exactly know if the chart would be the same if the microphone were listening to a recording. I think it is fair to assume that it would, but there will be variability even in that scenario if we look close enough.

That also leads us to:

it's an interesting assumption to say that microphones will always capture 100% of the audio information being fed into them. a spectral analysis is also dependent on the information being captured, and microphones are imperfect receivers, even at the higher end of the available spectrum.
...

Right, I would be VERY careful interpreting any of this as reproduceable to a different recording setup. This type of analysis is probably best used to highlight when something is the same or different within the detection window. There are multiple things that can filter or alter the input as seen, but if those are unchanged for a particular set, the differences can be see and scaled to provide directional information. Different systems could have a different repeatability value for the same performer and instrument... this can be quantified, but this step is skipped in every study I have ever seen.

Repeatability and Reproduceability get at why we can rarely get out of our way on these studies; just because you saw a higher point on one chart does not mean anything other than it was higher this one time on that one performance. Maybe it is actually lower on average, but a higher variance on that setup? That is something that I often feel about horns, some help me be consistent, some require more out of me.

Cheers,
Andy

[boneagain]

Do we know that the difference in the 36 and 42 graphs shown is greater or lesser than the random variation of just playing the same thing over again on the 42?

Great question. And, drat, I lost the file where we did JUST that! The overlays were pretty startlingly the same. But I can't honestly ask anyone to just take my work for that

Which is why I'm going to try just the multiple notes with different positions trick. More than once. Comparing notes that the tuner and my ear says are the same with each other.

I WILL normalize volume in Audacity. As you say, the fewer variable the more comparable the results.

RE: Williams note about microphone accuracy: I know I did not assume a microphone would capture 100% of the audio information. I know for a fact that ALL the microphones I have are "colored" in one way or another. Thing is, if that coloring is applied the same to both traces, it's still a usable comparison. I think sampling rate can have as much impact on results as microphone. These were, IIRC, 44k Hz. For the comparisons I'll do on single notes I'll go over double that, so none of the interesting overtones will suffer from Nyquist limit challenges.

RE: Bruce's comment about getting back to the OP.... I did not read his first post that way. He DID make assertions about "real basses" having a certain fundamental pitch. The comments about "real" are just as valid a response as those about length-based families. And I think discussion of timbres of different bore/length combinations fits nicely into both side of that. Of course, that is only my opinion.

[brassmedic]

The player should normalize the volume, not the software. The volume changes the sound spectrum, and merely reducing the volume with the software will only compress the graph. Ideally you would be able to play a note many times, making sure they are all the same volume, and generate a graph that averages them all together. Then when you compare two graphs, you will have eliminated the random variations.

[boneagain]

The player should normalize the volume, not the software. The volume changes the sound spectrum, and merely reducing the volume with the software will only compress the graph. Ideally you would be able to play a note many times, making sure they are all the same volume, and generate a graph that averages them all together. Then when you compare two graphs, you will have eliminated the random variations.

Just learned that the hard way

Having a tuning meter right there eliminates one variable.
I think I have a solid sound level meter around somewhere. I'd use the computer screen, but it gets in the way of being able to hold the horn and watch the tuner (and listen to the simultaneous drone.)

MUCH better results using unequal volume "raw" stuff than computer normalized, EXACTLY as you point out.

Expecting even better results after having the speedo and tach, er, tuner and volume.

Interestingly, the tips of the FFT spikes seem to stablize across volumes for a given position, with the valleys between those peaks going up and down with volume. Obviously, more experimenting required...

[MrHCinDE]

Where does this fit on the scale of realness:
https://www.ebay-kleinanzeigen.de/s-anz ... 9-74-2041?

[baileyman]

Re the Bartok gliss, if I recall he was poking fun at Shostakovich with that, a guy he thought was basically a fraud. It's funny how players obsess over it. Conceivably, Bartok may have written a basically unplayable line as part of the joke.

On the sound of a long bass, it seems like there are two things happening here. First is the overtone series, which on an air column are pretty exact in frequency until the very high multiples. Second is the trombone formant, which is characteristic of trombones in general and also individual horns, as a result of "errors" in the air column. These make internal reflections and "color" the overtone series as a series of weights in the series (different coefficients). For any fixed position, the formant is constant but changes with slide length. "Bach sound" seems likely to be from a characteristic formant created by slide joint diameter mismatch, tuning slide receiver diameter jumps, slide bow diameter jumps, for instance. So basically, the formant for an F bass with long slide should still say "trombone" but would also say "different" compared to a short slide bass, even when playing the same note with the same amount of tubing at work.

[BGuttman]

I like your idea. But who's going to notice the difference? The conductor? Probably not most. The audience? Probably even fewer. Only a few (and I want to emphasize few) trombone players will notice.

[Posaunus]

Re the Bartok gliss, if I recall he was poking fun at Shostakovich with that, a guy he thought was basically a fraud. It's funny how players obsess over it. Conceivably, Bartok may have written a basically unplayable line as part of the joke.

Fascinating thought. I really like the music of both Bartok and Shostakovich. I wonder what Bela thought was fraudulent about Dmitri?

[Posaunus]

On the sound of a long bass, it seems like there are two things happening here. First is the overtone series, which on an air column are pretty exact in frequency until the very high multiples. Second is the trombone formant, which is characteristic of trombones in general and also individual horns, as a result of "errors" in the air column. These make internal reflections and "color" the overtone series as a series of weights in the series (different coefficients). For any fixed position, the formant is constant but changes with slide length. "Bach sound" seems likely to be from a characteristic formant created by slide joint diameter mismatch, tuning slide receiver diameter jumps, slide bow diameter jumps, for instance.

Interesting insight. Thanks for that. I really don't know (or, I guess, appreciate) the "Bach sound" because I've never played one since I sampled several Bach 42Bs many years ago and found those trombones inferior to my wonderful Conn 88H (which I still play and love). I'm sure the Conn "formant" is rather different from the Bach's. Apparently I approve of Conn's "errors!"

[hornbuilder]

Re the "Bach Sound". That sound comes from the bell. Not the rest of the horn. Putting a 42 bell on a Conn chassis gives a horn that sounds like a Bach 42. Similarly, an 88 bell on a 42 Chassis gives a horn that sounds like a Conn 88. Observation based on having done the experiment.

[HowardW]

Re the Bartok gliss, if I recall he was poking fun at Shostakovich with that, a guy he thought was basically a fraud. It's funny how players obsess over it. Conceivably, Bartok may have written a basically unplayable line as part of the joke.

Unplayable and part of the joke? Hardly. Actually, Bartók also used the BB to F bass trombone gliss in other places. For example 22 times in The Miraculous Mandarin, in the full ballet version (one measure before rehearsal no. 97 to one before no. 101 -- the first two times are BB to F, then 20 times up-and-down F to BB to F.

I remember when the Boston Symphony played Bluebeard's Castle at the Lucerne Festival in 2001, Doug Yeo played the bass trombone part on the F-bass that he also always used for the glissando in the Concerto as well as for the Mandarin. I don't have a score of Bluebeard, so I can't say if there is also a BB to F gliss there too, but apparently the part fits well on the F-bass.

See also the so-called "glissandi" in all three trombones in the ad lib finale of the Violin Concerto. They are not actually glissandi, but the slide position numbers in the bass trombone part correspond to those of an F-bass.

I don't know why Bartók thought of the bass trombone as an instrument in F. I have a couple ideas, but I've never been able to find any tangible evidence ...

Moreover, I don't know if Bartók thought that Shostakovich was a fraud, but he surely thought that the theme Shostakovich repeated ad nauseam in the Seventh Symphony was extremely trite: hence the scornful laughs assigned to the trombones after his quotation and persiflage of it.

Howard

[sungfw]

Interesting thread on whether or not Bartók was parodying Shostakovich 7 over on the Good Music Guide Classical forum (including the observation that Shostakovich parodied Bartók's Sonata for 2 Pianos and Percussion in the 13th.)

[brassmedic]

Where does this fit on the scale of realness:
https://www.ebay-kleinanzeigen.de/s-anz ... 9-74-2041?

Pretty low on the scale. That's about the most inauthentic sackbut you could find. At least it doesn't have a valve.

[baileyman]

Re the Bartok gliss, if I recall he was poking fun at Shostakovich with that, a guy he thought was basically a fraud. It's funny how players obsess over it. Conceivably, Bartok may have written a basically unplayable line as part of the joke.

Unplayable and part of the joke? Hardly. Actually, Bartók also used the BB to F bass trombone gliss in other places. For example 22 times in The Miraculous Mandarin, in the full ballet version (one measure before rehearsal no. 97 to one before no. 101 -- the first two times are BB to F, then 20 times up-and-down F to BB to F.

I remember when the Boston Symphony played Bluebeard's Castle at the Lucerne Festival in 2001, Doug Yeo played the bass trombone part on the F-bass that he also always used for the glissando in the Concerto as well as for the Mandarin. I don't have a score of Bluebeard, so I can't say if there is also a BB to F gliss there too, but apparently the part fits well on the F-bass.

See also the so-called "glissandi" in all three trombones in the ad lib finale of the Violin Concerto. They are not actually glissandi, but the slide position numbers in the bass trombone part correspond to those of an F-bass.

I don't know why Bartók thought of the bass trombone as an instrument in F. I have a couple ideas, but I've never been able to find any tangible evidence ...

Moreover, I don't know if Bartók thought that Shostakovich was a fraud, but he surely thought that the theme Shostakovich repeated ad nauseam in the Seventh Symphony was extremely trite: hence the scornful laughs assigned to the trombones after his quotation and persiflage of it.

Howard

The gentleman yields the rest of his time on this topic to Howard.

[rickfaulknernyc]

And it isn’t necessarily volume or else I suspect big bands would be packed with 547 horns too.

I remember once reading a quote from somebody, talking about high school and college jazz bands using large-bore tenors: "sounds like a squadron of baritone horns." In a jazz context, a section of large-bore tenors would sound too thick and not have enough bite.

BTW, a section of three tenors was the norm in most big bands until at least the 50's. I love the sound of the three-tenor Duke Ellington section, especially in the 50's: tight unisons and ringing triads.