• FAQ  • Search  • Memberlist  • Usergroups   • Register   • Profile  • Log in to check your private messages  • Log in 

Sound Experiment


Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8  Next
 
Post new topic   Reply to topic    trumpetherald.com Forum Index -> Fundamentals
View previous topic :: View next topic  

The louder you play, the less it carries!
True
38%
 38%  [ 52 ]
False
61%
 61%  [ 82 ]
Total Votes : 134

Author Message
mcamilleri
Heavyweight Member


Joined: 25 Oct 2001
Posts: 2076
Location: New Zealand

PostPosted: Fri Feb 24, 2006 2:27 pm    Post subject: Reply with quote

Dave, the lip vibration is already a square wave (or close to it). THe way the embouchure works is that as it buzzes it sends a series of puffs of air into the mouthpiece, which in terms of a pressure wave is very close to a square wave. Basically it is an open/shut valve between the high pressure in the mouth/lungs and the low air pressure in the horn. The square wave has a high harmonic content and this causes the resonance of the high harmonics.

If the lips produced a sine wave then there would be no harmonic content and the trumpet would produce a pure sine wave - fundamental with no overtones. Basically it would sound like a cheap door chime.

Many other intruments are driven by square waves - all the brass, single and double reed instruments, the violin family (the bow grips then slips with each cycle).

If you try to sythesise the sounds of these instruments you start with a square wave and then filter it. The acutal intrument is a 'filter' that modifies the sound.

Derek there are a few published studies out there on audience perception of sound. There is one on blind listening and playing tests of identical flutes with different materials (no sig. difference between plated steel and platinum!). Have a search around. If you can access acoustic/music journals at a university you should turn up some more.

Whatever you do start small with a pilot study.

Michael
_________________
Alpha Angles
Besson Loyalist
Back to top
View user's profile Send private message
_dcstep
Heavyweight Member


Joined: 05 Jul 2003
Posts: 6324
Location: Denver

PostPosted: Fri Feb 24, 2006 8:33 pm    Post subject: Reply with quote

Michael, I believe you're right, I'm merely repeating the good professor. They're saying that at lower volumes the lips are a sine wave. Maybe they really mean something else, but that's what I think they said.

When you free buzz your lips or buzz the mpc I don't think the result is a square wave, because of the lack of feedback. Despite that, the tone is somewhat "brassy" or "buzzy" and it's easy for me to imagine the trumpet smoothing that out somewhat and producing a mellow tone, free of "sizzle".

Dave
_________________
Schilke '60 B1 -- 229 Bach-C/19-350 Blackburn -- Lawler TL Cornet -- Conn V1 Flugel -- Stomvi Master Bb/A/G picc -- GR mpcs
[url=http://www.pitpops.com] The PitPops[/url]
Rocky Mountain Trumpet Fest
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Don Herman rev2
'Chicago School' Forum Moderator


Joined: 03 May 2005
Posts: 8951
Location: Monument, CO

PostPosted: Fri Feb 24, 2006 10:38 pm    Post subject: Reply with quote

Uhhh... I don't think free buzzing and sine/square waves get into the picture that much. I can make a pretty brassy buzz... I'm also pretty sure that not just odd harmonics are generated, but will have to go back and review the spectral plots (too tired, too late to do it now; I'm remembering older research). An ideal square wave has only odd harmonics; a triangle wave only even; a sine wave none (only the fundamental).

A flute produces almost a pure sine wave, and tends to add even harmonics when pushed. It sounds smoother, cleaner than a brass instrument, which does indeed have a square-wave sort of harmonic series (closed vs. open pipe, among other things). I can well imagine that, at low volumes, higher harmonics are not excitied, or not loud enough to be heard. At very high frequencies, tense/tight tissue has a harder time sustaining higher-frequency vibrations, and again the sound becomes more "pure" (less "brassy").

Too late for physics, even for an acoustics engineer - Don
_________________
"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
Back to top
View user's profile Send private message
dbolton
New Member


Joined: 02 Feb 2006
Posts: 3

PostPosted: Tue Feb 28, 2006 12:25 pm    Post subject: Re: Sound Distortion Reply with quote

dbolton wrote:

I noticed that the picture "Benny playing Pictures (2nd D) (far microphone)" shows distortion (the sound wave hits the top of the grey waveform area). Would this affect the frequency analysis?


I did a little experimenting, and the results show that clipping does indeed affect the frequency analysis. In short, a "pretty-looking" frequency chart doesn't necessarily equal a pretty sound.

F at the top of the staff (no clipping)


F at the top of the staff (amplified to force clipping)
[/img]

In case you're interested: here's the media clip the sample was taken from: Arban Page 192, N.5
_________________
David Bolton
Back to top
View user's profile Send private message Visit poster's website
Don Herman rev2
'Chicago School' Forum Moderator


Joined: 03 May 2005
Posts: 8951
Location: Monument, CO

PostPosted: Tue Feb 28, 2006 1:54 pm    Post subject: Reply with quote

Good analysis, David, though I don't think anybody would question that clipping, or any other (nonlinear) distortion would corrupt the frequency response. Gotta' keep the levels down, all through the chain -- don't want to overdrive the mic, preamp, or anything else before and including the ADCs.

An interesting sidelight is that, for music, the peak to average power ratio is about 17 dB (less for highly compressed music, like heavy rock, and maybe a little more for some orchestral pieces). That's a 50:1 power ratio, which helps explains why so many studios, performers, and stereophiles use relatively large amplifiers... - Don
_________________
"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
Back to top
View user's profile Send private message
Derek Reaban
Heavyweight Member


Joined: 08 Jul 2003
Posts: 4221
Location: Tempe, Arizona

PostPosted: Wed Dec 06, 2006 8:16 am    Post subject: Reply with quote

I bumped into this picture a while ago at this site. Clearly the plots that I have personally recorded do not match the “ideal” picture of resonances shown in this diagram. Also, everyone else that has taken the time to record a sample and presented the harmonic plot seems to look very similar to mine (see two posts up for example).




Here’s my D in the staff (4th line on C trumpet) from an earlier plot:



I’m sure there are reasons why my “harmonic spectrum” doesn’t align the Ideal plot, but if I was looking at these two charts with no description of what they were I would say that they were representing completely different instruments!

Am I really that far away from an “Ideal” sound?
_________________
Derek Reaban
Tempe, Arizona
Tempe Winds / Symphony of the Southwest
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Hipster
Veteran Member


Joined: 01 Sep 2006
Posts: 136

PostPosted: Wed Dec 06, 2006 8:24 am    Post subject: Reply with quote

Derek. I personally thought that "ideal" spectrum was of the fundamental resonances at each of the higher harmonics of a particular set of valve combinations, i.e. pedal c, low c, g, c, e, g etc., not a spectrum of a particular note and the overtones.
Back to top
View user's profile Send private message
kalijah
Heavyweight Member


Joined: 06 Nov 2003
Posts: 3257
Location: Alabama

PostPosted: Wed Dec 06, 2006 9:14 am    Post subject: Reply with quote

d wrote:
Quote:
I’m sure there are reasons why my “harmonic spectrum” doesn’t align the Ideal plot, but if I was looking at these two charts with no description of what they were I would say that they were representing completely different instruments!


To add to what hipster said...

The harmonics of a tone are not the resonances of the instrument. I guess you were expecting them to be. But , of course, the fundamental frequency of the tone IS a natural resonance of the instrument.

Any freqency can transmit thru the instrument and out the bell. The fact that a frequency is not a resonance does not mean it is attenuated, it just means that it can not reflect or resonate.

I recommend the following articles:

http://www.trumpetguild.org/pdf/2003journal/0306science.pdf

http://www.whc.net/rjones/jlynch/natres.html

to quote Lynch in the above article:

"The main points are that the trumpet's resonances are not actually harmonics of any actual note.".

Stop worrying about getting a "resonant" sound and just work toward getting a good sound.

The R-word is thrown around quite a bit here with some confused understanding of what it actaully is.
Back to top
View user's profile Send private message
dbolton
New Member


Joined: 02 Feb 2006
Posts: 3

PostPosted: Wed Dec 06, 2006 9:52 am    Post subject: Reply with quote

Derek Reaban wrote:
I bumped into this picture a while ago at this site.



The graph you showed of a so-called "ideal" resonance is displayed using log frequency (notice the numbers along the x-axis are not equally spaced).

You're own frequency analysis will look a lot closer if you switch from linear to log frequency (There's an option to do this in Audacity's frequency analysis window).
_________________
David Bolton
Back to top
View user's profile Send private message Visit poster's website
grooveduke
Heavyweight Member


Joined: 16 Dec 2005
Posts: 721
Location: Houston, TX

PostPosted: Wed Dec 06, 2006 11:39 am    Post subject: Reply with quote

Derek, if I understand this correctly, the chart you posted does more to explain why higher notes get progressively harder to play. But doesn't really address tone quality, at all.

In other words, the shape of the horn modifes the resonances (in the physics meaning) so that they roughly correspond to the harmonic series, giving us many usable notes to play on each valve combination.

That is the meaning of that chart in layman's terms, as I understand it.
_________________
Check out my loveletter to Rhythm & Blues:
Heavy Mariner
Back to top
View user's profile Send private message Visit poster's website
Derek Reaban
Heavyweight Member


Joined: 08 Jul 2003
Posts: 4221
Location: Tempe, Arizona

PostPosted: Thu Dec 07, 2006 10:10 am    Post subject: Reply with quote

Thanks for the comments everyone. When I saw the picture that I posted yesterday, it differed significantly from the recorded examples that I had analyzed. That 7th overtone (harmonic) was much larger than I was expecting to see. I was also confused when there was no scale associated with the vertical axis. I’m not sure what they’re showing (I had assumed decibels).

Anyway, my point with this experiment in the first place was to provide a starting point for a player to “see” their sound, and get a feel for how “resonant” it is. Resonant in this case would be rich in overtones. Based on my original experiment, I think an “ideal” sound would be close to what is shown in the chart below (Player A). I have shown 10 overtones, and the fundamental (note being played) is in the first position on the left for 3 players of different abilities. (Click on the picture to see a bigger, clearer view!)



If you took a snapshot of a top-flight symphony player, their “sonic signature” would look like Player A for every note that they played. Many of us would fall into the Player B picture, and have vibrant, resonant sounds for many notes, slipping in various degrees between Player A and Player C depending on the music being played. Player C represents a beginning student (lots of fundamental, very little energy in the higher harmonics).

Player A simply has more “energy” in their sound. Look at the area under the curve! There’s just MUCH MORE trumpet signal to be heard. And when these overtones have a good deal of energy in the 2000-4000 Hz range, human hearing really picks up on them at a distance. I’ve shown the 4th overtone (right in that 2-4K range) as being as strong as the fundamental for Player A (3 dB), much less for Player B (-30 dB) but still able to carry fairly well at a distance, and non-existent for Player C (-60 dB).

This is what I think this Sound Experiment should show. If a student would go through this experiment and find that they want to have a vibrant sound like Player A, but are really more between Player B and Player C then they would hopefully be much more receptive to suggestions from their instructor on reducing harmful tension in their playing, or understand that overblowing (masked as their idea of “more air”) will never get them to their desired “ideal” sound goal.
_________________
Derek Reaban
Tempe, Arizona
Tempe Winds / Symphony of the Southwest
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Hipster
Veteran Member


Joined: 01 Sep 2006
Posts: 136

PostPosted: Fri Dec 08, 2006 8:40 pm    Post subject: Reply with quote

Actually Derek, just from a physics point of view, the total energy, i.e. area under the spectra, should be constant at a fixed hmm intensity (sound volume). I personally would expect that instead of the variation in peak heights you are presenting here, that instead the peaks of the more experienced player would be sharper, up to the point of the "resolving" power of the particular instrument. Whereas the peaks of the less experienced players would be more spread, leading to a mushier sound. So, given a hypothetical perfect player and perfect instrument, all of the energy at a given frequency would be coalesced into a delta function at that frequency and it's overtones for the instrument and those of less perfect players and instruments could be approximated by gaussian functions of some value of sigma for that same set of frequencies.
Back to top
View user's profile Send private message
Jim-Wilson
Veteran Member


Joined: 11 Oct 2005
Posts: 415
Location: Fort Smith, AR

PostPosted: Sat Dec 09, 2006 3:37 am    Post subject: Reply with quote

As many of you know, I have great interest in all of this. Ultimately, only the sound that is perceived as it comes out of the end of the horn is important. But, our "understanding" of things can help and motivate us in our quest for better sound.

I really appreciate Derek's thoughtful time in encouraging others in this regard. I certainly think he's on track with the "carrying power" of world class player's resonant sound. How close we are to describing it via "captured waveforms" is another matter.

I'd like to re-post two spectral patterns I posted some time ago at the end of the thread "Visualizing Resonance". I do so only because the first spectral pattern is so unique and introduces into my mind ever more questions about spectral analysis and resonant sound.

[img][/img]

Has anyone else reproduced the "undulating" spectral pattern in the first analysis? I'm able to reproduce it only occasionally. I'm never able to reproduce it early in a practice session. It only occurs when I feel the note is very centered and I'm very relaxed. I really don't have a clue why it comes out as it does and wonder how it factors into the issue of the "sonic signature" of what I think is a full, resonant tone.

Anyhow, I don't mean to add to the muddle - I'm still trying to learn though.

Jim
Back to top
View user's profile Send private message
Derek Reaban
Heavyweight Member


Joined: 08 Jul 2003
Posts: 4221
Location: Tempe, Arizona

PostPosted: Mon Dec 11, 2006 9:36 am    Post subject: Reply with quote

Hipster,

Thanks for your comments. While I know that experiments like these venture into technical discussions about physics and advanced math concepts, it would really be nice to simply be able to present a diagnostic picture of a student’s "tone" quality. A chart of the student’s "sonic signature" side by side with an "ideal" orchestral tone picture would highlight that more overtones are ringing in the pros sound.

If it was presented so clearly that a Jr. High student could understand the idea, I think it could be used by almost any player that was interested in knowing if they were getting the most out of their “long tone” practice.

If a player has found that great vibrant and resonant sound on a second line G, and the "picture" of their tone is very similar to the "ideal" picture, they would have a visual cue that they are on the right track. If they continue their practice up to the 4th space E and still "feel" relaxed but are playing a little louder to get the notes to speak easily, they might believe all is still going well. But when things get tight or sound tense at a G or A at the top of the staff, they may feel that their sound production issues need to be addressed in this range. What would happen if the "picture" of their tone stopped lining up with the "ideal" picture around a 4th space E or even a 3rd space C?

I know that ultimately our ears must guide our progress, but this visual guidepost could help a young player realize that the train may have come off the tracks much earlier than they thought it did (along with guidance from a competent private instructor to get things back on track). How much more beneficial would long tones be to a player if they "knew" that they were ringing the higher overtones, while striving for higher notes (or lower notes for that matter)?

I’m just thinking out loud here, but wouldn’t stopping to fix sound production issues at the first note (or a small range of notes) that has a significantly different “sonic signature” allow the player to carry the relaxed energy of efficient sound production in a range that is easy to play gradually into the higher registers. For the player that can play a High C but discovers that their "sonic signature" deviates from a great orchestral tone at a 3rd line B, they will have some serious soul searching to do if they really want to make serious inroads to improved sound production!

As the dedicated student "sees" a great sound via these "spectral plots" and corresponds this sound to what they are hearing from behind the bell, eventually they would hone in on this vital component of the sound, and they would simply know that their sound had the energized, ringing overtones that make up a quality orchestral sound.

Making it simple to use is the task here (provided my logic is "sound")! I’m not sure I’m the one to do this, but its fun to think about what is possible with free online tools that are available to everyone simply by plugging a microphone into the back of a computer.


Jim,

Glad you’re enjoying this project. Thanks so much for turning me on to the Audacity tool. It’s amazing what you can get for free on the computer these days!

Quote:

Has anyone else reproduced the "undulating" spectral pattern in the first analysis? I'm able to reproduce it only occasionally. I'm never able to reproduce it early in a practice session. It only occurs when I feel the note is very centered and I'm very relaxed. I really don't have a clue why it comes out as it does and wonder how it factors into the issue of the "sonic signature" of what I think is a full, resonant tone.


I haven’t experienced the "undulating" pattern in anything that I have recorded. I wouldn’t rule out that something special is happening in your sound, but it seems like something electronic would be making that pattern appear. This is exactly why a process on how to use this tool would be great to have. You selected 4096 as the default for presenting your spectral plot and I chose 512. Maybe I should plot my 2nd line G at 4096 and compare it to your "sonic signature" and see where we are similar and different.

The thing that is most intriguing about your plot to me is that the higher overtones are stronger in magnitude than your fundamental. To me that says, "Way to go Jim! You’re on the right track!"


I may have some time over the Christmas break to look at this some more. I’m too busy practicing for jobs to think about this right now.
_________________
Derek Reaban
Tempe, Arizona
Tempe Winds / Symphony of the Southwest


Last edited by Derek Reaban on Fri May 03, 2013 8:43 am; edited 1 time in total
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Hipster
Veteran Member


Joined: 01 Sep 2006
Posts: 136

PostPosted: Mon Dec 11, 2006 9:37 pm    Post subject: Reply with quote

Derek Reaban wrote:
Hipster,

Thanks for your comments. While I know that experiments like these venture into technical discussions about physics and advanced math concepts, it would really be nice to simply be able to present a diagnostic picture of a student’s "tone" quality. A chart of the student’s "sonic signature" side by side with an "ideal" orchestral tone picture would highlight that more overtones are ringing in the pros sound.


Yeah sorry. I'm basically a physics geek most of the time and wanna be trumpeter the rest.

Jim's pictures of his resonant vs. non-resonant sounds however are a very good pictoral representation of the point I was making. If you look a the peaks between 23Hz and 7KHz notice how wide the peaks are in the non-resonant vs. the sharpness of the peaks in the resonant. I personally think that feature is more important towards a "resonant" sound than all of those other overtones.

Just to geek out again I'm wondering if the secondary wave in the peaks of his aren't an aliasing artifact or some sort of background hum. It would also be really nice if you guys could post up the sampling rates you used to digitize your clips.

Edit: Please forgive further geeking. Could you guys also redo those images with the windowing turned off. The window just by itself is going to add in some odd sidebands.
Back to top
View user's profile Send private message
Jim-Wilson
Veteran Member


Joined: 11 Oct 2005
Posts: 415
Location: Fort Smith, AR

PostPosted: Tue Dec 12, 2006 3:42 am    Post subject: Reply with quote

Hipster and Derek,

I'm on my way out of town until Monday so will have to give further thought/reply to things later. A brief thought on the "semi-sinusoidal" nature of the first spectral pattern - both of the samples were from the same note/same sampling setup, etc. I suppose there could be some transient artifact involved but I truly have the sense it is more a reflection of something acoustic going on. I'll try to reproduce it next week sometime. BTW, I should have a Harrelson demo horn in hand to play with next week as well - might provide some new interesting "feedback".

Jim
Back to top
View user's profile Send private message
MannyLaureano
Veteran Member


Joined: 11 Oct 2003
Posts: 405
Location: Plymouth, Minnesota

PostPosted: Tue Dec 12, 2006 5:49 am    Post subject: Reply with quote

So, is all of this to say that a player who plays efficiently and in the pitch center of a given note will be heard better than a less efficient player on the same note at any volume?

ML
_________________
Liviu Librescu-1930-2007- Holocaust survivor and hero during VA Tech massacre
Back to top
View user's profile Send private message Send e-mail Visit poster's website
grooveduke
Heavyweight Member


Joined: 16 Dec 2005
Posts: 721
Location: Houston, TX

PostPosted: Tue Dec 12, 2006 7:05 am    Post subject: Reply with quote

OK, You guys are way out there in the physics language for me! But after re-reading hipster’s posts, I begin to wonder.

Quote:
Jim's pictures of his resonant vs. non-resonant sounds however are a very good pictoral representation of the point I was making. If you look a the peaks between 23Hz and 7KHz notice how wide the peaks are in the non-resonant vs. the sharpness of the peaks in the resonant. I personally think that feature is more important towards a "resonant" sound than all of those other overtones.



Here is my question:
How long was the sample?
Was vibrato used?
If it was, which part of the vibrato did you sample?
If the software analyzes the harmonic content of the entire sample over a given period of time, wouldn’t the wider peaks in the sample be more of a function of the type of vibrato used, than the "resonance" of the sound itself? Or am I misunderstanding how the software analyzes the sound?

In other words, your notion is intriguing, but I still believe it is the presence of those overtones (which indicate that you are playing through the center of the horn) and the relative strength of each overtone (as exhibited in Derek’s post of Thu Dec 07, 2006 10:10 am ) that contribute to the “resonant” sound.
_________________
Check out my loveletter to Rhythm & Blues:
Heavy Mariner
Back to top
View user's profile Send private message Visit poster's website
Derek Reaban
Heavyweight Member


Joined: 08 Jul 2003
Posts: 4221
Location: Tempe, Arizona

PostPosted: Tue Dec 12, 2006 7:32 am    Post subject: Reply with quote

Manny,

Quote:
So, is all of this to say that a player who plays efficiently and in the pitch center of a given note will be heard better than a less efficient player on the same note at any volume?


Almost. In close proximity, a “loud” player (lots of fundamental and less strength in the overtones) may seem to dominate the efficient player who is in the pitch center of the note and has lots of vibrancy. This would be the sound you hear from behind the bell.

In the hall, the “loud” player will no longer be dominant, and the efficient player in the pitch center of the note will be heard very clearly at the back of the hall.

All of this, of course, is related to an orchestral sound concept.

If you have a recorder, you can try this with a couple of your students during a lesson. You don’t have to do any of the math, just look at the “peaks” on the indicator during the playback and see how different your students sound is than yours at a distance. Depending on the maturity of the student, they may register a 1/3 on the scale and your sound would be pegging the needle.

Getting them to understand that “the louder you play, the less it carries” is certainly not something that is intuitive as indicated by the poll results here.

Thanks for stopping by!
_________________
Derek Reaban
Tempe, Arizona
Tempe Winds / Symphony of the Southwest
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Jim-Wilson
Veteran Member


Joined: 11 Oct 2005
Posts: 415
Location: Fort Smith, AR

PostPosted: Tue Dec 12, 2006 7:47 am    Post subject: Reply with quote

Mark Cornell wrote:

Quote:
Here is my question:
How long was the sample?
Was vibrato used?
If it was, which part of the vibrato did you sample?


Mark,

I didn't actually keep the "raw" sound file but my recollection was that the sustained tone was about 10 secs. I "sampled" spectral analyses at varying time frames within the 10 sec sample and captured just the two you see above. My practice is to highlight a .15sec segment for sampling. Hipster makes reference to not using "windows" which I think relates to your question. Within Audacity you have several choices as to how the software "analyzes" the sample which relates to the "window" approach. I think this has to do in some fashion with a "time averaging" of the sample provided to the software for analysis. I'm a rank amateur in all of this and will let the engineers and physicists among us address the windows/sampling issues. I just opted for a short time segment and the highest sampling rate I could. My recollection is I sampled at 44Khz. Also, there was no vibrato involved (at least non intended!).

A thought concerning the "semi-sinusoidal" nature of my "resonant" sample - it's certainly possible that what was captured related to vibration in the horn and not something exclusive to the air column/radiated sound itself. But, I don't know.

Manny Laureano wrote:

Quote:
So, is all of this to say that a player who plays efficiently and in the pitch center of a given note will be heard better than a less efficient player on the same note at any volume?


I'll let Derek respond to this in more detail since he's really the one that has been the driving force behind this discussion. But, yes, I think that is his point - to some extent. I'm sure the "any volume" issue would need to be clarified in some fashion. As a 4th grade player I could make some pretty loud, lousy sounds that might obliterate a beautiful virtuoso pianissimo!

Jim
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    trumpetherald.com Forum Index -> Fundamentals All times are GMT - 8 Hours
Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8  Next
Page 4 of 8

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group