M/Pieces - Ligs Mouthpiece Material matters

[Morgan Fry]

Rather, the matter of whether mouthpiece matter, erm, matters --

OK, this isn't an invitation to beat a dead horse -- it's an opportunity to find out for yourself whether you can tell the difference.

Nobody has really done a very thorough study of this, in part because nobody markets mouthpieces that are identical except for the material.

So I made some.

Several mouthpieces, identical in every possible way -internal dimensions, external dimensions, facing, everything -- out of some different materials (I don't want to skew the results by revealing how many mouthpieces or made of what, except to say at least one is brass and at least one isn't) I took them down to Dave O'Higgins' studio in London. We recorded multiple clips of him playing each mouthpiece with otherwise the same setup.

8 randomly selected pairs of clips of Dave playing the same tune. Some are different clips with the same mouthpiece, some are different clips with different mouthpieces. If we get enough participation it might be interesting to see whether we have a significant result.

It might not be the most scientifically rigorous possible test, but I think it's more important to see what happens in real life, with real players playing real music.

The blog post
The Mouthpiece Challenge
The details

Give it a go and let me know what you think!

 

[aldevis]

Maybe I am slow. Do I have to register to take the test or only to see the results?

 

[Morgan Fry]

Maybe I am slow. Do I have to register to take the test or only to see the results?

Have to register to take the test -- the quiz module I'm using doesn't keep the data from anonymous users properly so their results skew the statistics. You're welcome to register with a nonsense email address if you're concerned but I'm not sharing any emails with anybody.

 

[aldevis]

I don't think you will spam me with mails like "enlarge the size of your Bflat"

 

[aldevis]

I had a fairly good score, I decided to listen to the samples in a rush, sticking with the first impression.
Quite consistent sound, though. I hope we will eventually know the opinion of the player.

 

[kevgermany]

Was really difficult to hear much difference between many of the tracks. I got 5/8, about what you'd expect from random replies without listening...

 

[Chris]

There wasn't a lot of difference between the tracks. Interesting test though..

Chris

 

[Nick Wyver]

I answered that they all sounded the same, Picking out the different ones would have been pure guesswork.

 

[daveysaxboy]

Every player will hear,feel much more than any 1 else.When i play a metal mp i do.Same with a HR piece.Thats all what matters in the end.Its fun going about the science/physics of it but it dont mean i thing in the end.I cant settle of metal alto mps.Why well i think they sound metal.I find the same with HR tenor pieces.They dont have that zing.I have had a few thin profile like metal hr tenor pieces that were great but never stay with me long.Unlac horn vs lacquered horn.If you feel it does thats the best result.These topics are never ending.Keeps us talking so its a good thing.

 

[Morgan Fry]

For those who have already taken the quiz, I have updated/upgraded it to include the original wav files. Dave brought to my attention that the mp3 compression affects the sound enough to matter. So if you want to try again, you might be surprised at the difference if you have good sound setup to listen on.

 

[Merryfisher]

time everyone got out a little bit more!

 

[Nick Wyver]

time everyone got out a little bit more!

Why? If nobody ever thought this sort of stuff mattered you wouldn't be playing on a saxophone now - or using a computer for that matter.

 

[aldevis]

For those who have already taken the quiz, I have updated/upgraded it to include the original wav files. Dave brought to my attention that the mp3 compression affects the sound enough to matter. So if you want to try again, you might be surprised at the difference if you have good sound setup to listen on.

The risks to go off topic is high; I think there is a HUGE difference between a recorded saxophone and a live saxophone.
But on the other hand we select the best mouthpiece in the world for our best horn, and after spending days in finding the right reed we walk on stage.

Where a sound engineer will decide that a bagpipe sound would be better.

 

[kevgermany]

Morgan, suggestion:

To get something meaningful out of this, suggest you ask around for the services of a statistician. Bound to be one on one of the forums who'd help analyse the data properly.

Tried to listen to the reposts earlier, but too slow on WLAN, will try later on cable lan.

 

[Morgan Fry]

But on the other hand we select the best mouthpiece in the world for our best horn, and after spending days in finding the right reed we walk on stage.

Where a sound engineer will decide that a bagpipe sound would be better.

Lol don't even get me started on sound guys. That's one thing I like about acoustic small group gigs, you sound how you sound. The rest of the time you only ever sound as good as your sound guy --which is fine if he's on the ball, I just ran into too many of the other kind.

Morgan, suggestion:

To get something meaningful out of this, suggest you ask around for the services of a statistician. Bound to be one on one of the forums who'd help analyse the data properly.

That's a good idea. Will do that when I get stuck and realize I don't remember my college statistics class as well as I though I did

 

[jbtsax]

I took the test twice. The first time some of the .wav files didn't play. I am trying to process what the results of the test will tell, if anything. If a majority of participants say that clips of mouthpieces of different materials are the same mouthpiece that would tend to show that the material doesn't matter.

What if a majority of the participants say that two clips using the same mouthpiece are different? Will that reflect that the participant is not a skilled listener? Or might it reflect the fact that the player played with a slightly different sound/volume/intensity on one of the takes?

I would have liked to have heard long tones on the same pitch in each example, and notes throughout the range of the instrument. Also I would have preferred to hear exactly the same music passage in each example. It will be interesting to hear from the player as to whether the mouthpieces "felt" different as he played them.

The bioacoustic feedback to the player is one of the elements of this acoustics question that is the hardest to measure, but appears to be very real by the abundance of anecdotal evidence from very accomplished players.

 

[aldevis]

What if a majority of the participants say that two clips using the same mouthpiece are different? Will that reflect that the participant is not a skilled listener? Or might it reflect the fact that the player played with a slightly different sound/volume/intensity on one of the takes?

It means that the variations of the player (let us not forget the axis response of a microphone) influence the sound more that the material.

Under the epistemological point of view (sorry for the word) Morgan is doing the only thing that makes sense.

On these forums there is alway something claiming that the only scientific method would be having the same exact conditions EXCEPT the mouthpiece.

This is a 19th century concept of science (Positivism) and cannot be applied to some disciplines (astronomy, geology, social sciences...).

Statistic incidence is not an absolute truth, but is the closer we can get to a prediction of the behaviour of a mouthpiece.

It would be interesting to hear the same test with different players.

I know of similar tests (by a different mouthpiece maker) that approached statistically the spectrum of several mouthpieces/players finding consistent peaks on some harmonics (on one particular make) whoever the player was.

 

[aldevis]

It will be interesting to hear from the player as to whether the mouthpieces "felt" different as he played them.

Hopefully we will eventually know.

 

[Morgan Fry]

What if a majority of the participants say that two clips using the same mouthpiece are different?

I don't think it matters, wrong one way is as good as wrong the other way.

Will that reflect that the participant is not a skilled listener? Or might it reflect the fact that the player played with a slightly different sound/volume/intensity on one of the takes?

I'm interested in this, so in the questionnaire at the end I ask about playing experience as well as listening equipment.

I would have liked to have heard long tones on the same pitch in each example, and notes throughout the range of the instrument. Also I would have preferred to hear exactly the same music passage in each example.

I don't think there's any point to doing it this way. The way I see it is either you do a completely objective test -- no player involved -- an artificial embouchure and apparatus to eliminate every variable -- embouchure pressure, air pressure, ligature pressure, reed placement, etc. -- and play a few notes and compare the sound spectra. If you want to eliminate the variations due to the player, eliminate the player.
Or, you do a practical test -- a real player playing real music and you ask a bunch of people what they hear. This I have the technology to do, and it's more fun.

 

[aldevis]

I don't think there's any point to doing it this way. The way I see it is either you do a completely objective test -- no player involved -- an artificial embouchure and apparatus to eliminate every variable -- embouchure pressure, air pressure, ligature pressure, reed placement, etc. -- and play a few notes and compare the sound spectra. If you want to eliminate the variations due to the player, eliminate the player.
Or, you do a practical test -- a real player playing real music and you ask a bunch of people what they hear. This I have the technology to do, and it's more fun.

That is what I tried to explain in my post, but the word "epistemology" put me on a slightly different path.
What about trying different players too?
Of course I volunteer for testing, with my band Aldevis and the Guinea Pigs.