Is the research on violin acoustics a viscious circle?

[Andrew tkinson]

5 hours ago, Marty Kasprzyk said:

One of my violins was in a blind listener and player test along with five other violins.  Nobody in the listener group of about two dozen people picked mine as their favorite.  The only one who liked it best was the player.

 

 

 

I hope nobody minds me mentioning guitars! Many years ago I made two guitars to try to make an extreme comparison to try to judge the effect of the size on sound and volume.

Needless to say things are a bit too complicated but what I did learn was that the appearance of an instrument has a bigger effect on how people perceive it than I had previously thought and I reckon if guitar is basically good, people will like the one that they think looks best and will grow to adapt to and like how it sounds. 

Here is a photo of the more conventional guitar, my 'single elephant' model (see inlay at 12 fret)

This guitar sounds nice, but to me, and most listeners - when both guitars were played behind them so they couldn't see - its bigger brother (the double elephant - modelled on an early 20th century Washburn "Conservatory Monster" model) was generally agreed to sound nicer. But sadly when they turned around and could see it, in all of its glory they seemed to change their mind for some reason! 

Here it is, photographed today, as was the other one, in the front garden in the sun!

I have to admit the real thing that puts people off it is they can't hold it to play. Luckily I am about 6ft 4" so can manage it ok. It is is a bit awkward to carry on buses but not as bad as my double bass!

[David Burgess]

5 minutes ago, Andrew tkinson said:

This guitar sounds nice, but to me, and most listeners - when both guitars were played behind them so they couldn't see - its bigger brother (the double elephant - modelled on an early 20th century Washburn "Conservatory Monster" model) was generally agreed to sound nicer. But sadly when they turned around and could see it, in all of its glory they seemed to change their mind for some reason! 

Not atypical of human behavior.

I don't think that there is any human, no matter how “enlightened”, who is totally and completely immune to peer pressure, social pressure, and current or prior indoctrination.

[Bill Merkel]

16 minutes ago, Andrew tkinson said:

I hope nobody minds me mentioning guitars! [...] what I did learn was that the appearance of an instrument has a bigger effect on how people perceive it than I had previously thought

It's well established that guitar players hear with their eyes, or at least it's a well-established joke

[Bill Merkel]

I have a mild touch of synthesia.  I kind of hear partly with my inner eye or imagination.  I might go to a recital and the violin has produced a purple cloud over the stage in my mind's eye, or maybe a shower of sparks.  Would a different violin do the same thing?  A different violinist certainly might not...  Perlman didn't do anything except play great that I remember, while a Tchaikovsky competition also-ran was a light show...

[David Burgess]

20 minutes ago, Bill Merkel said:

I have a mild touch of synthesia.  I kind of hear partly with my inner eye or imagination.  I might go to a recital and the violin has produced a purple cloud over the stage in my mind's eye, or maybe a shower of sparks. 

Are you sure that wasn't a Jimi Hendrix concert, in which he played "Purple Haze", and then smashed or burned his guitar?

[Bill Merkel]

Could be my eyes and ears are both just really bad and that's what was going on.

[Andrew tkinson]

53 minutes ago, Bill Merkel said:

I have a mild touch of synthesia.  I kind of hear partly with my inner eye or imagination.  I might go to a recital and the violin has produced a purple cloud over the stage in my mind's eye, or maybe a shower of sparks.  Would a different violin do the same thing?  A different violinist certainly might not...  Perlman didn't do anything except play great that I remember, while a Tchaikovsky competition also-ran was a light show...

I was reading about synaesthesia on wikipedia and found this quote from a man called Solomon Shereshevsky who apparently seemed to have all of his senses linked in some way

"One time I went to buy some ice cream ... I walked over to the vendor and asked her what kind of ice cream she had. 'Fruit ice cream,' she said. But she answered in such a tone that a whole pile of coals, of black cinders, came bursting out of her mouth, and I couldn't bring myself to buy any ice cream after she had answered in that way"

I wonder if the vendor had a sore throat on that day and her voice was a bit extra husky? Perhaps if he had caught her on a different day he would have bought some? Or perhaps something in her appearance, something he had just smelled or maybe he had a bad experience on his way to the shop caused this unpleasant effect in him?

As you can see from my clumsy sentences, I can't comprehend what it must be like to have such mixing of senses. At the other end of the sensory scale, perhaps, I once worked with someone who said he "didn't see the point of music"!

[Bill Merkel]

19 minutes ago, Andrew tkinson said:

I can't comprehend what it must be like to have such mixing of senses. 

Thanks for spelling it right.  What I wrote was some Microsoft product.  For me it's like when you read a story and picture the scene going on in your head.  Take that but change it to involuntary shapes and colors that go with some sounds, particularly violin sounds

[Andreas Preuss]

2 hours ago, Andrew tkinson said:

I hope nobody minds me mentioning guitars! Many years ago I made two guitars to try to make an extreme comparison to try to judge the effect of the size on sound and volume.

Needless to say things are a bit too complicated but what I did learn was that the appearance of an instrument has a bigger effect on how people perceive it than I had previously thought and I reckon if guitar is basically good, people will like the one that they think looks best and will grow to adapt to and like how it sounds. 

Here is a photo of the more conventional guitar, my 'single elephant' model (see inlay at 12 fret)

This guitar sounds nice, but to me, and most listeners - when both guitars were played behind them so they couldn't see - its bigger brother (the double elephant - modelled on an early 20th century Washburn "Conservatory Monster" model) was generally agreed to sound nicer. But sadly when they turned around and could see it, in all of its glory they seemed to change their mind for some reason! 

Here it is, photographed today, as was the other one, in the front garden in the sun!

I have to admit the real thing that puts people off it is they can't hold it to play. Luckily I am about 6ft 4" so can manage it ok. It is is a bit awkward to carry on buses but not as bad as my double bass!

Thanks for posting this. Though it has nothing directly to do with violins, your experiment demonstrates what perception musicians have towards instruments. Yes, we live in a very visual oriented world. And maybe this is also triggered in our brains through films, advertising etc. where with the means of pictures plus sound an ‘ideal’ and ‘desirable’ object is shown. 
 

Otherwise it might be a new field to study psycho-visual acoustics.

In violin making most experimental violins changed something pretty visibly and radically. There are reasons why they didn’t work acoustically and visually some of them don’t look like what we would expect. 

 

I think that slight deformations bear much more chances of success. Most famed violins are not mirror-precise symmetric and IMO this should give us some ideas. 

[Andreas Preuss]

19 hours ago, jezzupe said:

Oh I'm not here to accuse people of "not getting it" nor defend my position or opinion really, again I just state the facts as they seem to be known,and that seems to be that in numerous blind test's that have been done over the years "great Italian" violins do not stand out and are not something that "just anyone" can easily clearly pick out as "superior" to other instruments be they modern or older, French, Italian, or another nationality, and I'd just let that "statement" do the talking, unless of course someone has links to some test that was done where it was shown that "they" clearly were better and picked out among others?

That being said I'm more concerned about what is not being focused on and that is developing, proving and having contests about what does matter SUPERIOR SPECIAL HEARING ABILITIES...to me instead of the same old "let's try to prove Strad or other great makers are the best" I would be much more interested in trying to prove, go toe to toe, mano o' mano against other peers in "listening contests" where a person is given a gold star award and an all you can eat buffet at Denny's for proving their ears are "best"

So I am much more impressed by a person listening to say 7 to 10 violins blind. no names given, only numbered, give them 2 minutes of listening to each, in a room where they can walk around and hear it at a distance and close and then simply play them again after the introduction in random order, say 30 seconds each and then let them list and tell us which one is which, several rounds over several days with various instruments could be done and the folks with the best record of correct picks would be crowned EAR MASTER !!! not to be confused with beast master

That to me matters more than "is Strad sound the best" because then when it comes right down to it at least "we" or one could have some actual brevity behind their opinion if in fact one could do as I suggest, it may be that no one can, and it's all just dumb luck, but I have a feeling that certain people would stand out if such contests took place

EDIT; thinking about this some more I would say that these contest's should start with just 2 violins and then work their way up in numbers in an elimination process where several listeners participate at the same time to speed it along, that being said, I bring up this developing and proving,or demonstrating "superior hearing" because I want to give Strad and the old Italian guys the benefit of the doubt that it's not them, it's us. Meaning there very well may be listeners that can identify "Italian sound" and maybe very specifically. 

It's just that if we were to consider "blind listening test's" a "scientific" endeavor, I would say in the Italians behest that it has not been a very big control group as far as "numbers" and proof of quality of participants.

I really do wonder what "category" person would be EAR MASTER ? I would think it would be between a maker and a player? but who knows, maybe someone who is an advanced music listener who does not even play would win? perhaps being blind would help?

I don't know but I sure would like to see the development of such skills fostered in a more professional, say Oberlin like environment and I think contest's are a great way to do it, as they are somewhat like a performance 

Because, well maybe Strads do sound superior or markedly different, we just haven't had the right people at these blind test's?

I think you outlined pretty precisely how participants of a sound evaluation test should be selected. 
 

On the bottom line I think that the opinion of the listener is absolutely irrelevant because

I never heard any concert goer saying that he/she didn’t like the performance because the instrument didn’t sound well. There are clearly other objectives in a live performance which are by far more important to concert goers, namely the interpretation of a piece with all its aspects. And if the instrument doesn’t sound right, most of the time remains are sought other than the instrument itself, often just the climate in the room. Or, if there are some people in the audience thinking the instrument was bad, they belong to a negligible small minority of concert goers. 

However

the positive sideline of the experiments by Claudia Fritz was clearly that modern instruments can be appreciated by players. Maybe for other reasons than the old ones, but as sound has no objective criteria for ‘really good’ this should be good enough. It would have been interesting though to include a violin which was not constructed to standard norms.

The detailed discussion on this goes into a never ending opinion battle about the pros and cons of those experiments and does not really belong into this thread.

[David Burgess]

45 minutes ago, Andrew tkinson said:

"One time I went to buy some ice cream ... I walked over to the vendor and asked her what kind of ice cream she had. 'Fruit ice cream,' she said. But she answered in such a tone that a whole pile of coals, of black cinders, came bursting out of her mouth, and I couldn't bring myself to buy any ice cream after she had answered in that way"

Might have been one o' them cross-dressers, sometimes readily recognized from the pitch and overtones of their voice. I became semi-skilled at that when I was living in Hollywood, because otherwise, "doing the Hoky-Poky" could involve more surprises than I was looking for.

So that's how I will claim to have  refined my aural sorting skills, under the pressure of great peril.

 

[Andrew tkinson]

I hope nobody tells me off for 'not doing my homework', (I did do a quick search and couldn't find anything) but I wonder if it would be fruitful to perform some useful acoustical research on violins using non-traditional materials of a homogenous and easily shapeable nature?

If a homogenous material such as a thermoplastic was used to make various soundboards of differing arching shapes or f-hole spacing etc, they could then be attached to a violin body and set up and tested. The homogenous nature of the plastic would perhaps mean that the effects of the changes in arching, sound hole spacing etc on the acoustic performance can be measured and can be seen to be attributed to these changes without any interference from the variability of natural materials and workmanship ?

I am not suggesting that these materials be adopted but maybe the results from such experiments could be useful, in the design stages, when making traditional instruments, in the customary and well proven materials? 

I am hoping that this idea may have been seen as being too 'stupid' to have been done before, (I seem to be becoming a specialist in my own April Fool School of instrument cogitation) but surely makers of carbon fibre violins must have experimented to get the best results?

Perhaps computer modelling can do this type of modelling well enough already, I have to admit that my mind seems to switch off when confronted with anything much more than basic arithmetic so I fully admit to having not done any homework in this area.

Again, please be gentle on me if this type of experiment has been done or it is genuinely too stupid!

[David Burgess]

59 minutes ago, Andreas Preuss said:

I never heard any concert goer saying that he/she didn’t like the performance because the instrument didn’t sound well.

I have, but didn't place much faith in it.

[sospiri]

16 minutes ago, David Burgess said:

Might have been one o' them cross-dressers, sometimes readily recognized from the pitch and overtones of their voice. I became semi-skilled at that when I was living in Hollywood, because otherwise, "doing the Hoky-Poky" could involve more surprises than I was looking for.

So that's how I will claim to have  refined my aural sorting skills, under the pressure of great peril.

 

We owe you a debt of gratitude for such dilligent research in such perilous locations as those major intersections in Hollyweird.

 

4 hours ago, David Burgess said:

Are you sure that wasn't a Jimi Hendrix concert, in which he played "Purple Haze", and then smashed or burned his guitar?

Purple Haze, then Wild Thing, then the ceremonial pyre. Monterey Festival June 18 1967. No, I wasn't there either.

[David Burgess]

25 minutes ago, Andrew tkinson said:

If a homogenous material such as a thermoplastic was used to make various soundboards of differing arching shapes or f-hole spacing etc, they could then be attached to a violin body and set up and tested. The homogenous nature of the plastic would perhaps mean that the effects of the changes in arching, sound hole spacing etc on the acoustic performance can be measured and can be seen to be attributed to these changes without any interference from the variability of natural materials and workmanship ?

The Oberlin Violinmaking Workshop did something along those lines in 2017, incorporating many variables into a bunch of instruments (and then evaluating the results), but still using wood. So their use of wood might have been the biggest fly in the ointment, if one wanted to use a less variable material.
See what you think of that effort:

http://www.itzelavila.com/blog/2017/7/2/oberlin-2017

[Don Noon]

2 hours ago, Andrew tkinson said:

I wonder if it would be fruitful to perform some useful acoustical research on violins using non-traditional materials of a homogenous and easily shapeable nature?

A couple of years ago I made a back out of MDF.  It wasn't too bad, but maple was definitely better.  An MDF top I think would be horrid.  

I doubt that this kind of research would be too useful, as it is too far from "what works good".  After all, we have some pretty detailed acoustic studies of very good instruments, mode shapes and frequencies and all that... which hasn't resulted in any great practical advances that I'm aware of.

[Andrew tkinson]

3 hours ago, Don Noon said:

A couple of years ago I made a back out of MDF.  It wasn't too bad, but maple was definitely better.  An MDF top I think would be horrid.  

I doubt that this kind of research would be too useful, as it is too far from "what works good".  After all, we have some pretty detailed acoustic studies of very good instruments, mode shapes and frequencies and all that... which hasn't resulted in any great practical advances that I'm aware of.

Hello Don. What I meant was the idea of experimenting with a very homogenous material to produce a series of differently arched tops or same top with differently spaced f holes for example. Then fixing these, each in turn, to a single 'demountable' body test unit, starting with the flattest and working upwards and performing  a set of tests on each one.

I was thinking that if a very even, homogenous material was used then the changes detected would be attributable only to the changes in the arching not the variations introduced the material or in workmanship

I then thought, even though the plastic or whatever would not behave in such a complex way as real wood that any general trends if observed could be considered as moving in the same direction with a real spruce soundboard. I was thinking that it would remove the chance of anyone objecting to the observed results by saying the variations in the materials and in the test pieces used may have muddled any test results?

What I was envisaging is a bit like when an economist or scientist simplifies a system or looks at one part of a process and tests what happens when only one thing is varied.

I envisaged plastic would be a good choice as it is homogenous and it could be used to produce a lot of identical test pieces quite quickly. If a college has a CNC machine and acquired some 3d violin modelling software, if such a thing exists, a series of internal soundboard moulds with gradually rising arching could be machined from MDF. Then after the moulds have been prepared a suitable thermoplastic could be vacuum formed over the mould to produce a series of test soundboards which changed incrementally but were materially identical. Once the moulds have been made, experimental soundboards could be made easily so a lot of tests could be done each varying only one thing. Maybe a thermoplastic can be obtained which has a sort of grain like wood which isn't disturbed by the heating and forming process? Also, for arching tests the greater stretching and tensions caused by moulding higher archings would reduce the homogeneity of the plastic and would have to be considered and avoided if possible.

I was reasoning perhaps optimistically that the general trends observed when varying only one aspect of a plastic soundboard would maybe move in a similar direction with real wooden soundboards and maybe such information may help some people when trying to come up with their own slight variation on a violin design

Apologies for the long explanation of my thoughts. I sometimes get the feeling that the more I try to explain the less clear I am being!

I fully realise though, that well trained and highly skilled instrument makers seem to manage quite well using their experience and hard won skills and have no need for more pages of graphs and lists of frequencies etc.  Another slight worry I have is the risk of incurring the wrath of Greta Thunberg for seeming to desire to increase the amount of plastic waste. So if this experiment is actually worth doing I would hope that the plastic used must all be recycleable?

I must admit that being a simple man I am more excited by the idea of making and experimenting with my own luthier's "schnitzbank" !

[Andreas Preuss]

6 hours ago, Andrew tkinson said:

I hope nobody tells me off for 'not doing my homework', (I did do a quick search and couldn't find anything) but I wonder if it would be fruitful to perform some useful acoustical research on violins using non-traditional materials of a homogenous and easily shapeable nature?

If a homogenous material such as a thermoplastic was used to make various soundboards of differing arching shapes or f-hole spacing etc, they could then be attached to a violin body and set up and tested. The homogenous nature of the plastic would perhaps mean that the effects of the changes in arching, sound hole spacing etc on the acoustic performance can be measured and can be seen to be attributed to these changes without any interference from the variability of natural materials and workmanship ?

I am not suggesting that these materials be adopted but maybe the results from such experiments could be useful, in the design stages, when making traditional instruments, in the customary and well proven materials? 

I am hoping that this idea may have been seen as being too 'stupid' to have been done before, (I seem to be becoming a specialist in my own April Fool School of instrument cogitation) but surely makers of carbon fibre violins must have experimented to get the best results?

Perhaps computer modelling can do this type of modelling well enough already, I have to admit that my mind seems to switch off when confronted with anything much more than basic arithmetic so I fully admit to having not done any homework in this area.

Again, please be gentle on me if this type of experiment has been done or it is genuinely too stupid!

The problem with maybe all artificial materials is that they are homogeneous. It seems that the top plate needs different stiffness along and across the grain. And within those parameters spruce is additionally extremely light.

In a certain way you could look on spruce like a nature made sophisticated composite material where the hard winter grain and the soft summer grain are made of two different materials to achieve a very specific result in terms of stiffness, damping and speed of sound.
 

If anything could be replaced it would be everything except the top plate. In fact violin makers have used other materials for the back, the ribs and the neck.

[Andreas Preuss]

6 hours ago, Don Noon said:

After all, we have some pretty detailed acoustic studies of very good instruments, mode shapes and frequencies and all that... which hasn't resulted in any great practical advances that I'm aware of.

Yep.

Wouldnt this be reason enough to change research in more practical user-defined direction?

 

[martin swan]

29 minutes ago, Andreas Preuss said:

Yep.

Wouldnt this be reason enough to change research in more practical user-defined direction?

 

Sorry to sound like a broken record but what do you feel are the problems that such research might solve?

[Andreas Preuss]

4 hours ago, martin swan said:

Sorry to sound like a broken record but what do you feel are the problems that such research might solve?

Most research describes how it is and which role it plays in the whole but doesn’t answer the question how to build certain sound features.

Research on what constructional alteration creates which alteration for the sound would be much more useful. 1)
 

There seems to be no constructional alteration which is able to change one specific feature in the overall sound spectrum, so there are always positive effects bound with negative effects.  This in mind you could develop a better building strategy to trigger at the right moment the right things in a more controlled way. 
 

At the side this kind of research might open windows to different construction methods which don’t affect the classical forms too much. For example (IMO) research on the rib structure as sound determining component is largely neglected. There are IMO other aspects.

————————————————

1) from the top of my head only few acoustic research papers deal with this question. One must include Carleen Hutchins tap tone theory, one book written by Konrad Leonhard in collaboration with the physicist Müller at the Mittenwald school, the Gruyère violin experiment by Hutchins, a book by Prof. Güth and a research paper by Jansson in Sweden about arching and overtones. There are probably some more which are not on my radar.

[GeorgeH]

6 hours ago, martin swan said:

Sorry to sound like a broken record but what do you feel are the problems that such research might solve?

This is a great question. In my opinion, the problem of how to build a good violin has largely been solved, and has been for hundreds of years, though the definition of a "good violin" is highly subjective and individual, and does not appear to be objectively measurable using comparative sound spectra.

[Andrew tkinson]

10 hours ago, Andrew tkinson said:

Hello Don. What I meant was the idea of experimenting with a very homogenous material to produce a series of differently arched tops or same top with differently spaced f holes for example. Then fixing these, each in turn, to a single 'demountable' body test unit, starting with the flattest and working upwards and performing  a set of tests on each one.

I was thinking that if a very even, homogenous material was used then the changes detected would be attributable only to the changes in the arching not the variations introduced the material or in workmanship

I then thought, even though the plastic or whatever would not behave in such a complex way as real wood that any general trends if observed could be considered as moving in the same direction with a real spruce soundboard. I was thinking that it would remove the chance of anyone objecting to the observed results by saying the variations in the materials and in the test pieces used may have muddled any test results?

What I was envisaging is a bit like when an economist or scientist simplifies a system or looks at one part of a process and tests what happens when only one thing is varied.

I envisaged plastic would be a good choice as it is homogenous and it could be used to produce a lot of identical test pieces quite quickly. If a college has a CNC machine and acquired some 3d violin modelling software, if such a thing exists, a series of internal soundboard moulds with gradually rising arching could be machined from MDF. Then after the moulds have been prepared a suitable thermoplastic could be vacuum formed over the mould to produce a series of test soundboards which changed incrementally but were materially identical. Once the moulds have been made, experimental soundboards could be made easily so a lot of tests could be done each varying only one thing. Maybe a thermoplastic can be obtained which has a sort of grain like wood which isn't disturbed by the heating and forming process? Also, for arching tests the greater stretching and tensions caused by moulding higher archings would reduce the homogeneity of the plastic and would have to be considered and avoided if possible.

I was reasoning perhaps optimistically that the general trends observed when varying only one aspect of a plastic soundboard would maybe move in a similar direction with real wooden soundboards and maybe such information may help some people when trying to come up with their own slight variation on a violin design

Apologies for the long explanation of my thoughts. I sometimes get the feeling that the more I try to explain the less clear I am being!

I fully realise though, that well trained and highly skilled instrument makers seem to manage quite well using their experience and hard won skills and have no need for more pages of graphs and lists of frequencies etc.  Another slight worry I have is the risk of incurring the wrath of Greta Thunberg for seeming to desire to increase the amount of plastic waste. So if this experiment is actually worth doing I would hope that the plastic used must all be recycleable?

I must admit that being a simple man I am more excited by the idea of making and experimenting with my own luthier's "schnitzbank" !

When I went to bed last night I realised that my idea of making many quick and easy variations in plastic to test the effects of, for example, raising the arching from flat to high and then testing them would not be as easy as just using a computer and CNC to make a range of sound board moulds.

I forgot about the bass bar which would have to be fitted to each plastic test piece. This would take time, making doing many many experiments I proposed much more tedious to accomplish. The workmanship and material of a bass bar would also would add variables back  to my supposedly materially identical test pieces!

Also, although I was not proposing using plastic in 'real' violins, just for lots experiments testing only one variable and observing trends/changes, I think now that its homogeneity may mean that any trends/results that were observed would, as Don stated above, be too far removed to be applied to Spruce. I wonder if a ribbed thermoplastic would be a bit nearer to wood structure to make any test observations more realistic?

Ah well, at least I tried! Now back to the schnitsbank!

[Don Noon]

1 minute ago, Andrew tkinson said:

I wonder if a ribbed thermoplastic would be a bit nearer to wood structure to make any test observations more realistic?

I see two major problems remaining:

1)  Acoustic properties of arching curves are strongly dependent on the speed of sound of the material.  Thermoplastics are very low for speed of sound.  Making a ribbed thermoplastic would not help this aspect.

2)  Damping.  Plastics are very high damping, and I think this variable is hugely important in the tonal balance and the judgement of the listener or player.

[Andrew tkinson]

1 hour ago, GeorgeH said:

This is a great question. In my opinion, the problem of how to build a good violin has largely been solved, and has been for hundreds of years, though the definition of a "good violin" is highly subjective and individual, and does not appear to be objectively measurable using comparative sound spectra.

Hello Martin and George.

When the climber George Mallory was asked why he wanted to climb Mount Everest he answered

 "Because it is there"