Jump to content
Maestronet Forums

Scientific investigations of Stradivarius violins--an updated review article


Recommended Posts

  • Replies 564
  • Created
  • Last Reply

Top Posters In This Topic

Bruce Tai, I don't understand some of your arguments.  You have rightfully pointed out that auditory memory is short, and you claim that players or listeners therefore lack the ability to judge or discriminate between instruments in Fritz's work.  But yet, they clearly did discriminate, and the successfully distinguished between old and new instruments.  (They got "new" versus "old" backwards, but they did discriminate successfully.)

Moreover, you point to anecdotal evidence of people choosing instruments after very short playing periods.  For example, Perlman is supposed to have chosen his violin after three notes.  I had the same experience with my violin.  Although I would have a difficult time making sound adjustments on a violin because of the short-term sound memory problem, I'm quite certain that both Perlman and I were comparing violins against memories of other violins.

So you have used the short-term memory problem to argue against Fritz's results.  How can you then claim that they have the memory to be so sure that old Cremonese instruments are so superior?  Conversely, if you argue that they have memory to judge the quality of old instruments, surely they must have the memory to judge new instruments as well.

Link to post
Share on other sites
48 minutes ago, La Folia said:

Bruce Tai, I don't understand some of your arguments.  You have rightfully pointed out that auditory memory is short, and you claim that players or listeners therefore lack the ability to judge or discriminate between instruments in Fritz's work.  But yet, they clearly did discriminate, and the successfully distinguished between old and new instruments.  (They got "new" versus "old" backwards, but they did discriminate successfully.)

Moreover, you point to anecdotal evidence of people choosing instruments after very short playing periods.  For example, Perlman is supposed to have chosen his violin after three notes.  I had the same experience with my violin.  Although I would have a difficult time making sound adjustments on a violin because of the short-term sound memory problem, I'm quite certain that both Perlman and I were comparing violins against memories of other violins.

So you have used the short-term memory problem to argue against Fritz's results.  How can you then claim that they have the memory to be so sure that old Cremonese instruments are so superior?  Conversely, if you argue that they have memory to judge the quality of old instruments, surely they must have the memory to judge new instruments as well.

It's like wine tasting competition, there is always some winners. The next week you hold the same blind test, the results could be very different. That means the results are actually quite random or could be affected by unexpected, uncontrolled factors like the temperature or the music being played. 

In the tests by Fritz el al., the judges show some preference for some violins. That's not to say it's not random or affected by unexpected factors. What we want to test is their ability to judge on projection and timbre. But there is no positive control in the experiment. 

Human memory is a lot more complicated than we think. As shown in the wind tasting experiment, adding red tasteless colorant to white wine will totally mess up the taster's response. When seeing red wine, the wine taster's brain will likely search the "red wine" memory bank to see if it is merlot or  pinot noir. If it is Riesling dyed red, the brain will be totally confused, at least initially. 

Once I ordered a chocolate ice cream and and they gave me a coffee ice cream, I thought to myself this chocolate is weird and special. It does take 30s or so to realize the store person gave me the wrong flavor. So "what you think you got" vs. "you are clueless in a blind test" will alter the way your brain searches into the long-term memory databank. If our timbre memory lasts only 15 s, the brain may not have enough time to do a good search in a blind test. 

Before Perlman plays the Menuhin Soil Strad, his brain was already searching the Strad sound database and the Menuhin playing sound database, putting such memory into the cache memory. Once the signal comes in, comparisons can be made on the fly. That's only an hypothesis. We simply do not know enough about how auditory memory works. I am not a memory expert. But I was in a lab that had a project on human memory studies by putting electrodes into epileptic patient brains to record neuronal response. We don't know how memory works even if we can put some electrodes in there. 

I never doubted the integrity of the data published by Fritz et al. There are many different ways to interpret that data, though, especially because we don't know how auditory memory works in such complex situations.

It's like all of a sudden a Tesla cannot identify a huge truck in a blizzard on the highway and crash into it. It's really hard to know why without looking into the code and hardware design. 

Link to post
Share on other sites

I don't why my thread has initiated some personal attacks against David Burgess. I may not know much about modern violin makers and I apologize for my ignorance. So correct me if I am wrong to think that David is one of the most respected violin makers in the world. He belongs to a small group of elite makers who, as history will prove, have a chance to challenge the supremacy of Stradivari and Guarneri del Gesu. How they plan to get there and climb the Everest is their freedom and trade secret. If my research can help them in anyway (either by learning from my data or by realizing my follies and picking a better road), I would feel gratified. I don't know many dealer friends but the cello dealer I know best told me that David Burgess instruments are not only nice but also built to last. These are his words but not mine. While I have never seen one myself, I don't doubt that David has put in good knowledge and good research to ensure the durability. Cremonese masters have also put in good knowledge and good research to ensure the durability, as my data suggest. 

Link to post
Share on other sites
1 hour ago, Bruce Tai said:

I don't why my thread has initiated some personal attacks against David Burgess.  

I'm guilty too, Sorry David! I even used Jeffery to get a bigger punch :(
 

Some years ago, a thread started to go the same way for me. David actually stepped in to help.

Peronal attacks are very childish behaviour, I will do better..

Link to post
Share on other sites
24 minutes ago, Peter K-G said:

I'm guilty too, Sorry David! I even used Jeffery to get a bigger punch :(
 

Some years ago, a thread started to go the same way for me. David actually stepped in to help.

Peronal attacks are very childish behaviour, I will do better..

Peter, rest assured that regardless of whatever VicM may try to claim, I do not hate Finnish people. ;)

Link to post
Share on other sites
12 hours ago, Peter K-G said:

Chemical treatment is an interesting thought and quite possible too. (considering all discussions and publications over the years, where there is smoke there is usually fire)

It would be a mistake to repeat the same like

-> Varnish is the secret to the sound

-> Chemical treatment is the secret to sound

The real secret is, making a great violin is the secret to sound.

 

 

Like.

 

I think I see the problem with some of this thread's discussion.  I think GeorgeH. is trying to apply methods used in industrial processing, mostly for repeatability and traceability vs science. Science, and good science, can certainly be done in areas where one hasn't completely characterized everything in a system, or even understood most of it: cf, the discovery of tRNA in cell-free systems; the RNA tie club (somewhat analogous to Overlin Violin workshop?); the ability of Berg and Adler to discern much about the bacterial flagellar motor, before understanding anything about its exact mechanisms, via "simple" physical experiments (tethering the flagellum to the cover slip, and watching the bacteria spin through restricted media), etc.  In the molecular biology world, examples abound.  

Conversely, in attempting to explain the scientific method, particularly regarding trying to characterize unknown, not "forensically" pristine samples, we do have some hand-waving and other non-scientific statements regarding the whys and wheretofors (Strads sound better, Cremonese instruments are different from other instruments, etc.). Sound quality and preference is subjective.  On a slightly related note, some of the best work on discerning preference for general music quality might be had from Revel Lab's (Harman International) (Floyd O'Toole early work; now, Sean Olive and his co-workers) studies on headphone and IEM preference amongst widely varying sample (and characterized) populations.  https://www.researchgate.net/publication/332210798_A_Statistical_Model_that_Predicts_Listeners'_Preference_Ratings_of_Around-Ear_and_On-Ear_Headphones is a recent summary of some of this, and esp. references therein. They had both trained and untrained populations from different groups of people; training was done at Revel and on line. Only selected people who passed their test were included in the trained population.  Summary - all populations had similar broad preferences for  "good" and "bad;" the audiophiles differed from untrained listeners only in slight preference for some music samples in the intermediate levels, and it was a relatively slight difference. 

I digress a bit. The relevance is that there are broad, similar sound preferences, if I may slightly over generalize, (yes, this was for headphones, but they used certain, broad music samples. We could do the same for violins, but this work would neeed to be funded, and I can't see violin makers coming together to do this. It has been 10 -20 years' of work or more at Revel (Harmon International), with Sean Olive and co-workers). 

1) We are never, in this lifetime, going to be able to characterize that preference re violins to a degree that GeorgeH thinks is necessary before we apply the scientific method to his satisfaction. 

2) We can and should continue investigation into wood charcteristics using state of the art equipment. We may find some wood charcteristic that explains the notion of Cremonese sound, or old instrument sound, or some form of wood treatment, accidental (preservative) vs purposeful, etc.  We will never be able to do statistically valid experiments involving destructive testing, because of private ownership, and moral principles, amongst other factors.  We may not need statistically valid experiemnts.  The exception can also prove the rule...

3) The language of music preference, subjective reviews, etc. has progressed in the last 20 years or so (see The Absolute Sound, Stereophile, etc.) , and we should not stop trying to define that vis a vis violins.  To wait until we have broad consensus on what is good vs not so good violin preference should not hinder investigation into possible methods of older violin construction.  

4) We should be careful about applying overly broad statements or analogies ("wet towels") about Cremonese sound in attempting to justify research on specific sub categories of violins.

Link to post
Share on other sites
4 hours ago, La Folia said:

Your remark is shockingly unprofessional.  Bruce Tai has shown you a small sample of his data, and only that.  To think that you could judge his work from about 1 square inch of the paper is astonishing.

It's a shame that we can't see the paper yet, and I frankly have no idea what it shows.  But neither do you.

As I recall, you have used the word "pseudoscientific" very liberally when describing anything Bruce Tai has ever written.  He is a well-qualified researcher.  Whether or not he is correct, he has done a lot of work on this difficult subject.  As far as I know, you have published no knowledge at all on this subject.

Informed scientific opinions sometimes differ, but there are rules and careful forums for those opinions.  You claim that Tai's published papers are all nonsense.  Since you are disputing statistics, you have ample opportunity to publish a valid response in the same journals that published his papers.

It's telling that you claim to have published "numerous" scientific papers, but apparently have never published a response to one of Tai's papers.  That's how it works.  I'm sure you will supply the references if you actually have.  Valid scientific criticism does not come from internet trolling.

Like.

Link to post
Share on other sites
5 hours ago, La Folia said:

Your remark is shockingly unprofessional.  Bruce Tai has shown you a small sample of his data, and only that.  To think that you could judge his work from about 1 square inch of the paper is astonishing.

It's a shame that we can't see the paper yet, and I frankly have no idea what it shows.  But neither do you.

Who has seen the paper? Let's see...according to Tai:

"For the last two years it has been rejected by Nature, Science, Nature Communications, Science Advances, Nature Materials and PNAS, and we only got one peer review which was pretty disturbing.

His last PNAS paper was riddled with unsupportable claims, inadequate data, flawed methodology, and bogus statistical analyses. Hopefully, PNAS won't make that mistake again

And one only has to see the "small sample of his data" to know that the conclusions he is making from it are absolute nonsense.

By the way, when I wrote "This article is pure pseudoscientific nonsense" I was referring to this article, which is pure pseudoscientific nonsense. 

Link to post
Share on other sites
1 hour ago, l33tplaya said:

I think GeorgeH. is trying to apply methods used in industrial processing, mostly for repeatability and traceability vs science.

No, just basic science and sound statistical analysis.

As Carl Sagan said: "Extraordinary claims require extraordinary evidence."

Tai is full of the former with absolutely none of the later.

In fact, in a typical pseudoscientific fashion, he makes claims of phenomena that have not even been proven to exist, and then suggests that only he can prove what causes them.

And this:

16 hours ago, Bruce Tai said:

So we know what Cremonese masters did not use but not what they really used. That creates a huge problem for violin makers. I am sorry about this. But we are getting closer to the forgotten truths about Cremonese violin making. What happened happened, no matter how it was kept so secretive. 

You see, nobody has ever made a violin as good as the old Cremonese makers in 400 years, and Tai is going to now tell all you violin makers why that is with his flawed experiments, bogus statistics, and extraordinary claims. He and he alone knows the secrets.

And he claims this is "a huge problem for violin makers." All your work was for naught.

But he is sorry about it. 

Nobody should be swallowing this self-aggrandizing nonsense.

Link to post
Share on other sites

I find inquiries into the way the famous old violins were made to be interesting, regardless of whether or not these instruments are proven to be superior, or whether or not I will ever utilize these techniques in my own making.

Link to post
Share on other sites
1 hour ago, David Burgess said:

I find inquiries into the way the famous old violins were made to be interesting, regardless of whether or not these instruments are proven to be superior, or whether or not I will ever utilize these techniques in my own making.

I was wondering the same thing yesterday about your museum example.  Did he trace around an existing plate in the shop or did he use a plan?  Then I thought that is still pretty good for early 1980's regardless.

Link to post
Share on other sites

GeorgeH admits that he can't meet the golden standard of 6-sigma statistics (used by many global corporations and experimental physicisyts) in his own research. According to his own obsessive-compulsive standards, he was doing pseudoscience throughout his career with underpowered statistics.

GeorgeH does not even know how many violin manuscripts I am preparing and he is confusing them altogether, making blanket criticisms on all of them, none of which he has even read. 

GeorgeH is calling Philip Ball's article in Nature Materials "pseudoscientific nonsense." Philip Ball is a great science writer with many inetresting books on Amazon. Ball's breadth of knowledge in science is astonishing. Philip Ball interviewed several experts on violin research before writing the article. Ball was well informed and well prepared, a true professional. 

Ball's book Bright Earth is very useful for varnish makers trying to understand pigments and colors. I read it for my varnish researech: 

https://www.amazon.com/Bright-Earth-Art-Invention-Color/dp/0226036286/ref=sr_1_1?dchild=1&keywords=bright+earth+philip+ball&qid=1620042210&s=digital-text&sr=1-1

I will stop responding to GeorgeH in this thread. He apparenttly has no insight to offer about fine violins.  

Link to post
Share on other sites
1 hour ago, David Burgess said:

I find inquiries into the way the famous old violins were made to be interesting, regardless of whether or not these instruments are proven to be superior, or whether or not I will ever utilize these techniques in my own making.

There is always a curiosity to know what kind of instrument is being used by Paganini, Sarasate, Heifetz, Milstein, and Yo-yo Ma. What is it made of? 

I never advise violin makers on how to make instruments. In my Strad article in 2017, I specifically warned makers about wood treatment experiments. It can do more harm than good if not properly controlled. I never advise makers on how to make varnishes. Analytical chemistry can detect the remaining traces of many ingredients. That's very different from knowing the original formulation and process. 

Link to post
Share on other sites
43 minutes ago, Bruce Tai said:

GeorgeH admits that he can't meet the golden standard of 6-sigma statistics (used by many global corporations and experimental physicisyts) in his own research. According to his own obsessive-compulsive standards, he was doing pseudoscience throughout his career with underpowered statistics.

I never claimed that you needed to reach six-sigma certainty. Never. You're simply lying and blowing smoke. 

43 minutes ago, Bruce Tai said:

GeorgeH does not even know how many violin manuscripts I am preparing and he is confusing them altogether, making blanket criticisms on all of them, none of which he has even read. 

Your last PNAS paper was full of grandiose claims, faulty experiments, and bogus statistics. I covered all that in a previous thread.  I am not confusing anything. You said "For the last two years it has been rejected by Nature, Science, Nature Communications, Science Advances, Nature Materials and PNAS, and we only got one peer review which was pretty disturbing."

Those are your words, not mine. And I don't find that disturbing; I find it hopeful.

43 minutes ago, Bruce Tai said:

GeorgeH is calling Philip Ball's article in Nature Materials "pseudoscientific nonsense." Philip Ball is a great science writer with many inetresting books on Amazon. Ball's breadth of knowledge in science is astonishing. Philip Ball interviewed several experts on violin research before writing the article. Ball was well informed and well prepared, a true professional. 

I don't really care about who the author is; I care about the content. And the content is pure pseudoscientific nonsense. There is no known measurable phenomena of "awakening violins," but you think "that stress-induced deformations will redistribute water molecules in the wood. 'Although precise measurements are difficult, it is not far-fetched to attribute the awakening of old instruments to these factors.”

Using scientific tests to try to prove a cause of a phenomena hasn't even been shown to exist is pure pseudoscience. 

Link to post
Share on other sites

Science = Knowledge

Data => Information => Knowledge(Science) => Wisdom

Data = rubbish

Information = Structured rubbish (Data)

Knowledge(Science) = Applied Information

Wisdom = Silence

 

Link to post
Share on other sites
9 hours ago, Bruce Tai said:

Guarneri del Gesu was pretty desperate when he soaked his wood into something like 5% alum solution to get a lot of aluminum ion to penetrate his spruce and maple. 

Even if this is true, so what?  The implication is that a god of violinmaking did something, then it is meaningful.  If you can show that aluminum ions do something significant to the relevant acoustic properties of the wood, great.  Until then, I would assume it is just a small amount of dead weight, not enough to be significant.  I think it helps to keep in mind that bound water in the wood can change the weight of the wood by far more than any of your trace elements, depending on humidity.

Please prove me wrong, and show that any of these trace elements actually DO something, instead of speculation that there was some special knowledge and intent, now lost.

 

 

Link to post
Share on other sites
46 minutes ago, GeorgeH said:

 There is no known measurable phenomena of "awakening violins," 

  That may of been me saying the awakening bit, not Bruce Tai.

  Though one will need a cause afterwards one can get a copy of caprice no. 21 {20) 6th line and a good player who can play the music.

  Let him/her play whatever music they want for awhile, put the freshly played fiddle on a rack for testing and take note of what's happening.

  Next, have the player run through that section of the caprice ten or so times.  Now retest.

If it's not called awakening it's gotta be called something.

Link to post
Share on other sites
2 hours ago, Bruce Tai said:

GeorgeH admits that he can't meet the golden standard of 6-sigma statistics (used by many global corporations and experimental physicisyts) in his own research. According to his own obsessive-compulsive standards, he was doing pseudoscience throughout his career with underpowered statistics.

 

Reading between the lines of what he said earlier, his XRD experience probably extends to putting prepared powder samples in a Rigaku and feeding the digital output to Jade to read the peaks for him.  Six-sigma concerns would never have entered into it.  He's simply being unprofessionally beastly.  Ignoring his outbursts is best.   :)

I did find the Sagan quote amusing in this context, but then, I know the backstory on it.  :lol:  Overenthusiastic skepticism can be a serious barrier to valuable research, as well as a smokescreen for a hidden agenda. 

 

7 hours ago, Bruce Tai said:

I don't why my thread has initiated some personal attacks against David Burgess.

It's one of the diverting traditional pastimes on MN.  Any excuse will do, and David gives as good as he gets.   You didn't start it, so don't be concerned.  :lol:

Link to post
Share on other sites
11 hours ago, Bruce Tai said:

They had other goals in mind. And they were willing to experiment for a century to meet those goals. Guarneri del Gesu was pretty desperate when he soaked his wood into something like 5% alum solution to get a lot of aluminum ion to penetrate his spruce and maple. That's not  just for anti-fungi. The Guarneri family business was doing poorly compared to Stradivari. I guess del Gesu was desperate for a competitive edge. 

I do admit that I have no experience in building mind-reading time-travel machines that Dr. Tai apparently has.

Link to post
Share on other sites

I think we could all use some real-work exercise at this point in time (after a long winter compounded by pandemic stress, etc.). Consider taking a nice long walk, maybe engaging in some ditch-digging or a little goat wrestling, etc...

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    No registered users viewing this page.




×
×
  • Create New...