Can an expert panel of violinists pick the old Italian Violins from modern instruments?

January 3, 2012 at 02:28 AM · http://www.npr.org/blogs/deceptivecadence/2012/01/02/144482863/double-blind-violin-test-can-you-pick-the-strad

Replies (72)

January 3, 2012 at 04:00 AM · I love the first violin. It has the freshness of a young person; attractive and personable. I would like to take it home.

The second violin sounds tired, but still retaining its charm and glory. It is something to be admired and respected, but not to own.

January 3, 2012 at 05:20 AM · The second violin sounds more tight to me, I'll pick the first one~

January 3, 2012 at 05:47 AM · After seeing the result, I am not really surprised. I heard three strads in concert last year and I wasn't impressed by any of them. Perhaps all the fuzz has built too high of an expectation. Plus there are so many out there it's hard to say which ones are the best. There is also a way to play it aka. not to press too hard on the strings.

If I could choose, I would choose the first one, brighter sound. second one sounded strangled or having difficult produce sound. but if I were an investor, I would probably go with 2nd one? ha ha

January 3, 2012 at 02:11 PM · Sorry but I'm going to reveal who's playing the violin! Though it has nothing to do with the blind test. ;-)

Violinist revealed...

January 3, 2012 at 02:31 PM · Using pretty decent earbuds to listen to my computer the 2nd violin seemed to me to be the Strad. It seemed stronger in the mid-range and had those characteristic overtones. So my guess was right.

Andy

January 3, 2012 at 03:10 PM · Number one seemed to have a youthful freshness and vigor. The second struck me as having a lush maturity. I guessed number two, and yes, that was before I had scrolled down for the answer - honest !

January 3, 2012 at 03:50 PM · I wonder if this thread will reach 100 responses.

It was so obvious to me when I first listened! I knew within five seconds of the first excerpt. I was so certain that I was nervous I was wrong. The second violin seemed like it filled the space; the first one's sound was much narrower and more directional to me.

I read one striking comment about this story, whether on NPR or Cozio or somewhere - that in order to *really* be able to judge instruments, we'd need to do it over a long period of time, in a variety of halls, in a variety of repertoire, in a variety of weather conditions, probably with a variety of strings and bows and players. I'm leaning toward that viewpoint, too; it's just too short of an experiment to draw concrete conclusions from. A modern violin may very well win such an challenge. (I'd like that.) But we have insufficient data now.

Who wants to bet that even if it could be scientifically proven over a series of long-term experiments that to ninety-nine percent of the population modern fiddles are just as preferable if not more so than Strads...Strads will still retain all their mystique, allure, and value?

January 3, 2012 at 04:54 PM · this is the same old result of all these types of studies, whether blind, double blind, or seeing-eye dog. Would be same even if all strads. One thing that's proved over and over again is how powerful the strad name is to put otherwise non-newsworthy items in public eye.

January 3, 2012 at 05:07 PM · One issue you might have is the strings used. How were the violins strung? All with the same strings? All with the strings that best bring out their characteristics?

I am not surprised that at least some folks were able to tell the difference. I once read Perlman to say that when he is in a room at an auction where violinists are trying out expensive violins, he can always tell clearly when someone is trying out a Strad.

January 3, 2012 at 05:31 PM · I heard the story on NPR yesterday and found that on their website. Luckily, I listened to the links before scrolling down to the answer on the bottom.

There wasn't a huge difference I could hear (on my laptop) in tone quality. The difference I did pick up is that one of them had a measure of flexibility that the other didn't have. Could have been setup, of course, but it was enough to suggest that it was the older one. And as it happens, I was correct.

January 3, 2012 at 06:15 PM · Before listening I'd like to say that I'm a big believer in modern as well as classic instruments. I have about a dozen violins and only one of them is old. A number of my contemporary violins are very competitive with good classics - in quality as well as sonority. So I won't presume to be sure which is which, but will be glad to say which one I like, and why. I won't be surprized if the one I like better is the Strad, and I won't be surprized if the one I like better is the modern...

OK, just listened. First of all, I'd like to compliment the violinist. I really like the vibrato, phrasing, etc., even though on my computer and earphones at least, the sound was too soft and distant. But anyway, I like violin #1 better. It had more focus, presence, and sparkle - and the violinist somehow sounded more comfortable and free with it. That said, had another violinist played the same excerpt on both, I might have preferred violin 2. No violin plays itself. The Tchaik. is a very good choice if you're going to play just one excerpt - the opening of the Beethoven, too. it goes up and down all the strings. I wish this excerpt had continure a bit just to the F#, where the orchestra come in.

How do we find out which is which, w.o. spoiling it for anyone who doesn't want to know yet?

January 3, 2012 at 06:28 PM · I was expecting this test to be difficult, but it only confirmed my opinion: you can't capture the sound of an old violin in a modern violin. It was so obvious. And I prefer the second (which is unfortunate for my wallet...).

January 3, 2012 at 06:29 PM · On the first clip, it sounds as though the violin was closer to the mic and/or there is more sound from the top of the violin versus the side.

A slight angle or distance change can certainly change the recorded tone, this would not affect the results in the room but will in listening to the clips after as we are doing.

January 3, 2012 at 06:46 PM · That was fun. Thanks for posting! :D

I guessed correctly. And I preferred the first violin. To me, it sounded fuller.

The second sounded 'thin'.

January 3, 2012 at 07:36 PM · Yikes! Where are the audiophiles amongst us? Digitized MP3 files are lossy, even uncompressed ones-- when compared to analog (high quality LP or tape). The differences between violins would be more evident if the resolution of the recordings was greater. Violinists can easily detect tonal differences between violins when played live, although preferences may vary.

January 3, 2012 at 07:53 PM · I doubt that most violinists, or even, more particularly, makers, have the foggiest clue as to what is the "best" type of violin sound.

Players know what they like under the ear.

Makers know what sells, but can't guarantee it.

If a maker says they can guarantee a certain sound, they are either self-deluded or lying. I would love to be proven wrong in this last judgement, but I can't afford to commission ten fiddles from each of ten makers to do the experiment with a hundred fiddles.

I recently met a maker who said that he had spent a great deal of time trying to get rid of the very frequency range that we need for good articulation.

What am I supposed to make of that?

What I made of it was that the players who had given him feedback on his instruments wanted "smooth" "mellow" tone to please themselves under the ear, without a thought for the poor listener who needs to hear the clarity of the phrasing.

Most dealers aren't that interested in sound. There are enough people with different tastes to ensure that almost all fiddles will sell eventually, at the right price.

You have to spend plenty of time playing hundreds of violins, and learn how to listen to what is reflected by the room or hall, to understand how a violin works as a performing instrument. Very few players, makers or dealers do this, either by limited time, or limited inclination.

January 3, 2012 at 08:01 PM · I listened to both clips once and straight away I thought the second was the Strad. I then listened to them again, and concluded for myself that I was sure the second was. AND I preferred the second... More depth behind the sound for my ears. :-)

January 3, 2012 at 08:37 PM · Such test are nothing new, nor is the outcome, but this one is a little unique among those which have been published.

If I understand correctly from having read several articles, this one drew its conclusions not from an audience, but from the players themselves, 21 one of them in all.

My guess is that it was done this way to address a couple of common criticisms of other tests where the megabuck old instruments also didn't prevail: That the listeners in the audience who evaluated the tone weren't educated enough; or that it isn't important whether the audience can tell the difference or not, because the advantage of a fine old violin is in what it does for the player. This experiment certainly calls these excuses into question, because it was the players themselves who couldn't identify whether a fiddle was new or old, and ultimately preferred a modern fiddle over a Strad.

There have been other such experiments where the players perspective was weighed, but this is the first I'm aware of which has received wide publicity.

January 3, 2012 at 09:35 PM · I did not hear the violins in this test because when I went to the website it caused my computer th freeze up, so I have avoided it.

But at the risk of being unpopular I would think there is a strong chance for the listener to be deluded, and I can't ever take seriously any test which we do not witness live. Recorded sound will never really tell us anything much. You can make no judgement as the the carrying power of the instruments, and as someone that has seriously recorded very fine players - I know how much gain twidling and mic distance can have, not to mention graphic equalisers and any equalisation or processing.

I would agree with the comments Graham Clark made about players in particular not often having good judgement, and many, particularly less experienced players, prefer a nice sound under the ear. This usually translates into a thin or feeble sound a little distance away where the listener may be positioned. These "nice under the ear instruments" usually fail to cut through an orchestra with concerto playing, and are not strong enough in chamber music, especially for the first violin part, to be heard properly.

So there is in my opinion no point in listening over the Internet to dubious recordings with poor resolution and sound quality. It really is just a waste of time.

January 3, 2012 at 09:49 PM · Since it was held in Indy, I wonder if it was done in Sept during the competition? My understanding was that there was one listening session with players, and with the results from that session they took the top two "old" instruments and top two "contemporary" instruments (from the invited makers) and played those for the audience while the judges were going over the final player judging results.

January 3, 2012 at 09:53 PM · Here's another article about the same blind test:

http://blogs.discovermagazine.com/notrocketscience/2012/01/02/violinists-can%E2%80%99t-tell-the-difference-between-stradivarius-violins-and-new-ones/

One of the participating musicians have posted two clarifying comments in the discussion that follows below the text.

Btw, listening with studio headphones I had absolutely no doubt as to which violin was the strad. It had the typical "cardboard" sound of old violins, while the new one had a more solid sound, to the extent that I found it a bit strenous in the upper mid range. I personally preferred the strad.

January 3, 2012 at 09:58 PM · I see it another way. You can't always blame the internet or the equipment.

I have some professional experience in audio, I recorded more than 20 albums, from the "Winterreise" to orchestra. Obviously bad equipment limits the audible information, but even when I listen to Oistrach thru the "speakers" of my cell phone the magic is instantly there. You see the beauty of Ingrid Bergmann no matter if it's on the movie screen or on a fax printout.

It should not be ignored that whe "hear" many things that are not there in reality; for example the frequencies below 300Hz in a phone call or the fundamentals of the first few notes of any violin. The brain replaces a lot of information that really is not there. It makes only few mistakes, for example it never replaces missing frequencies in a way that a male voice is recieved as a females's. That's why we can judge a relevant part of the tonal quality of a fiddle even with MP3s. A decent MP3 file has everything the ear needs. And the questioned recordings are sufficient for comparing.

January 3, 2012 at 10:03 PM · OK - so it is known that the 2nd one is the Strad. And I still prefer the 1st in that short segment with that particular player. Anybody know who the player is and which Strad and which modern violin was used?

I was also going to make the point that what might work one way in a studio recording might work another way live in a hall. Good Strads are known to carry beautifully in a hall and work with it, so to speak. In the excellent DVD, "Homage" in which James Ehnnes plays extensively on the violins of David Fulton, Ehnnes made a particularly interesting comment in comparing Strads and del Gesus: he said that with a del Gesu, what you hear under your ear is pretty much what you'll get, but that a Strad plays the hall, and that in close quarters he might prefer a del Gesu, whereas in a great hall he'd probably prefer a good Strad. Even one small space can affect a violin differently than another - and not just in terms of more live or more dry: specific notes may be more or less enhanced. I just experienced that the other day.

At first I somehow expected that the test here would involve several instruments. With only two, the testing should have been more extensive: the Tchaik. opening, followed by characteristic passages on each respective string separately. Then maybe passages with chords, etc.

Does anybody have the DVD set, The Heifetz Concert Series? It includes a recital by Erick Friedman. He plays the whole recital on what he later reveals to be a Joseph Curtain. The program includes the slow mvt. of the Goldmark concerto, which he plays beautifully. Then for an encore he does an unusual thing: he plays the Goldmark again - only on a Strad that a friend lent him. There I prefer the Strad , but not by a huge margin. The Curtain is quite good, with clarity, focus and a ringing tone. The Strad is warmer with a broader spectrum of colors and more complexity. It's not easy to switch back and forth as the encore Goldmark is on the same track with the Rondo Capricciosso.

January 3, 2012 at 10:29 PM · Some interesting quotes from the various articles:

"Kai-Thomas Roth, secretary of the British Violin Making Association, said that double blind tests, where neither experimenter nor musician knows which violin is played, had already shown people cannot distinguish a modern violin from a priceless work of art.

There's some myth-making that helps old instruments, Thomas said. If you give someone a Stradivari and it doesn't work for them, they'll blame themselves and work hard at it until it works.

Give them a modern violin, and they'll dismiss the instrument straight away if it doesn't work for them. That's the psychology at work. Modern violins are easily as good, but even a good maker can make an instrument that doesn't work out."

Another article:

"Only 8 of the 21 subjects chose an old violin as the one they’d like to take home."

And another:

"In the old-to-new comparison, a Stradivarius came in last and a new violin as the most preferred."

Yet another, or do some of these overlap, and I've lost track? LOL

"It puts cold water on some old myths and should certainly be good news to young musicians who yearn for violins that they will never afford."

January 3, 2012 at 11:47 PM · One problem with this test is that no sensible violinist(including the ones in the study I guess and hope) would buy a violin given such a short trial of an instrument as was the case when preferences were expressed in this experiment. Normally I would hope a buyer would take at least two weeks to asses an instrument before making a purchase. Often I see a player's initial response to a fiddle change over this period. Sometimes a fiddle that initially wows later might not after a few days or one that did not wow so much at first can offer hidden depths when the player gets to know it. Also it is (as we all know) a good idea to try it in a concert hall and get someone you know to play it to you etc etc...all rather hard to do in a double blind test. That makes me a bit skeptical of this test itself although I agree that many contemporary violins are very fine musical instruments indeed..and valid works of art themselves...probably more so than at any other time since the mid 18th Century and this is good news for players as David Burgess says above.

January 4, 2012 at 12:51 AM · I too would wholeheartedly recommend that a player spend more time with an instrument before making a purchase decision. However, in my experience, at least with pretty good players, the impression after 30 seconds of playing tracks pretty well with their decision after two weeks. Maybe that's because most of them are better equipped to know how an unfamiliar instrument "wants" to be played right away?

However, I've had a few of them change their mind after six months or a year. So how long should one of these tests last, and how long should an instrument maker's or dealer's "approval" period be? A year?

January 4, 2012 at 12:54 AM · Try listening to the 2 excerpts comparing little phrases at a time, easy to do as they are placed parallel. You will find that the violinist did not use the same fingering , the 2nd sample has more glissandos and the glissandos also sound better than in the 1st sample. The 2nd violin has that velvety sound to the glissandos that good violins have. I wonder if the first violin is brand spanking new, it has that rough woody edge to it. It is hard to get those velvety glissandos on a very new fiddle, it took a while for my fiddle to get that.

January 4, 2012 at 01:18 AM · Something to keep in mind:

With those sound tracks, we're listening to a violin recorded through an unspecified microphone, and an undefined recording path. Both can color the outcome.

The players gave their live impressions, with no recording involved, so that might be the first thing to hang your hat on.

With that in mind, I preferred the first violin. It was less "fuzzy". Section players probably won't agree with my preference, which is largely based on the best solo instruments. These aren't the best thing, if the desire is to "blend" into an ensemble or orchestra.

January 4, 2012 at 01:22 AM · I think we can be pretty sure that the recordings are done professionally and with good equipment. I am guessing these are either done by Joseph or Claudia, both having good recording equipment and mics. The recordings could possibly be bineural.

There is a path behind this too, and mics does of course not give the same impression as playing the fiddles do.

January 4, 2012 at 01:31 AM · Anders, with respect, you and I both know that there is no recording or playback path which doesn't color the sound. Compensations for the anomalies of the equipment can do a lot, but not everything.

I use recordings quite a bit in assessing sound, and it's a valuable tool. When I really want to cut to the chase, we do live player and listener tests, often in an auditorium.

I really appreciate that the people who did the test provided sound files, but that pales in comparison to the judgements of those who made the live assessments.

January 4, 2012 at 01:41 AM · David Burgess wrote

I too would wholeheartedly recommend that a player spend more time with an instrument before making a purchase decision. However, in my experience, at least with pretty good players, the impression

after 30 seconds of playing tracks pretty well with their decision after two weeks. Maybe that's because most of them are better equipped to know how an unfamiliar instrument "wants" to be played right away?

However, I've had a few of them change their mind after six months or a year. So how long should one of these tests last, and how long should an instrument maker's or dealer's "approval" period be? A year? [Flag?]'....

I am not sure I totally agree with what you say David. There are some very fine players who are not comfortable with any instrument but their own and there are some who love trying everything they can and adapt fast and have a broad field of reference... If we don't know the players or their experience...what can we find from this? I can think of infinite ways to rig a double blind test with great old Cremonese that really work and moderns that don't and vis -versa ...

January 4, 2012 at 01:53 AM · Fair enough. I don't happen to think that this test was rigged, knowing some of the people involved, interfacing it with other tests, and having trivial knowledge about why various specific instruments were involved, but it's always a good question to ask.

Melvin is an uncommonly fine maker, and I value his input, if that puts any disagreements in context.

January 4, 2012 at 02:13 AM · Raphael - I don't know which violin was used, but check out my post above to know who's the violinist.

January 4, 2012 at 02:30 AM · Thanks, Casey. No wonder I liked his playing.

One thing we shouldn't lose sight of is how inportant the individuality of each instrument is. It's not just a question of old vs new or Strad vs del Gesu.

January 4, 2012 at 03:09 AM · re earlier post: some of it not so much glissando as audible shifting.

January 4, 2012 at 03:54 AM · just outta curiosity, can you really tell the price and maker just by listening to the violin, even a rough estimate?

January 4, 2012 at 04:11 AM · Here is the URL of the more detailed article in The Guardian (U.K.): http://www.guardian.co.uk/music/2012/jan/02/how-many-notes-violinist-stradivarius. Here is a brief summary of the article:

The test was done at the Indianapolis international violin competition in 2010. The article does not specify whether or not the Strad players were contestants.

All the instruments were played in a large hall.

The instruments were 3 modern violins, 2 Strads, and one Guarneri.

There were 21 musicians playing all 6 instruments used in the test. Each player had a chance to play each instrument in the test and rate them for "playability, projection, response and "tone colours." (Presumably they were given more time than the brief sound samples published by NPR).

Here is my perspective as an ex-scientist:

Did the players' responses regarding any one of the factors (playability, projection, response, tone colors) show a difference for Strads vs non-Strads?

How different were the players from each other? Did any one factor, such as number of years experience playing the violin, have a major effect on the results?

Did any of the players in the test have previous experience playing a Strad?

Did the players all play the instruments in the same order? (I don't think so.) Did the players favor the first violin they tried, the violin which they played most frequently, etc.

What were the effects of age and gender on the players' choices?

Did each player play the same passage of music on each violin? Did they all play the same passage? Some violins prefer certain kinds of music.

What about the bows used?

(I used to be a pretty tough scientific reviewer.)

Here is my reaction as a listener:

I could hear a difference between the two violins, but I was not sure which was a Strad. The first violin sounded like it gave an even sound through all the registers. The second violin sounded a bit thinner in the higher registers than in the lower ones.

Finally, I note that similar tests have been run many times, and they all give similar results.

January 4, 2012 at 06:29 AM · Player vs Listener tests. Interesting player evaluation test by Claudia Fritz and others, but NPR's blurb for listeners was totally inadequate - far too brief and for only two violins. Does this reflect prejudice against good listener tests, which may be a belief of many players? and many violinmakers? Compare this NPR blurb with the superb Landon 2010 Ole Bull listener comparisons, with remarkably consistent and excellent playing of 50 violins made by reputable international violinmakers using the same model of violin. The Landon comparisons of contemporary instruments is far superior to the old Ricci comparisons and the mid 1970's BBC comparisons by Beare, Stern and Zuckerman. I attended the double blind (rarely done)comparisons of old vs. new cellos, organized by Robert Cauer at the 1990 Amercan Cello Congress. See the article by Allan Coggins, "Blind Faith" in the The Strad Feb., 2007. I believe the Landon comparisons are even better than comparisons of 30 Strad vs. del Gesu violins in "The Miracle Makers", by Bein and Fushi. Why so little comment under the Christophe Landon discussion below regarding his remarkable listener evaluations?

January 4, 2012 at 06:30 AM · Wise man say:- player about to buy violin need reality cheque.

January 4, 2012 at 01:39 PM · Pauline - all very valid questions! BTW, over the years I've played on about 5 Strads, 2 del Gesus, a few Guadagninis and Amatis, etc. They are all different. When people talk of a Strad vs a Del Gesu sound, it is a generalization - not without meaning, but still...

Even old vs new sound. Yes, violins need time and playing to reach full maturity. But some contemporary violins have a warmth and mellowness that some might find surprizing. On the other hand, from personal experience I can say that the Kreisler del Gesu and the Molitor Strad have an edginess and, esp. in the case of the Molitor, almost a rawness that again, many would find surprizing.

January 4, 2012 at 01:49 PM · "NPR's blurb for listeners was totally inadequate - far too brief and for only two violins."

Actually Charles, I think you've pointed out one of the strengths of NPR's presentation.

While what Landon did is great, few people will take the time to wade through recordings of 50 violins. NPR presented more of a summary, with a couple of brief examples, which require little more than mild curiosity and a time budget of two minutes to explore. I don't think there's much question about which will get more traffic. If something is complex and time consuming enough, it can be largely ignored, and that can reduce its ultimate utility.

January 4, 2012 at 03:05 PM · My husband and my mom, both non-violinists, took the test and they both knew which one was the Strad. but this is after I spent the past three weeks going back and forth between a modern and an old violin, so they were kind of trained already.

January 4, 2012 at 03:14 PM · I also thought the Strad was easy to pick out...

However, that doesn't mean that I don't like the first instrument. In fact, it tickles that formant-focus quality throughout it's range which I like. It has a certain aural energy to it.

But I would really have to play them! There are some "X" factors that go on when you actually play a violin, or are in the room with it, at least :-)

January 4, 2012 at 03:38 PM · Raphael: But LISTEN to Kreisler - discs 2, and 4, of Kreisler playing the great violin concertos, and discs 6 and 7 of the EMI Box set (at nominal cost) playing the great violin sonatas - all in the mid 1930'S. Did he use the del Gesu of his name for some of these?

David: Maybe this is the reason why there had been so many responses to the four violins present by BBC? But how long does it take to make a violin? How long for a proper evaluation of its sound? How long did it take Anders Buen to evaluate the Kreisler del Gesu, compared with others? How long does it take to play the violin, reasonably well or expertly? I think it is worth the effort to listen to something great - like Landon's Ole Bull presentation, as well as the Miracle Makers.

January 4, 2012 at 04:42 PM ·

Some of the comments here are quite interesting in terms of what it takes to have a "mature" sound in the instrument. The recent development of "Stradivarius replicas" created using CT imaging technology made a lot of people question whether the the luthier's talent could be truly duplicated with modern science.

At the same time, the Fungus Violins generated even more controversy with the idea white-rot fungus could actually produce a substrate with superior tone quality by destroying specific structures in the wood. Raphael's comment about aging an instrument isn't too far off from this. The difference is that the aging is manmade instead of synthetic in these cases.

Of course, there is always the perspective that only the combination of talented construction and time can truly lead to an instrument with a true "million-dollar sound."

January 4, 2012 at 05:42 PM · Maybe I have missed something in this conversation, but has anyone mentioned sound projection? for instance some of my students' violins sound as good or better than my Amati in my studio. However, in a hall there is no comparison.For the Amati, the larger the hall, the better it sounds. Also, anecdotally during the 70's when I did a little session work in NY, some of the regular players said they preferred to play on for instance a Vuillaume, rather than an old Italian, because the Vuillame sounded better miked close up.

January 4, 2012 at 06:29 PM · Bruce, there have already been many tests involving audience perception, and the outcomes have been somewhat similar to this test, in that overall, the old expensive instruments have failed to rule the day.

This particular test was focused on whether the players themselves could tell the difference.

When some of the audience-evaluated tests came out, the reaction from many was, "OK, maybe the audience can't tell the difference, but the player can. The advantage of an expensive old violin is in what it does for the player."

Now, this too has been called into question.

January 4, 2012 at 07:29 PM · David wrote: "Anders, with respect, you and I both know that there is no recording or playback path which doesn't color the sound. Compensations for the anomalies of the equipment can do a lot, but not everything".

I think we disagree there. I am most familiar with measurement systems, units and mics. These are designed not to colour the recordings or the analyses, all the way to the analysis screen. A recording done using the same system will behave in the same manner. Many of the good recording mics are also made not to colour the sound, e.g. B&K condenser mics and the like.

There can of course be colorisation in recording systems, mics and possibly other devices, but not always, as you seem to assume. In our measurements there are never colorisation involved. If it was, they would not be done properly, and the measurement instrument would not work according to its spesifications.

Reproduction of recordings is a less standardized side, and I think colorisation there are more likely than not.

January 4, 2012 at 07:38 PM · I see lots of people feeling good about their 50/50 guess as to the real thing. If the research in this on going experiment is true (remember, Claudia has allegedly conducted many such experiments with various other instruments as well) the evidence shows that you'd likely not be any better at recognizing the "real" thing any more than the pro players could in the experiment. Saying it was so easy to recognize the Strad in the article recording is tantamount to saying "Well, if I were part of that experiment, I could easily pick the old italians every time!" The evidence seems to overwhelmingly suggest that you simply could not.

The probability of choices changes drastically when going from a choice of 2 things that are each 1 type to 6 things with the possibility of each one being one of several types. ;)

(Sorry this was my first post, but I just couldn't resist adding my useless opinion) lol!

Ryan

January 4, 2012 at 07:45 PM · Actually, there's a follow-up that needs to be done. If the ability to choose correctly is consistent with random guessing then you should try again with the people on the right end of the curve. If the second set of results by those successful listeners/players is random, then we can conclude they were guessing the first time.

Otherwise, all you've proven is that the ability to choose is normally distributed. Like height, IQ, and other traits.

January 4, 2012 at 07:49 PM · Excellent, Stephen! I wonder if they run the same people through the experiment several times to control for random guessing. I wonder what other controls they might possibly have employed to improve accuracy.

Interesting stuff, nonetheless!

Ryan

January 4, 2012 at 07:55 PM · I remember reading an Art History text and a question was posed:

Do we actually know what we like, or do we like what we know?

This debate reminds me of that...

January 4, 2012 at 08:32 PM · NPR included the sound recordings to create interest for their listeners but the recordings have nothing to do with our actual experiment.

Our results were published in the Proceedings of the National Academy of Sciences. The paper requires a fee, but the abstract is free, as well as the supplementary information which has detailed descriptions of the experiment, including background of the testers, the value of their own violins, and additional analysis of the results:

http://www.pnas.org/content/early/2012/01/02/1114999109.full.pdf+html

The experiments took place in a relatively dry (acoustically) hotel room. The room was darkened and the players wore very dark goggles. I presented the violins to the players and I also wore dark goggles so I could not identify the specific instruments by sight. (That’s the double-blind part.)

The experiments took place during the final week of the 2010 Indianapolis Competition. We cannot provide information on the relationships, if any, between the instruments in the experiment and those used in the public playing demonstration at the end of the Indianapolis Competition. We arranged the public demonstration to maximize audience participation and interest without regard to its scientific validity. See Laurie Niles' blog about the playing demo:

http://www.violinist.com/blog/laurie/20109/11672/

Laurie was one of the test subjects in our experiment and she has just commented on her experience in a new thread:

http://www.violinist.com/discussion/response.cfm?ID=21663

The most important point(the point of our study!) is that if you are truly interested in the sound and response of a violin, learn to play and evaluate them with your “eyes closed.” You will learn a lot more, and it may challenge some of your pre-conceived notions.

Fan Tao

Director of R&D

D'Addario & Company

January 4, 2012 at 09:25 PM · Charles - I'm not sure what point you're making to me, but perhaps this story is relevant:

Once a wealthy amateur was looking for the ultimate del Gesu, which he felt should play as closely as possible to Kreisler's. He thought he'd finally found it at Wurlitzer's, and took it home for a trial. Meanwhile he heard Kreisler in concert. The next day he returned the del Gesu back to Wurlitzer's, saying that it wasn't as good as Kreisler's after all. Shortly after that, Kreisler, himself paid a visit to Wurlitzer's. "Mr. Kreisler," they ruefully said, "you cost us a sale". "It gets better," replied K. "I was using my Vuillaume that night!"

BTW, I've tried the Kreisler del Gesu.

January 4, 2012 at 11:29 PM · All this talk about people being able to identify a Strad or a del Gesu or a modern fiddle is total bull**it.

I heard a guy playing a Strad tonight in a concert hall in London AND he was using Heifetz's bow!! (Allegedly, but how many times have I heard that about Heifetz's bow!)

It was the most miserable small insignificant sound you are likely to hear.

So was it because this misreable example of a Strad was used with a certain bow, or was it because the fiddler was a useless player?

Answers on a postcard please.

The truth is that no one can tell. We all hear sound differently and we all have different expectations. We all play differently.

So you can hear violins in test conditions until the cows come home, and they (the cows) may end up with a better idea of what makes a good fiddle than we do.

I certainly can't tell a Strad or a del Gesu from a foghorn, but I know what I want in a fiddle I use.

January 4, 2012 at 11:42 PM · "All this talk about people being able to identify a Strad or a del Gesu or a modern fiddle is total bull**it."

You knows it, Peter.

A few years ago I played two del Gesus side by side and they were so different from each other that there is no way I could have known they were by the same maker from the sound. One was not that good - a decent old fiddle, but nowt special.

The other was extraordinary. Nearly as good as my 1840 Knilling ;)

January 5, 2012 at 01:12 AM · Price is a basic component of competition. A 20K dollars Ford car competes in the market with a 20K dollars Toyota.

So, contemporary violins are not competing with old violins in the same price range since old violins are much more expensive.

I make new instruments - mostly violas - and my competitors are my other friend makers, not old instruments.

For the price you pay for a good contemporary instrument you will get a mediocre modern Italian in the same price range (in general oversanded, with a dry varnish, poor model, poor sound, but made by a "dead maker").

January 5, 2012 at 02:38 AM · I wish some of us here could get together, with fiddles we like and swap and play and talk and work out what is going on .

We can't really make much sense of it by typing out opinions.

January 5, 2012 at 04:10 AM · Perhaps a third factor is the performers playing style with a particular instrument. Many violinists compensate their bow speed and pressure to fit the musical needs. Different fiddle, different style. After an hour or so, a weak fiddle can sound full, and a strident E can play smoothly. If the musicians did not have sufficient time to adjust their playing technique to the violin, they may not be getting the best tone from it.

January 5, 2012 at 04:32 AM · Peter,

I would press "like" button if there were one

Graham,

I like that idea but we would all have to be billionaires :(

January 5, 2012 at 08:14 AM · Here is the URL to another article about the test, also published in The Guardian (UK): http://www.guardian.co.uk/music/musicblog/2012/jan/03/stradivarius-v-modern-violins-study.

Stephen, I like your remark about randomness in the study.

Recently I heard a performance of Schubert's quartet "Death and the Maiden" played on 4 Strads. The room was relatively small (compared to a concert hall) with excellent acoustics. The sound of the music was Divine! I could hear each voice separately and combinations of voices. The music seemed to fill the air in the room completely. The sound was very rich and expressed a wide range of emotions beautifully. I don't know how other instruments would sound with the same music and the same room, but the music I heard was Divine.

January 5, 2012 at 09:03 AM · "I wish some of us here could get together, with fiddles we like and swap and play and talk and work out what is going on ."

Graham - if you are ever in London? I'm often around.

January 5, 2012 at 12:04 PM · Peter, I often come down to London. I had a monthly gig at a pub in Bushey, but that is being stopped for the winter.

They are holding a fiddlers' gathering there called "Fiddle Hell" at the end of January. I might go to that.

January 6, 2012 at 12:18 PM · The 1st sounds fantastic in the upper range. It is focused and clean.

The 2nd has better lower range but sounds fuzzy on the upper register.

It will be nice to have a violin that has the low range of the 2nd and the upper quality of the 1st.

I think in the violin world, beyond certain objective threshold (good intonation, good phrasing, and etc), everything becomes subjective and is prone to cognitive bias. Remember the "Halo Effect"? A strad or a famous violinist sounds better to you than a modern violin or unknown violinist simply because of their name.

January 6, 2012 at 05:14 PM · A member of the Hallé Orchestra (UK) was called Sidney King. Because the names were printed surname first on documentation, "King Sidney" was given VIP treatment on arrival somewhere on the African continent when on tour with the orchestra, this despite being in the company of a crowd of plebs and an absence of a crown or any other ceremonial regalia.

This "halo effect" can be such a a dangerous thing !

January 6, 2012 at 06:09 PM · I had an odd reaction to the two NPR sound clips, my ear preferred the first violin the eveness of the sound across the strings and the clarity of tone is what sold me but I have to admit that there was something interesting about the second fiddle it had a bit of "growl" to the sound and was a bit punchy I'd be curious to hear it in a hall under concert conditions.The first fiddle had such a pleasant sound but would fiddle #1 contrast and carry over a heavy piano accompaniment or orchestra as well? My violin definitely sounds different out in the hall than it does under my ear I'd have to ask the experts when you record a fiddle with microphones up close do you get more of an "under the ear" sound than what you'd actually hear out in the hall?

-M

January 6, 2012 at 09:32 PM · I don't have access to data about the recording, but it sounds like maybe a 3 meter distance in a room with dry acoustics, which means it's not the "under the ear" sound, but not an indication of how the sound "expands" in a large hall either. It's important to know that carrying power can originate from different characteristics. A Gagliano could have a focussed, somewhat nasal sound that "punches" through the air, and a Strad might rely on a broad sound with rich upper harmonics which also will come across well to the audience. The same is true for different modern violins of course.

January 7, 2012 at 11:41 AM · I heard someone live playing the Korngold concerto the other night on a Strad (1711 if I remember correctly) and the sound was tiny. But it was probably (almost certainly) the violinist. I was sitting only a few rows away.

January 7, 2012 at 01:30 PM · That could well be the violinist - or the violin - or the acoustics, and any combination. It could be that the sound really started to bloom a little further down the hall.

Once I heard a violinist - her instrument pedigree was not that distingushed, but she is well-known - oh what the hell: Leila Josefowics. It was a good hall - New York's Town Hall. At first I heard her from near the back of the "orchestra" section - i.e. the ground floor, and she sounded fine. I hoped to get closer as a fellow fiddler to see what made her tick, so during intermission I looked for an empty seat closer up. Ideally, I would have liked to be in the 5th or 6th row, but the only available seat was in the very first row, right in front of her. The sound was not weak at all, but dry as dust. Again, that is a fine hall, but where I was , there was nowhere for the sound to bloom.

January 7, 2012 at 02:43 PM · In this case Rapahel, the leader of the orchestra, a young lady, a student of about 17 or 18, also had solos and in other works they did as well, and she was sitting only 3 feet from him. Her sound was bigger and more focussed ... so yes, it was definitely him!

Also a colleage told me he heard him do the Mendellsohn previously and it was just as bad.

January 7, 2012 at 03:58 PM · The player can't be ruled out. I've heard one very famous prodigy a few times in Symphony Hall and her sound seems not to carry at all, even to places where Pinchas Zukerman sounds like the voice from the burning bush. Of course, her del Gesu might not be as nice as his, but...

January 9, 2012 at 06:09 PM · The player interacting with the violin can certainly make a significant difference. Jacques Thibauld told this interesting story about his friend and former teacher, Ysaye. They made a test with Ysaye's Strad and del Gesu. Each one played on both, with the other one listening at a distance. When Y played, he projected better on the del Gesu; when T played, he projected better on the Strad.

BTW, just read through Laurie's blog on her own experience as a participent in this instrument contest, and highly recommend it. I also just saw the poll re whether we think Strads are better than modern instruments. I can't vote in a poll like that. WHICH Strads, and which moderns? It's so individual. And I find it very strange that even after the contest is over, they are not naming the modern instruments. In my personal experience, having tried about 5 Strads, 2 del Gesus, a number of Guadagninis, Amatis, etc. that my best modern instruments in my own collection would be quite competitive with them. The only exception would be a particular Amati that I tried a couple of years ago, that I wrote about in my blog re auctions. And of course, listeners have their own taste which may vary from how the violinist feels with the instrument under the ear. And as noted, a different player with the same instruments may produce different results with same instruments - so we can never have a completely 'objective' tests of this kind.

All that said, for me a more meanigful test would be half a dozen of the acknowleged finest classic violins vs 6 particularly first-rate contemporary ones. From personal listening experience and from what I've heard and read, for the 3 Strads, maybe the "Soil", the "Barrere" and the "Alard". For the del Gesus, maybe the "Kochansky-Rosand", the the "Vieuxtemps", and the "Lord Wilton". For the moderns, this gets trickier. But since this is my fantasy, I'd definitely include one from my own collection, and choose 5 others from many submitted to me from near and far. The violins would have specially selected excerpts played on each - first one across all the strings, then on each string, then chords, etc. They'd be played in pairs until each one was played against each other one - eg, at one point #2 vs #8, etc. I'd recommend a "10 point must system" as they often do in fighting contests, i.e. the winner must get 10 points, the other one gets 9 or less. One violinist - again, since it's my fantasy, ME, though if James Ehnness would like to do it, I'll bow out! - will play them all. I'm under no illusion about complete objectivity, but if in such circumstances one or two or three get a lot more points than the others, it just might mean something. At any rate, it would be very interesting. Many years ago a somewhat similar test was made, with Aaron Rosand as the test pilot, and the clear winners were a Bergonzi and a del Gesu (I think his own). But I think that some of the best fiddles made in the last 20 years would be more competitive.

This discussion has been archived and is no longer accepting responses.

Facebook Twitter YouTube Instagram Email

Violinist.com is made possible by...

Shar Music
Shar Music

Yamaha Silent Violin
Yamaha Silent Violin

Pirastro Strings
Pirastro Strings

Find a Summer Music Program
Find a Summer Music Program

Dimitri Musafia, Master Maker of Violin and Viola Cases
Dimitri Musafia, Master Maker of Violin and Viola Cases

Violinist.com Business Directory
Violinist.com Business Directory

Violinist.com Guide to Online Learning
Violinist.com Guide to Online Learning

Dominant Pro Strings

Antonio Strad Violin

Bay Fine Strings Violin Shop

Bobelock Cases

Fiddlerman.com

Fiddlershop

Los Angeles Violin Shop

Nazareth Gevorkian Violins

Violin-Strings.com

Metzler Violin Shop

Leatherwood Bespoke Rosin

Warchal

Barenreiter

Johnson String Instrument and Carriage House Violins

Potter Violins

String Masters

Bein & Company

Annapolis Bows & Violins

Laurie's Books

Discover the best of Violinist.com in these collections of editor Laurie Niles' exclusive interviews.

Violinist.com Interviews Volume 1
Violinist.com Interviews Volume 1, with introduction by Hilary Hahn

Violinist.com Interviews Volume 2
Violinist.com Interviews Volume 2, with introduction by Rachel Barton Pine

Subscribe