Once again new violins chosen over Strads in blind test

[Blumlein 88]

http://www.thestrad.com/players-fav...old-in-largest-ever-blind-testing-experiment/

http://www.pnas.org/content/early/2017/05/02/1619443114.abstract

This report has some sample audio of the violins.

http://www.npr.org/sections/health-...arius-violin-easier-to-hear-science-says-nope

 

[RayDunzl]

Did they test one of these?

 

[Blumlein 88]

Did they test one of these?

I don't think so.

I haven't been able to find the names or prices of the modern violins. This is the same team that have done two other blind tests. In those they used modern violins in the $5k-$10k range.

I do love when people make statements that defy physics and then act all hurt when it turns out not to be so.

Experts say these antiques possess the curious ability to sound quieter under the player’s ear, yet project sound farther and better in a concert hall than newer models.

I saw another professional violinist say, Modern instruments are incapable of projecting sound beyond the 6th row no matter how they are played

Does it really take a blind test to refute that kind of thinking? Wimpy violins project wimpy sound and it runs out of juice at the 6th row and dies. We need special speakers that do that so we could cut down on sidewall reflections and bass build up indoors.

 

[amirm]

As spock would say, "fascinating!"

 

[DonH56]

Carbon fiber instruments are also coming out, with similar cries of "horrible, terrible sound" etc. until they hear without seeing and/or some famous player uses one and reveals it later. Similar things happen with early trumpets; people pay a fortune for certain vintage instruments that often sound to a fellow player the same or worse than a modern horn. And those old hand-made instruments often have quirks, and you don't one made on Monday or Friday, and all that jazz... My horn is a custom by a modern designer who takes great pride in measuring and adjusting each horn he makes so it sounds the same as all the rest. He has special tools to clean and tweak the innards and such. Mass-produced horns have a fair amount of variance and antiques are even worse.

That said, it is pretty easy to tell a student horn from a professional model, at least most of the time, and to tell variations among horns that have not been carefully designed and tested. I'm sure it is the same with violins and other things.

 

[oivavoi]

Very interesting. Thanks. I skimmed through the original article, seems solid. If anybody would like a copy of the research article and doesn't have access beyond the academic paywalls, send me a pm.

Does anybody know of similar tests of grand pianos, btw? Asking since I'll probably buy a grand piano for my own use in the the next couple of years. I only know of one, in which a cheaper Chinese one bested a Steinway (the test is in French). https://www.feurich.com/no/special/awards/diapason-dor-2008/

For grand pianos I doubt that it makes much sense to blind test specific models though, since grand pianos are so complex that individual units may vary considerably. My sighted playing has told me me that Bösendorfer and Steinweg's (not Steinway's) often are the grand pianos I like the most. But it is difficult to know how biased that assessment is. Nevertheless, I only play sighted, of course. So the subjective appreciation of having a certain brand shouldn't be dismissed.

EDIT: Among piano players, at least, I don't think there exists any myth about old pianos being better. Everybody knows that a piano deteriorates with age.

 

[Cosmik]

EDIT: Among piano players, at least, I don't think there exists any myth about old pianos being better. Everybody knows that a piano deteriorates with age.

Interesting... I have a 1905 upright which, to me, sounds like a 'proper piano' - although I have heard stronger bass notes from later pianos. It's got a metal frame so stays in reasonable tune, although it is by now a semitone down from normal pitch. Maybe it was a good one in its day. This catalogue dates from a few years after my piano.

I recently went round to friend's house where they have a brand new Yamaha upright (I think) and I was immediately struck by (a) how loud, and (b) how 'synthetic' it sounded. It played and sounded more like a typical electronic piano than a real one. In the humble world of uprights, at least, I would prefer my £50 antique over that particular modern one.

 

[oivavoi]

Interesting... I have a 1905 upright which, to me, sounds like a 'proper piano' - although I have heard stronger bass notes from later pianos. It's got a metal frame so stays in reasonable tune, although it is by now a semitone down from normal pitch. Maybe it was a good one in its day. This catalogue dates from a few years after my piano.

I recently went round to friend's house where they have a brand new Yamaha upright (I think) and I was immediately struck by (a) how loud, and (b) how 'synthetic' it sounded. It played and sounded more like a typical electronic piano than a real one. In the humble world of uprights, at least, I would prefer my £50 antique over that particular modern one.

Of course, there are exceptions to this "rule". Genereally speaking, many pianos were often manufactured in a better way back in the day - that goes for uprights at least. Many modern pianos in the budget range are not very good. So an old well-made piano, that has been properly maintained, may still be better than a modern one. But a piano will deterioriate with age. If your vintage piano was manufactured today in exactly the same way, it would have been better than the one you have. Here's a thread on a piano forum where the technical reasons for this are outlined: http://forum.pianoworld.com/ubbthreads.php/topics/266616/1.html

 

[Cosmik]

Here's a thread on a piano forum where the technical reasons for this are outlined: http://forum.pianoworld.com/ubbthreads.php/topics/266616/1.html

OK, that puts me in my place!

In my opinion, people who say they have hundred year old pianos that they say sound and play good have a high tolerance for poor sound and loose clattery actions.

 

[DonH56]

My wife plays piano (keyboard, organ). Most of the older pianos I have seen are a mess. An upright has a very different sound compared to a grand and construction is different, natch. Years ago when I got her a nice little grand I went with Boston, a Steinway company. I let her choose and she went by feel of the keys and sound. I paid attention to construction and some of the special features Boston offered that made it sound better than the competition. I expected to get a Steinway or maybe Yamaha but the Steinway's sound she did not like quite as much and it was about 30-50 % more costly. The sound was a little different and arguable neither was better, she just preferred the Boston (as did I -- the bigger tail made for a bit deeper sound). This was for a 5'10" model, not one of the bigger grands. The Bosendorfer was also nice, again a little different sound, but well out of my price range and when we looked the store didn't have a smaller Bosendorfer (we don't have room for a much larger model). Most of the better pianos we saw had heavy steel frames and complex multilayer sound boards to achieve a really good sound across the entire range.

YMMV - Don

 

[Jinjuku]

Vintage electric guitar may however do better:

1. True single slab bodies
2. Many made out of old growth wood species with a denser grain

At least that is that I have been told

 

[DonH56]

I have a friend who, in addition to being another anal design engineer, is a very good guitar player. He bought a couple of custom new guitars (one acoustic, one electric, about $4k each) a few years ago and schooled me on a lot of the aspects of guitar design. We also did some mic'ing and measurements, natch. Wide range of impact from body, neck, and pickups for vintage and newer models, and cheap to very expensive models. Kinda' like everything else, generally but not always did more money provide a better guitar, and differences tend to become much smaller after you exceed a certain price point. But, I'll bow to your avatar and assume you know waaay more about guitars than I! I did make some decent money making custom stomp boxes for the beasts decades ago.

 

[watchnerd]

I don't think so.

I haven't been able to find the names or prices of the modern violins. This is the same team that have done two other blind tests. In those they used modern violins in the $5k-$10k range.

I do love when people make statements that defy physics and then act all hurt when it turns out not to be so.

Experts say these antiques possess the curious ability to sound quieter under the player’s ear, yet project sound farther and better in a concert hall than newer models.

I saw another professional violinist say, Modern instruments are incapable of projecting sound beyond the 6th row no matter how they are played

Does it really take a blind test to refute that kind of thinking? Wimpy violins project wimpy sound and it runs out of juice at the 6th row and dies. We need special speakers that do that so we could cut down on sidewall reflections and bass build up indoors.

Oh, that's just the tip of the iceberg.

About 2 years ago we were doing a spot mic set up for a very famous soloist who plays a very well-documented (as in movies) Strad. We were setting up the local symphony standard for violin soloists, the Schoeps M222.

We did the initial setup using a modern violin from 2002. Then when we switched to the Strad, the soloist told us, "You need to cut the volume in half because this violin is twice as loud as that modern one."

We promptly did so...

"What happened?"

"We cut the gain in half like you wanted."

"Well it's too soft."

"Okay, well...how about now."

"Still too soft."

"Now?"

"Closer, but still not there..."

"How about now?"

"Yeah...that's good...that's just right."

At which point were actually +4 dB more gain than the reference level set using a modern violin.

 

[Phelonious Ponk]

There is some science to the quality of old stringed instruments. As the wood ages and dries, the cellular structure changes, and the wood takes on, and retains, less moisture. So the wood is stabler and, some say, more resonant. I know a lot of old acoustic guitars that have survived from the 30s and 40s are just louder, and that's a beautiful thing for a guitar player because such an instrument responds very immediately to a light touch, and just gets louder as you dig in. It gives you great dynamic control. But personally, I'm guessing it has more to do with how lightly they were build back in the day compared to modern factory guitars.

Tim

 

[Jakob1863]

<snip>

I do love when people make statements that defy physics and then act all hurt when it turns out not to be so.

Experts say these antiques possess the curious ability to sound quieter under the player’s ear, yet project sound farther and better in a concert hall than newer models.

I saw another professional violinist say, Modern instruments are incapable of projecting sound beyond the 6th row no matter how they are played

Does it really take a blind test to refute that kind of thinking? Wimpy violins project wimpy sound and it runs out of juice at the 6th row and dies. We need special speakers that do that so we could cut down on sidewall reflections and bass build up indoors.

It seems so, that blind tests are needed but it will take a long time to change what people think. I´ve read a statement of a professional violonist from 2016 who still asserted that a old violin projects much better. (given the context of the interview he apparently meant that old violin generally project much better)

But otoh it too suprises me everytime in which way the alleged scientific writers, trying to explain such study result to a broader audience, distort what was really done.

See for example an excerpt of your third link above:
"Dedicated NPR listeners and violin enthusiasts may remember a few years ago when a team led by the French acoustics researcher Claudia Fritz published a study showing that blindfolded professional violinists could not tell the difference between a so-called Old Italian violin (they tested instruments made by both Stradivari and Guarneri) and a new violin."
(bold tag inserted by me)

or a quote from the National Geographics website:
"But to Claudia Fritz from Sorbonne University, the search for Stradivari’s secrets is a “perennially fruitless one”… because they don’t exist. In two studies, she has shown that professional violinists can’t tell the difference between the so-called “Old Italian” violins and newly made ones."
(bold tag inserted by me. source: http://phenomena.nationalgeographic...violins-arent-better-than-new-ones-round-two/)

Although in both articles, much more detailed in the second on, a more detailed description followed, is the "can´t tell the difference" part the most quoted and repeated in headline or abstracts since the appearance of the first study by Fritz et al.

But it is simply wrong. Apparently the violonists could tell the difference, but that wasn´t the hypothesis/question examined.

 

[Cosmik]

It seems so, that blind tests are needed but it will take a long time to change what people think. I´ve read a statement of a professional violonist from 2016 who still asserted that a old violin projects much better. (given the context of the interview he apparently meant that old violin generally project much better)

If we think the world needs louder violins, why not test them objectively using, say, a calibrated spinning disc instead of a human player with a bow? I would imagine it will probably turn out that the playability, feel, smell and 'cachet' of an older violin is the key to the way a player perceives its sound, but clearly there is an issue of 'engineering' involved in building such a thing.

I declare that, for me as a listener, a violin is just a violin (likewise a piano) when it comes to the recorded sound. I think the position of the mic and the recording venue's acoustics have far more of an effect on the sound than whether it's a Stradivarius or not. I imagine the player can tell more about it, because they probably adapt their technique to maintain something like the sound they are accustomed to; if it's easier to play they will perceive it as projecting better.

 

[Jakob1863]

If we think the world needs louder violins, why not test them objectively using, say, a calibrated spinning disc instead of a human player with a bow? I would imagine it will probably turn out that the playability, feel, smell and 'cachet' of an older violin is the key to the way a player perceives its sound, but clearly there is an issue of 'engineering' involved in building such a thing.

I am neither a violonist nor a specialist in the sound or physics of violins, so have to be careful.
But, the statement that Blumlein88 quoted was a bit different in saying that "old violins project better while being softer on the violonist´s ear" . I stumbled a bit about his remark that such a statement defies physics because that´s not easy to evaluate. As far as i remembered the violin has vastly frequency dependent directivity and the opposite of the quoted statement is already known to be true, means violins indeed can project bad, although beeing hard on the violonist´s ear. So i wouldn´t have dismissed it right from the beginning as defying physics, more as being unlikely that all old violins should excel per se in this relation.

And yes there is a lot of physics and engineering needed to build a high quality violin, i remember there was once a study (funded by the European Union) trying to find what is important to violonists in the evaluation of a violin and in which way they described the different aspects. The survey included afair well above 100 violonists, but i´ve to dig in my archive as i can´t remember who had done it.....

I declare that, for me as a listener, a violin is just a violin (likewise a piano) when it comes to the recorded sound. I think the position of the mic and the recording venue's acoustics have far more of an effect on the sound than whether it's a Stradivarius or not. I imagine the player can tell more about it, because they probably adapt their technique to maintain something like the sound they are accustomed to; if it's easier to play they will perceive it as projecting better.

Of course it matters what mic is used and placement and room acoustics are important parameters, but a good recording imo will capture the difference in sound between different violins quite well. But i´d never believe that every Stradivarius sounds alike or on a comparable quality level (or any other old violin).
Btw, again afair after the first study published by Fritz et al. there was a website where one could evaluate two violin samples and the majority picked correctly the Stradivarius. Which is interesting....

 

[DonH56]

There was a study years ago (I cannot find reference, it was linked on a musician's forum) showing listeners could clearly pick out a Stradivarius from a modern violin. Further research revealed that the player actually played the selection better on the Strad, perhaps knowing the instrument was that revered old fiddle. They took the same player blindfolded and repeated the exercise, again requesting identical performances. The second time nobody could reliably tell Strad from modern. I don't know if it was the same study or even relevant, but wanted to relate a tale showing how hard it is to account for all the variables.

Chances are if I have a Blackburn (very nice, expensive) trumpet in my hand, and know it, I am going to try my best to play up to it whereas I might just blow through the selection on a student horn just because my mind "knows" what to expect from each horn.

When I was in college studying pre-med stuff one of the courses highlighted the Driver's Ed Effect. A bug study was done showing students with Driver's Education were better drivers and also had better grades and such. Later studies revealed that the student's who took driver's education were generally better students to begin with, and with or without driver's education the students with better grades were better drivers.

 

[oivavoi]

There was a study years ago (I cannot find reference, it was linked on a musician's forum) showing listeners could clearly pick out a Stradivarius from a modern violin. Further research revealed that the player actually played the selection better on the Strad, perhaps knowing the instrument was that revered old fiddle. They took the same player blindfolded and repeated the exercise, again requesting identical performances. The second time nobody could reliably tell Strad from modern. I don't know if it was the same study or even relevant, but wanted to relate a tale showing how hard it is to account for all the variables.

Chances are if I have a Blackburn (very nice, expensive) trumpet in my hand, and know it, I am going to try my best to play up to it whereas I might just blow through the selection on a student horn just because my mind "knows" what to expect from each horn.

When I was in college studying pre-med stuff one of the courses highlighted the Driver's Ed Effect. A bug study was done showing students with Driver's Education were better drivers and also had better grades and such. Later studies revealed that the student's who took driver's education were generally better students to begin with, and with or without driver's education the students with better grades were better drivers.

Very interesting!

I also think that I would personally play better if it was a grand piano that I knew had belonged to Keith Jarret. I think I would just treat the instrument in a different way. Even though a cheaper, newer model probably would have been technically better.

 

[DonH56]

Yup. Perceptual bias works in many areas outside listening tests.

Alas, the best trumpet in the world in my hands still won't make me play like Wynton Marsalis or Adolph Herseth...