Jump to content
Maestronet Forums

Scientific investigations of Stradivarius violins--an updated review article


Recommended Posts

24 minutes ago, avandesande said:

I guess then we should keep passing around 19th century marketing baloney and keep quiet.

It's 19th century marketing baloney between two slices of 21st century pseudoscience.

Link to post
Share on other sites
  • Replies 564
  • Created
  • Last Reply

Top Posters In This Topic

47 minutes ago, avandesande said:

Since you brought up recording, with the advent of the internet it should be straightforward to set up a double-blind test that the entire world can hear for themselves.

The player should not know what he or she is playing either. This is called double blind testing.

Link to post
Share on other sites
23 minutes ago, Alma said:

As a retired proofreader, I suggest you double check your spelling before posting about what is not correct.

.

Thanks! I deadlift 200 kg. What is your ability?

I have reviewed many articles through my career. Some have been rejected. 

Link to post
Share on other sites

Do we know how opera singers and stage actors develop carrying power in their voices? I think we do. 

They develop strong resonances around 3000 Hz, called singer formant or speaker ring. It can be measured in LTAS or by singing power ratio. A search in Google Scholar shows many articles related to these topics.  

The human voice does not show strong directivity because of our mouth shape. The sound spreads uniformly, left and right, top and down . 

Researchers have tried to find singer formant in Strad violins. Have they found anything conclusive? It seems not. Anders Buen may know this topic better than I do.  

However, at 3000 Hz, the violin sound is highly directional. Is it possible that a Strad violin is directly projecting a stronger singer formant to the audience? Wouldn't that lead to greater carrying power? 

We usually record violins using overhead microphones, not microphones at the audience position which is projecting at downward angles.  That's why I think it would be a fun experiment to put a dozen 4-channel Zoom H2n recorders at different seat positions. They are low-cost units running on batteries. 2 front channels for direct sound and 2 rear channels for reverberation sound. No messy cabling. 

 

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

We usually record violins using overhead microphones, not microphones at the audience position which is projecting at downward angles. 

Violins have been recorded from all sorts of positions and distances

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

Do we know how opera singers and stage actors develop carrying power in their voices? I think we do. 

They develop strong resonances around 3000 Hz, called singer formant or speaker ring. It can be measured in LTAS or by singing power ratio. A search in Google Scholar shows many articles related to these topics.  

The human voice does not show strong directivity because of our mouth shape. The sound spreads uniformly, left and right, top and down . 

Researchers have tried to find singer formant in Strad violins. Have they found anything conclusive? It seems not. Anders Buen may know this topic better than I do.  

However, at 3000 Hz, the violin sound is highly directional. Is it possible that a Strad violin is directly projecting a stronger singer formant to the audience? Wouldn't that lead to greater carrying power? 

We usually record violins using overhead microphones, not microphones at the audience position which is projecting at downward angles.  That's why I think it would be a fun experiment to put a dozen 4-channel Zoom H2n recorders at different seat positions. They are low-cost units running on batteries. 2 front channels for direct sound and 2 rear channels for reverberation sound. No messy cabling. 

 

Our acoustics consultant company had the responsibility for the acoustics consulting for the new Oslo Opera. During the acoustic rehearsal tests, I had a sound level meter and could see the spectrum from singers. And of course you hear the effect very well. I think it was Johan Sundberg at KTH who introduced this therm and have done a lot of research on that. He shared the department with Erik Jansson, who you probably know, Bruce. He was into the acoustics of guitars and of violins. I visited Jansson once and had my instruments measured at his lab. One of the features he was looking for was where the phase shift for the bridge impact force and recorded response at the bridge side came. My best instrument had a wider bridge body hill like on a Strad. (It was a large French instrument from late 1800, we think). I think his last article is on the bridge and bridge irland region response, as well as Cremonese long flat arch and French round arch. I am not quite up to date on the literaure. Jansson say that Strads in general have a flatter bridge body hill than many other instruments like moderns, because the central region is thin. So I think that it is more high frequency projection in general from modern instruments than Strads. I understand, e.g. from Curtin's presentations and work that if you disregard the lower frequencies, something important will lack in the sound. Of course also if its the other way around.  

You need both strong fundamentals or first fundamentals, and the "singer formant". I think already Herman Meinel knew this from his PhD work before WW2. You also see the same concusion from Saunders work with Heifetz also pre WW2 and after. Meinels main article has been translated from German and published lately in the VSA Journal or - Papers. I think Joes results in general support his findings. 
One may prefer more or less of everything. So do players. Some prefer a brighter instrument, some a deeper sound. 

I think that is is not only the acoustic side of the instruments that matters. The playability and how the instrument is to handle matters, maybe more than the sound, per se. I am speculating here. Others know better than me how players behave when choosing instruments. From my Master thesis work on violin modes and properties, I worked with a fine violinist and his instrument. He was curious about how the instrument sounded out in a hall. Maybe some will let others play and they go and listen themselves. 

I have done a bit of testing at Oberlin with a sound level meter and recorder. I have also used a Zoom H4n. In that set Strads, del Gesu, a Nepolitan and other old instruments, as well as new instruments were mixed. It was not controlled tests, more "playing" with the instruments and listening. We had some scales e.g. At that time I was preoccupied with the divisions of spectra after Dünnwald and managed to sort the spectra into more or less "Dünnwaldian spectra" (whet he characterized as old Italian sound). The Naepolitan came first there, but it is a bit weak, also an old instrument trait, using DW parameters as the metric. I did not dive much into the properties of the Strad in there. I know Don Noon e.g. have and have shared his experiences with that. 

It belgs to the story that I come from a different tradition, and used to prefer deeper sounding instruments to very strong highs. I actually improve Hardanger fiddle setup by cutting highs by changing the bridge design. However I become surprised to learn that I actually like pretty stiff instruments too. Some are really stiff, beyond normal violins. And I have no hurry to fix that. They will sound more bright, and probably the understrings will ring a little longer. 
At my first visit i liked the darkest sound violin in the hall. An instrument the maker brought to fix so he could get it sold. I think I am a bit more calibrated now. I liked the two preferred violins in the blind test with the Strad and del Gesu. They simply were more even, and powerful and it was more than just a little. The difference was loud and clear. 

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

We usually record violins using overhead microphones, not microphones at the audience position which is projecting at downward angles. 

This is probably the most common mic position. Often a mix of near and distant mics are used in more advanced setups, recording productions, television etc, then they mix the dry and reverberant channel at will. There is a trend in music for more clarity, which takes near mics.

My father runs a recording and publishing company. The first recordings they did in 1976 was with the sound producer from Oslo Concert house in our old style timber house on the country side. Later he bought a DAT recorder and two fine B&K mics and recorded himself. The mics were positioned as you mention, 2-3m distance overhead in a 45 degree cross. These mics were omnidirectional, so you can in principle point them in any direction. The usual thing is to point them downwards, towards at the source.

This position in a concert hall will probably not sound similar to the sound heard by a listener in a hall, nor for the player. In a smaller studio, say 50m3, it probably does not matter all that much because there will be plenty of reflections and clarity. 

In a hall there will be a seat dip effect due to the periodicity of the seats. It worsens the longer from the source you sit. I think it is in the 125-200Hz (depends on te distances and geometry), which may not matter so much for violin sound. Maybe it does for cello or viola, I do not know.
That may be a good reason to lift the mic a bit off the seats. 

Another factor is the damping in the air. Higher frequencies are more damped in dry air than in more moist air (higher RH). The effect is large enough for users of a large auditorium to detect this effect on the high frequency reverberation if the climate change enough. 
We also see some effects of this on the transferred sound. E.g. a near recording to a violin being played will contain more bow "hiss noise". This disappears or weakens at, say 10-15m in a hall. A listener does usually not hear this unless he sits close. 

This noise is one of the things I understand matter for violin projection. It should sound "edgy" and a bit "untolerable" under the ears. It will still sound smother in the hall. 

There is also effects of directivity, but it is more difficult to interpret without using arrays or repeating the measurements around. Joe does that. And will probably share info on this. Bissinger also measured violin directivity at Oberlin with his rig.

I guess more sound radiates a bit more from the top plate side than the back around 2-2,5kHz. The room reverb sort of collects and walls and elements reflects the radiated sound that does no radiate directly towards the listener, towards the listener. The sound just travels a bit longer. If that happens, within 80 ms of the direct sound, we are happy. 

 

Link to post
Share on other sites
50 minutes ago, Urban Luthier said:

interesting demonstration of the affect of microphone placement in the recording process

One important effect with such semi near mics is not only the directivity of the instrument but probably the comb filtering of the direct sound and floor reflection. There will be deep dips at the cancelling frequencies and tops where the sound waves meet in phase.  

Link to post
Share on other sites
9 hours ago, David Burgess said:

1. Carl, Anders is an acoustician, by profession.  What are your qualifications for describing his opinions as "nonsense"?

Thanks, David for supporting me, when I have run into trouble! It actually calms me down a little. 

Link to post
Share on other sites

Bruce Tai: Thank you for putting this report here and the information it has. I will be reading it in detail later. I was particularily interested in the statement about particle size in fillers, "If ground finely (<5 um), such fillers can enhance reflective brilliance without adding cloudiness," on page 13. It appears the transparentness of pigments is greatly increased when the particle size decreases. The wavelength quality of light seems to dominate then and the waveform simply goes around the particles when they are small. Now I have a quantitative particle size to put in the mental work. 

Edited by Greg Sigworth
spelling error
Link to post
Share on other sites
2 hours ago, GeorgeH said:

It's 19th century marketing baloney between two slices of 21st century pseudoscience.

That's a little more concentrated than my take, but I get whiplash from the extremes of biased opinion and conjecture alternating with detailed chemistry that is beyond understanding.  I can't tell if this is supposed to be a an article in Popular Mechanics or a Chemistry research report.

While I do appreciate Bruce making this article available for all, at the end of the day all I have is more uncertainty and mystery, and some things that bother me.  The obvious bias and praise of old instruments is certainly not appropriate if this is supposed to be an objective scientific investigation.  Off-hand dismissal of contemporary research that disagrees with the writer's bias is also a red flag.  And omitting age-related optical changes in the wood when discussing varnish aesthetics is plain sloppy (it is well-known that old violin wood is opaque, and newer wood is not).

My personal experience with old-vs-new comparisons is more along the lines of the Fritz/Curtin results. 

At VSA 2014, two well-known soloist Strads were played against 2 modern instruments in a blind listening.  I had a definite preference for the moderns, although I thought one of the Strads was pretty close.  Sure, I'm not playing them and I have my own tonal preferences.

While preparing for a concert, the solist was offered to use a Strad owned by a member of the orchestra, instead of the 1 year old modern one owned by the soloist.  The soloist, conductor, and owner of the Strad all decided that the modern would be the better instrument for the situation.  (my favorite anecdote, since I made the modern one)

In a private meeting, several good modern instruments were compared with several excellent Cremonese ones.  I won't say what they were, but you'll just have to trust me that these old ones were top soloist instruments.  It wasn't a blind test, but with this selection of the best, there was a definite difference between the old and moderns.  I'll give the old ones a win under these specific circumstances... but if you filtered all of the modern instruments as carefully to find the best of the best, maybe the result would be different.

Link to post
Share on other sites
7 hours ago, Don Noon said:

it is well-known that old violin wood is opaque, and newer wood is not

This is new to me. Can you explain a bit more? 

Is old wood more opaque before varnishing or after varnishing? 

Why is the cause? Is it because if yellowing?

Link to post
Share on other sites

While it may not be apparent to some readers in this thread, what we having here is a healthy debate between old acquaintances but not quarrels of any sort. I have been sharing my research articles on this forum since 2007 (my first VSA varnish review). Different kinds of criticism have shown me many new ways to look at the problem. Through my interactions with the knowledgeable members of this forum, I have been educated on many aspects of violin making and violin properties. 

Based on personal experience, there are believers and non-believers in the superlative qualities of Strads. Some believe in the "blind tests," but I also pointed out in my article that these are flawed and cannot really address the tone quality problem. 

Instead of getting trapped in the "Strad vs. best modern" violin debate, we could compare Strad vs. other old Italian. I think it is universally accepted that Stradivari and del Gesu were a class above other old Italian makers. If so, then why? What are the acoustic correlates of the favorable tonal qualities described by soloists and collectors? If we can actually answer these questions, we would be making great progress. 

 

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

This is new to me. Can you explain a bit more? 

Is old wood more opaque before varnishing or after varnishing? 

Why is the cause? Is it because if yellowing?

Don will explain. 
Meanwhile: While thinning a top plate, holding it against a light source will give some useful info on the graduation distribution and very efficiently. Thinner areas appeaar lighter than the thicker. White or varnished, relatively new, and thin wood will, to some extent, be translucent. As the wood ages, exposed to light and Oxygen or whatever, the wood darkens and finally becomes opaque, maybe after 100 years, or 70? The repairmen here knows. The time it takes may depend on the thickness, and possibly what the instrument has been exposed to over the years. maybe also the wood properties and the used varnish may play a role?

The chemistry part here should be interesting for you and maybe you tell us what happens after a while. I also think there is a mix of optics into this. Light and photons, possibly also invisible ones, may play a role. Humidity perhaps?

Added from a discussion board on this theme: 

While most wood we see comes from stain or oils, wood left on its own will darken. As far as this article suggests, this is caused by the polymerization of polyphenols from lignin into tannins which are dark red or brown in color when they are highly polymerized.
The specific content of the lignin in different woods will affect the density, strength and color of the wood.

Edited by Anders Buen
Added some info on the causes of wood darkening.
Link to post
Share on other sites
4 minutes ago, Bruce Tai said:

While it may not be apparent to some readers in this thread, what we having here is a healthy debate between old acquaintances but not quarrels of any sort. I have been sharing my research articles on this forum since 2007 (my first VSA varnish review). Different kinds of criticism have shown me many new ways to look at the problem. Through my interactions with the knowledgeable members of this forum, I have been educated on many aspects of violin making and violin properties. 

Based on personal experience, there are believers and non-believers in the superlative qualities of Strads. Some believe in the "blind tests," but I also pointed out in my article that these are flawed and cannot really address the tone quality problem. 

Instead of getting trapped in the "Strad vs. best modern" violin debate, we could compare Strad vs. other old Italian. I think it is universally accepted that Stradivari and del Gesu were a class above other old Italian makers. If so, then why? What are the acoustic correlates of the favorable tonal qualities described by soloists and collectors? If we can actually answer these questions, we would be making great progress. 

 

Remember that even Strads are different. They are probably more different than what most modern makers do put out. A whole lot of them are probably not good at all. I think it is more fruitful to discuss tonal traits in general and effects of variations of the input or the source on the resulting tone. I know you have support for your thoughts and probably play a lot ball with a retired chemistry professor, also doing quite a lot of "acoustics". Varnish and small changes to wood properties can't be very important acoustically. Damping is dominated by holding the instrument, and small variation in wood properties is what you also get from natural variation in wood properties. 

Link to post
Share on other sites
14 hours ago, Urban Luthier said:

interesting demonstration of the affect of microphone placement in the recording process

I enclose a graph showing the calculated combination of direct sound from a sitting violinist and the floor reflection at a mic at 2 m distance and 2,5m height. If the mic or player move, the curve will change accordingly. 

210417 Floor comb filter.jpg

Link to post
Share on other sites
17 hours ago, Don Noon said:

it is well-known that old violin wood is opaque, and newer wood is not 

9 hours ago, Bruce Tai said:

This is new to me. Can you explain a bit more? 

7 hours ago, Anders Buen said:

Don will explain. 

I'm not sure how well I can explain it... but if you have been around new and old wood and cut some shavings and look at them, it is obvious that wood becomes darker and more opaque with age, with some possible photo-bleaching depending on age and exposure.  Or look at old violins where the varnish is worn down to the bare wood and then polished over, and compare it to a similarly finished new piece of wood.

I recall a video (which I can't immediately locate) where Robert Cauer was checking for a Strad authenticity, and the first test was to shine a light on the top and look inside to see how much light came through.

As for the chemistry and mechanisms for opacity, I can only point to exposure to light, air, and other natural chemical reactions.  Your paper points out chemical reactions that occur spontaneously with time, and if chemistry changes, optics can change with it.

I am astounded that someone in chemistry writing articles about old violins would be unaware of this fundamental fact.

 

 

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

While it may not be apparent to some readers in this thread, what we having here is a healthy debate between old acquaintances but not quarrels of any sort. I have been sharing my research articles on this forum since 2007 (my first VSA varnish review). Different kinds of criticism have shown me many new ways to look at the problem. Through my interactions with the knowledgeable members of this forum, I have been educated on many aspects of violin making and violin properties. 

Based on personal experience, there are believers and non-believers in the superlative qualities of Strads. Some believe in the "blind tests," but I also pointed out in my article that these are flawed and cannot really address the tone quality problem. 

Instead of getting trapped in the "Strad vs. best modern" violin debate, we could compare Strad vs. other old Italian. I think it is universally accepted that Stradivari and del Gesu were a class above other old Italian makers. If so, then why? What are the acoustic correlates of the favorable tonal qualities described by soloists and collectors? If we can actually answer these questions, we would be making great progress. 

 

 

This is from the article...

"First, fewer than 20% of the 500 surviving Stradivari violins remain in top conditions fit for concert violinists, but we do not know the quality of instruments entering these stud-ies. Secondly, short-term memory for timbre quality only lasts for seconds, but it takes min-utes for the players to change violins. By the time the next instrument is played, the mem-ory for the previous instrument has severely decayed. Thirdly, the loudness has to be strictly matched to carry out timbre comparisons, but this is not feasible in a listening test with live performances. Louder violins often give the initial impression of sounding better in blind tests, masking timbre dif-ferences[5]. Therefore, it is not surprising that blind tests carried out by Fritz et al. were inconclusive."

Since you mention that the quality differences are apparent using recording, doing blind tests using recordings should address the last two issues and the first is solvable as well.

 

As the saying goes the burden of proof rests on those making extraordinary claims. The problem is nobody in the industry really wants to dispel these myths since it goes against their financial interests of those that have bought into it. Even makers benefit as collectors keep many good instruments out of the hands of musicians.

 

 

Link to post
Share on other sites
5 hours ago, Anders Buen said:

I enclose a graph showing the calculated combination of direct sound from a sitting violinist and the floor reflection at a mic at 2 m distance and 2,5m height. If the mic or player move, the curve will change accordingly. 

Wow that is a huge impact - thanks for sharing. Is mic placement one of the reasons that strings instruments don't sound as natural as they should in recordings vs what we experience live in a good concert hall? If it comes down to microphone placement, why don't more engineers utilize a binaural microphone setup (ie. a dummy head with mics placed in the ears. I understand some Chesky recordings were made this way and they 'give the illusion' of sounding quite natural. 

 

Link to post
Share on other sites
25 minutes ago, Urban Luthier said:

Wow that is a huge impact - thanks for sharing. Is mic placement one of the reasons that strings instruments don't sound as natural as they should in recordings vs what we experience live in a good concert hall? If it comes down to microphone placement, why don't more engineers utilize a binaural microphone setup (ie. a dummy head with mics placed in the ears. I understand some Chesky recordings were made this way and they 'give the illusion' of sounding quite natural. 

The Norwegian Broadcasting Company (NRK) do live productions, every now and then. One of these was the opening of the Oslo Opera. Our company had the acoustics consulting for that with an English company, Arup Acoustics. We wondered if we could have some sound files for our auralisation in the room acoustics program we use. One of the recording engineers there told that in an orchestra they will record every musician with their own mic in addition to some room mics, I guess, for the reverb. Then they mix it until it sounds as good as possible. It does not have to sound like the real auditorium experience. From this occasion I would guess it was close, because the Oslo Opera is almost reverberant like a concert hall.

I once attended such an event with NRK present, as well as that particular recording engineer. My cousin Per Anders was the artist, and the fiddle he used is one of the best regarded instruments in the family. A dark-ish and even sounding instrument, with a violin top. Technically, it is a Hardangerfiddle. It is recorded on several CD's and was used by my, now passed, uncle Hauk. I know that sound well. I like it very much. And my cousin had it perfectly in tune, and managed the wind. It was outdoors, the instrument was miced, as well as one for speech. Maybe there was more.

We listeners got both the natural and the amplified sound without noticing the difference, and it sounded great! I do not know how the recording and radio program became. 
The recording engineer did a great job, and I told him. 
The NRK have their own house symphony orchestra, KORK, and the "Store studio" (The large Studio) is their rehearsal and concert hall. The recording engineers "main home" is the control room there. He and his collegues does a lot of musical instrument recordings and productions. 

One of the major performers in Norway on Hardangerfiddle, uses a floor mic in addition to a few others in air. Close mics give less comb filtering. Smaller rooms also give less coloring, although one can hear that a room is smallish or biggish, I think.

Edited by Anders Buen
Improving the text, and corrected spelling errors
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...