Jump to content
Maestronet Forums

Cremonese verus the rest


Argon55
 Share

Recommended Posts

quote:


Originally posted by:
David Burgess

When it comes to resulting statistics being meaninful, I'm sure some will default to their old beliefs, some won't, and some will find fault with any possible methodology.


Perhaps because any methodology applied to something as subjective as sound judgement, even in the well thought out context you propose, would, ultimately, possess flaws.

I cannot, as I've said previously, see how such a "contest" can account for all variables in a truly replicable and scientific (and hence meaningful) way.

Link to comment
Share on other sites

  • Replies 332
  • Created
  • Last Reply

Top Posters In This Topic

quote:


Originally posted by:
falstaff

Perhaps because any methodology applied to something as subjective as sound judgement, even in the well thought out context you propose, would, ultimately, possess flaws.

I cannot, as I've said previously, see how such a "contest" can account for all variables in a truly replicable and scientific (and hence meaningful) way.

Any experiment has flaws, real or perceived. They can't be completely eliminated. All I can offer to do is bias them heavily in favor of the old instruments.

Real or perceived flaws don't prevent an experiment from having validity. Sometimes multiple "flawed" experiments can present a useful preponderance of evidence.

Can you cite even one experiment in all of human history that hasn't been arguably flawed?

If we had no tolerance for "flawed" experiments, we'd probably still be living in caves, certain that the earth was flat.

One needn't throw the baby out with the bathwater.

Link to comment
Share on other sites

quote:


Originally posted by:
falstaff

All other aspects aside, no one has addressed MD's question of how you will limit the voters in such a trial to folks who have previously demonstrated the ability to tell one instrument from another.

Without some confidence that is demonstrably justified in the voters, I say the test is bogus.

My suggestion about giving the players a vote in no way obviates the need to have an audience that can hear a difference if there is one.

To qualify the listeners, at random times, play the same instrument as A and B. Only those who can truly tell the difference will be able to say, "hey, that sounded like the same instrument." If the test is old versus new. Have them specify old or new without comparison to another instrument. Sometimes, don't switch between category. Play both old or both new and see if they can consistently tell the difference. Also, make sure it is statistically valid, involving many trials.

Link to comment
Share on other sites

Last Chair described a method for selecting expert listeners. And there will be other valid methods also. But really, since musicians play for audiences who generally aren't experts, the solution which is most like the real world is a panel of non-experts. If there is a significant difference in sound, then they'll be able to pick it up. If not, then we're talking about subtleties not apparent to the majority of the population. If the latter is true then this leaves the higher echelons of the violin trade open to the accusation that the king has no clothes.

But as I've often repeated, we'll only know if a well-designed study is carried out :-)

Link to comment
Share on other sites

quote:


Originally posted by:
Argon55

But really, since musicians play for audiences who generally aren't experts, the solution which is most like the real world is a panel of non-experts.

Winemakers make wine that is mostly consumed by non-experts too... does this mean we should pick our wine judges from the front bar of the local pub? Or our film critics at random from the Saturday afternoon matinee?

No... I'm with Michael and others on this one. If the outcome is to be at all credible then it will need to be judged by a panel of people who are accepted "experts" in the field.

Link to comment
Share on other sites

We already have the only jury that really counts...ie that of top

international performers and the instruments they perform on &

who could be better judges than these players. The majority

persist in using old Cremonese instruments some do use modern

instruments & I can think of one who performs on modern

exclusively...others own modern instruments but do not perform on

them so much but if they genuinely purchased these modern

instruments that is a great accolade for some modern making.

Link to comment
Share on other sites

quote:


Originally posted by:
yuen

Hi all,

The only fair test but never been performed is this one: take out the labels, price tags of the

violins and play them (take your time) to test if any one suprior than others.

That's a pretty good test.

Some of us have done this informally by handing players different instruments, both Cremonese and copies with fake labels, or labels covered, or in a room dark enough that labels can't be seen.

quote:


Originally posted by:
Melving

We already have the only jury that really counts...ie that of top

international performers and the instruments they perform on &

who could be better judges than these players.

That might be the last word if we could eliminate all the variables other than the way the instrument sounds and plays. There's the placebo effect, the fact that some people buy instruments because they have significant money to invest, and the cachet of saying you own one or perform on one.

People will pay a lot of money (not just for fiddles) to have something that other people admire and covet.

Look at the money people will pay for an old master painting. It doesn't do anything, just hangs on the wall..........

Link to comment
Share on other sites

quote:


Originally posted by:
lastchair

To qualify the listeners, at random times, play the same instrument as A and B. Only those who can truly tell the difference will be able to say, "hey, that sounded like the same instrument."

That's a great idea. Score sheets from people who gave significantly different scores to the same instrument played as A and B could be thrown out.

Link to comment
Share on other sites

David Burges writes.....

'There's the placebo effect, the fact that some people buy

instruments because they have significant money to invest, and the

cachet of saying you own one or perform on one.

People will pay a lot of money (not just for fiddles) to have

something that other people admire and covet.

Look at the money people will pay for an old master painting. It

doesn't do anything, just hangs on the wall..........'

Yes You are right there David.

I've had players say to me that they feel that part of the

attraction for the audience is that the price of a ticket includes

the privilege of hearing a Strad, Guarneri or quartet of Old

Cremonese etc....

Link to comment
Share on other sites

If we're trying to determine whether there is a difference between old Cremonese instruments and modern ones (as opposed to whether the average concert-goer can perceive the difference), then we need the expert panel -- people who can demonstrate the ability to reliably distinguish instruments.

One thing I can't figure my way out of though is: if the members of a panel are good enough to distinguish various types of instruments, how would such an experiment control the members' voting for certain instruments because they can recognize them (at least into classes of instruments) and vote according to their preferences (or bias)?

Said another way, we need the experts on the panel but, by the fact that they're experts, might we not be pre-determining the results?

Link to comment
Share on other sites

I don't know if the instrument atracts the audience, a good part of the audience is a bit ignorant even about the pieces and composers that will be played in a concert, I think. And they are a bit ignorant about the price of violins in general. They will be surprised that an instrument made in 1970 can be sold for 100k, and are unaware about the prices of Strads, and the name Del Gesù is ignored by them, I think.

Link to comment
Share on other sites

quote:


I don't know if the instrument atracts the audience, a good part of the audience is a bit ignorant . .

I have a friend that had lent one of his instruments to Sidney Harth. After a concert, he met Sidney in the dressing room and asked him "You weren't playing your Strad were you?" Sidney said "I never say what I am playing or not playing. People are paying money to hear me play the Strad."

Link to comment
Share on other sites

One thing that needs to be clarified. This idea is consistently described as a "test" or an "experiment."

In the final analysis this is a POLL, not a TEST.

You will be soliciting opinion from the panel -- however it is composed. And therefore, as with any poll, there has to be some way to statistically account for error.

That is the fundamental flaw in thinking of this as in anyway definitive or "scientific." However interesting the results, they will still represent a polling of a particular group at a particular time and therfore the results will be unscientific.

If you performed this same test multiple times, the margin of error would be reduced, perhaps, but never eliminated.

Link to comment
Share on other sites

>We already have the only jury that really counts...ie that of top international performers and the >instruments they perform on & who could be better judges than these players.

I've been present at many of these play offs and have always found it a frustrating exercise. The average audience or even a very sophisticated audience can't seem to distinguish between a Strad and a good modern. Structuring a valid listening test is wickedly difficult. For instance, I've noticed that the 'context' of the listening experience can alter one's perception. If you've been listening to a series of bright sounding instruments and a very dark one is inserted, it stands out and is often perceived as better at first. My conclusion is that we, as instrument makers, must answer to the players and players in turn answer to the audience. Therefore I'm most interested in how players experience these instruments. The test that I would pay to see; get a number of good players and have them choose/rank a group of instruments that they play.

A working model:

1)a group of players ( mix of older/younger; plays old/new instrument; orchestral/chamber/soloist; deaf in one ear/ blind in the other )

2)selection of instruments old/new specifically the instruments normally used by the above group.

3)No time limit to the test.

4)Repeat the test three times over a period of weeks.

5)Part of the test should be to play the instrument in duet with a piano.

6)Change venues from stage of large hall to medium recital hall to living room

The Judges/Players should not know what they're playing (blindfolded)

They should not know the results until the end of the trial.

A work in progress........

Oded

Link to comment
Share on other sites

OK, I was getting too many votes to change back to the Corvette avatar, so I gave in. Now you know I'm a wimp.

Hulk Hogan is actually an intelligent, articulate guy when interviewed (the rest is just acting) but oh well.

I don't know if we'll ever come up with an acceptable format for "old versus new".

The biggest problem as I see it is this:

No past trials have been valid, because new instruments usually did very well, often embarrassing hideously expensive, highly revered old instruments.

No future trials, however meticulously planned can be valid, because the same thing might happen again! It's just too risky.

People seem to be hugely invested in their belief that the old Cremonese are vastly superior. It's a comfortable place to be, and I don't think they want anybody messing with it.

Me? Either way is OK with me. I'll just go wherever the evidence takes me. If I should need to admit that a half-million dollar instrument is a big step up from mine, I can't see it affecting my sales one bit.

The quality of past trials?

As Michael said (using himself as an example) there may have been a few people among the group of musicians voting in the past who were not qualified, but the event I was involved in did a lot of analysis. Spread sheets were generated showing all kinds of patterns and distributions. There was remarkable consistency between different voters. Sure, there were a few wild cards, and these odd voters might possibly have been enough to switch a second place instrument to a third in a 16 instrument field, but I don't think they were statistically significant enough to change the general outcome.

Link to comment
Share on other sites

Come on Falstaff, I'm sure you realize that any scientific test relies to a greater or lesser extent, sooner or later, on the obervation or interpretation of a human observer. So does this. If you want to call it a "poll", that's fine, but I don't think your position is strong enough to correct everyone else who might want to call it a "test".

Going back over your previous posts, you seem to most tenacious at wanting to invalidate any possible kind of test even before it happens, now to the point of even attempting to re-label it.

Link to comment
Share on other sites

Come on David, if you think these "trials" have in the past, or can in future, even with your generous criteria, qualify as scientifically controlled experiements as opposed to polls -- then so be it.

I don't see it.

Sure the trials of a scientific experiemnt contain human interpretation and observation. But they don't become DATA until they are replicable, and replicable by different observers over time. If they still rely upon the opinion of a particular scientist or group of scientists then they remain hypothesis, not proof.

It is sophistry to suggest otherwise. And I have seen or read notthing about this or other "trials" that would qualify them for true scientific method. They remain interesting -- and sometimes motivated -- polls.

As I've said before -- and has been echoed in other posts -- the closest thing we have to evidence in this question is the usage of performers over time. That is evidence from precedent. That has a statistical meaning. Still not DATA, but as valid, if not moreso, despite the placebo effect, as any "evidence" garnered by a particular trial.

Link to comment
Share on other sites

quote:


Originally posted by:
falstaff

Sure the trials of a scientific experiemnt contain human interpretation and observation. But they don't become DATA until they are replicable, and replicable by different observers over time.

So does this suggest that the pattern which has emerged as a result of different "new versus old" experiments has significance? Might even be called "data"?

quote:


Originally posted by:
falstaff

As I've said before -- and has been echoed in other posts -- the closest thing we have to evidence in this question is the usage of performers over time. That is evidence from precedent.

You've indicated that precedent has significance to you. But you feel that evidence from "violin testing" precedent should be ignored?

Tell us a little about yourself. Are you a statistical analyst, employed in the science field, a professional player, and amateur, a collector of rare instruments?

Link to comment
Share on other sites

quote:


Originally posted by:
Dr. Bearden Axelrod

How can anyone take seriously the pseudo-scientific ramblings of a "luthier" who thinks owning and driving a Corvette is still cool? Come on man!

It looks like this topic is always good for generating some hostility.

Would you care to go further into my psychological profile, my thoughts beyond what I've shared, or reasons for having owned many kinds of performance vehicles?

Another option might be to make a "pseudo-intellectual" response to specific points rather than "acting out".

Oh, by the way, it's not just any old "off the showroom floor" B flat Corvette.

Maybe you can have some fun with that.

Link to comment
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...
 Share

  • Recently Browsing   0 members

    No registered users viewing this page.




×
×
  • Create New...