Sign in to follow this  
Joaquín Fonollosa

Geometría fabrorum

Recommended Posts

On 10/12/2019 at 10:37 AM, David Beard said:

Once you understand the recipes that were traditional adhered to in each part, the whole design can be 'read'.

 

And we can reverse-read a "recipe" into almost anything, if we try hard enough. ;)

Share this post


Link to post
Share on other sites
13 minutes ago, Three13 said:

It seems to me that the easiest way to resolve whether the circle/ratio theory is correct would be to take a random selection of Cremonese instruments and later copies of Cremonese instruments, carefully copy their outlines, purfling lines, and soundhole locations, then subject the images to analysis. The Cremonese instruments should - theoretically - all correspond to circle/ratio geometry, while the copies should tend not to (obviously, the possibility exists that a traced example would coincide).

It would require a bit of work, but perhaps it would put an end to the repetitious cycle we seem to have here...

I've done this with over 90 examples.

However, I never introduce an extra step of tracing.  This only further obscurs the question.  I work with images of the instruments. The best images I can find.

I always had an instinct to work directly from images instead of tracing of unknown additional imperfection.  But the leason was driven home in a thread with Addie regarding the Bergonzi form.  Addie has made all these wonderful very careful tracings of the historical molds.  I didn't possess any good enough photos of the Bergonzi mold.  So, I tried to analyze the design from his tracing.  But the upper bouts did not seem to follow the normal geometry recipes seen in ALL OTHER OLD CREMONA examples.  So these appeared to be a  counter example, breaking principles found consistently in Cremona work.  So I contacted Addie and he sent me the very high quality image he had done the tracing from.  Lo and behold, the upper bout was in fact completely typical Cremona work.  There was some slight difference from photo to trace the  completely obscured the actual shape structure!

So I learned that lesson.  Be cautious with photos.  Double check them for true square perspective in all ways you can.  But remember that drawings are simply worthless to analyze shape structure. Drawings are never truly accurate anymore than photos are.  But drawings are ultimately worse because the distortions are human made in ways the defy detecting or understanding the deviations.  In contrast, the distortions in photos occur from the physics of the process.  Photo distortions can be detected and understood.

 

*******

Further labors by others to access my assertions will ultimately bring more people around to accepting the propositions.

But I have already done these labors.  

We live in a time of 'alternative truths' and 'facts resistance'.    

These are ideas that are new, and radical in terms of the status quo.  Anyone that will take the trouble can see what I'm talking about.  You can just walk along the trail I've cut.  Or you can explore the territory independently for yourself.  Just bring a machete in that case.

The problem is that most people either just believe ideas that are staus quo already.  Or they are less discplined and believe things as fantasies built from desires.

Either way, very few are going to adopt this idea before it grows to become statis quo.

 

Share this post


Link to post
Share on other sites

The only reason I suggested tracings would be so that the person analyzing the image wouldn’t be able to recognize an instrument. Perhaps there would be another way of masking cues.

Share this post


Link to post
Share on other sites

I regret not being able to offer you, for the moment, a compendium of the design system that I propose. However, I think I already gave them the necessary elements. I invite you to take it to practice to check its effectiveness. And to improve it with their own knowledge. I do not have the possibility to make a violin with my system, because I do not dedicate myself professionally to the luthería, I am only a big fan of this trade. So I urge you, if you think so, to build a violin whose mold has been designed with my system, and thus check if it works and is aesthetically attractive. Try to get all the information that the double spiral template can provide. I have come to obtain the measurements of all the elements and their position. If you decide to carry it out, stick to the method, to the lines provided by the template, straight, curved and tangent to curves. If I could, I would do it myself.

IMG_20191012_203242.jpg

Share this post


Link to post
Share on other sites
42 minutes ago, Three13 said:

The only reason I suggested tracings would be so that the person analyzing the image wouldn’t be able to recognize an instrument. Perhaps there would be another way of masking cues.

 

 

 

Why does that matter?

There are enough example instruments.

Make very good photos (square and true proportions) of a Cremona example not analyzed before.  Give the photos to three people that have learned the recipe and the choice ranges.

See if all three 'read' the same recipe choices.

*******

My initially research has been limited by budget and physical access.  I used the best images I could find on the web.  Also, my work was mostly in mode of 'discovery' rather than in a mode of proving on demonstrating to win over resistance.

The next step would be to more rigorously test out my propositions.

******

My original purpose was merely to understand the old shapes well enough to be able to make by the principle of 'do as they did'.   Along the way, I stumbled on to this whole thing of finding highly consistent 'recipes' across all the Old Cremona examples.  And arriving at the hypothesis that these 'recipes' were worked interactively with the building process.  And then lastly, I stumbled into the idea that the 'choices' presented by these 'recipes' are structured in a 'choose one of these traditional options' way.  And that this amounts to an 'encoding' of the whole design.  This sequence of choices is then 'readable', 'repeatable', and 'variable'.   All of this then raises the possiblity that this 'sequence of design choices' might actually act as the 'DNA' in a cultural mechanism of 'design by evolution'.  Indeed, these things might be the mechanism behind the 'design evilution' the 2015 MIT study pointed to.

All of these things can be framed as clear enough hypotheses to be tested and examined.

 

I have probed these ideas enough to satisfy my own needs.  And I am running short of deaire to continue giving the majority of my efforts in this direction.  But I do remain interested in helping these new ideas become public knowledge, and in seeing others get involved in testing the premises and carrying historical investigations using these tools further.

 

Share this post


Link to post
Share on other sites
29 minutes ago, Joaquín Fonollosa said:

I regret not being able to offer you, for the moment, a compendium of the design system that I propose. However, I think I already gave them the necessary elements. I invite you to take it to practice to check its effectiveness. And to improve it with their own knowledge. I do not have the possibility to make a violin with my system, because I do not dedicate myself professionally to the luthería, I am only a big fan of this trade. So I urge you, if you think so, to build a violin whose mold has been designed with my system, and thus check if it works and is aesthetically attractive. Try to get all the information that the double spiral template can provide. I have come to obtain the measurements of all the elements and their position. If you decide to carry it out, stick to the method, to the lines provided by the template, straight, curved and tangent to curves. If I could, I would do it myself.

IMG_20191012_203242.jpg

Joaquin, you aren't proposing a violin design system.  You're making intriguing pictures!

Every one of your posted images shows us some different 'system'.  Which one do you mean?  And all your images are full of extraneous visual elements.

Please!!!   Take pride in your artistic talent!!

Forget the fantasy stew of 'violin design systems'.   

Share this post


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

Quite the opposite.

I was asking if he was of of at least above average IQ or even more.

 

 

I do not know the IQ of A Amati (and I admit that this subject is of no interest to me) but if you admit that he knew how to count until 12 it is more than enough to make a violin.

We are very far from complex mathematics ....

 

Share this post


Link to post
Share on other sites
24 minutes ago, francoisdenis said:

You seem to be unaware that this work has and continues to be done...

I’m aware of the work that’s been done, I’m merely suggesting a public demonstration of this work using both Cremonese and non-Cremonese imitations to illustrate the validity of the theory.

Share this post


Link to post
Share on other sites
27 minutes ago, francoisdenis said:

I do not know the IQ of A Amati (and I admit that this subject is of no interest to me) but if you admit that he knew how to count until 12 it is more than enough to make a violin.

We are very far from complex mathematics ....

 

OK, so its more about knowing a method like yours ?

https://www.youtube.com/watch?v=7KwaoCLqXa0

Share this post


Link to post
Share on other sites
5 hours ago, Michael Darnton said:

If an outstanding question here is whether Stradivari's outline was built from circles or spirals, this question has a definite answer, there is software that can answer it, and I have thrown the software at the question. The answer is circles, WITHOUT A DOUBT.

What software did you use, and which of the variable input parameters did you use?

A spiral can be described as some combination of circles of varying radii, so I'm not sure how trying to make a distinction between the two is useful, without a lot more information and specifics

Share this post


Link to post
Share on other sites
15 minutes ago, David Burgess said:
5 hours ago, Michael Darnton said:

If an outstanding question here is whether Stradivari's outline was built from circles or spirals, this question has a definite answer, there is software that can answer it, and I have thrown the software at the question. The answer is circles, WITHOUT A DOUBT.

What software did you use, and which of the variable input parameters did you use?

...

I would also be keen to know, if you don't mind sharing.

But most of all, I would be highly interested to know how the resulting approximation(s) are being quantitatively assessed - the lack of an objective, quantifiable metric against which the different methods can be unambiguously tested is a glaring omission in this rather circular thread (...or should that be spiral?).

Share this post


Link to post
Share on other sites

We are not talking about planetary motion.  You can not prove the issue in any direction by digging in for accuracy.

I have used guess work, initution, other people's work, my work, and stubbornness to cobble together a complicated hypothesis that proposes:

Classical Cremona makers to man worked the sizes, positions, and shapes using traditional recipes of ratios and dividers and straight edge work, and in particular traditional ways.

It is not going to be possible to prove this in any absolute way, as there are many sources of deviations from ideal design in the historical work including:

*) The original work is quite loose and inaccurate

*) They actual followed from and built upon the deviations as the work continued

*) Wood materials move and wear

*) Repairs have altered things

*) Images are imperfect 

 

What can be done is to test each proposed recipe against historical examples.

Are there any counter examples that disagree with the proposed recipe within an accuracy appropriate to the example?

This is an answerable question.  I have examined many cases and not found such counter examples.

Does the accuracy of the proposed recipe shapes fot to the examples improve when the examples instrument and image are more detailed and clean?

This is a question that can be answered.  Yes.  Clean examples, clean fits.

Are there predictive aspects of the hypothesis?  Yes.  The recipe present only very limited options, and predict tjat the best fit will be found within those limited options.

After seeing the same recipes produce the best fits time after time after time, confidence grows.

 

But, there's lots of room for doubters to dance about.  With time however, the mountains of supporting data are just going to grow.  

Share this post


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

These are the tools necessary to design a violin mold. The measurement of the side of the initial square can be varied a little +/- 2mm.

IMG_20191013_234101.jpg

IMG_20191013_233830.jpg

IMG_20191013_234320.jpg

Good!!!!!   Now that shows a definite method to making your spiral, based on Phi.

Now.  What definite method to produce a new mold from that?  Not to trace a mold.  How to you use these tools you show and a blank piece of paper or plank of wood to create a new mold?

 

Share this post


Link to post
Share on other sites
4 hours ago, Three13 said:

And that wouldn’t be entertaining?

well nobody said that, I think we could make a novel out of it, Lord of the F hole or something like that. :lol:

Share this post


Link to post
Share on other sites
6 minutes ago, jezzupe said:

well nobody said that, I think we could make a novel out of it, Lord of the F hole or something like that. :lol:

If you called it that, I suspect a lot of people buying the book based entirely upon the title would be disappointed.

Share this post


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

What software did you use, and which of the variable input parameters did you use?

A spiral can be described as some combination of circles of varying radii, so I'm not sure how trying to make a distinction between the two is useful, without a lot more information and specifics

Not to speak for Mr.Darton, but it seems to me the distinctions usefulness is that it would expose that the "simplicity" of a compass was all that would be needed for design, and that spirals would be more time consuming and fussy to work with, and or the most obvious answer is the right one....? but I don't know what Mr. Darton exact reason for the statement...but I'm sure we all want to know.

enter popcorn eating emocon "here" 

Share this post


Link to post
Share on other sites
1 minute ago, Three13 said:

If you called it that, I suspect a lot of people buying the book based entirely upon the title would be disappointed.

We don't care if they're disappointed, as long as they buy the book, hey, we can't help it everyones mind is in the gutter.

Share this post


Link to post
Share on other sites

The software I used didn't work on large spans like someone trying to figure it out with with a compass or set of circles; it measures the curve of small segments and calculates them individually. It has been a while and I don't remember exactly, but I believe the segments I was using were about 1/2 inch long at the most, smaller in some areas, but it wasn't necessarry to drop the resolution beyond that to get a clear picture of what was going on.

Changes in circles on the first Strad mold I checked were clear and abupt, there was one section on the outline that was obviously hand drawn, and the c-bouts were a bit of a mess, especially in the armpits and mirroring spot at the bottom of the Cs. If spirals or similar were used one would expect irregularity comparing distant regions, but that was not the case. Stradivari's mold-making precision was remarkable, with the circles of the original drawing precisely preserved from the point where they started to the point where they ended. I am hoping that readers are clever enough to realize that if most of a bout consists of small segments of X radius, and the large circle also matches that, you are looking at a circle, not a French curve or spiral.

As anyone who has made a violin probably realizes, such a measuring strategy wouldn't work on a real violin, especially a worn old Cremonese one, due to the outline being several generations removed from the drawing, and executed in a way that would tend not to preserve the circles precisely beyond assuring that the curves flowed and looked OK, with wear and abuse on top of that. Fortunately, we don't need to measure violins--the molds are available, and Francois Denis thoughtfully provided his tracings of them on his own site. There are also several sources for good photos of them,  Francois' book and Stewart Pollens' also.

Since the idea and technique was by way of a private present from someone else, it's not mine to share and I will let people do their own research for a method.

Share this post


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

Good!!!!!   Now that shows a definite method to making your spiral, based on Phi.

Now.  What definite method to produce a new mold from that?  Not to trace a mold.  How to you use these tools you show and a blank piece of paper or plank of wood to create a new mold?

 

The rectangle has to be necessarily golden to be able to trace the spiral. Phi appears in the relationship between its sides. Nothing else.

IMG_20191014_004214.jpg

IMG_20190801_140333-01-01-02.jpeg

IMG_20191014_004724.jpg

IMG_20191014_004841.jpg

IMG_20191014_005231.jpg

IMG_20191014_004922.jpg

IMG_20191014_005404.jpg

IMG_20191003_145834.jpg

IMG_20191003_152934.jpg

IMG_20191003_145146.jpg

IMG_20191001_104107-01.jpeg

Share this post


Link to post
Share on other sites

Just out of curiosity--has anybody done outline shape (all scaled to the same length) overlays of violins starting from Andrea Amati to DG to see how much the violin's shape had changed?

Share this post


Link to post
Share on other sites

Marty, I have dozens--maybe hundreds--of violin outlines I have taken over the years. If you stack them on a window (budget light table) you can see that the curves and locations of important spots are very consistent over the whole period. Most variation, even within a single maker, is in the exact shape of the blocks. While overall dimensions vary a bit and make you think they are different, one doesn't realize how much the Cremonese have in common until you throw a Gagliano, or Venetian, or anything else on the stack.

I think that Francois did a very good job of explaining the overarching ideas of the overall layout, which other schools of that time very obviously didn't know the smallest thing about.

Share this post


Link to post
Share on other sites
37 minutes ago, Marty Kasprzyk said:

Just out of curiosity--has anybody done outline shape (all scaled to the same length) overlays of violins starting from Andrea Amati to DG to see how much the violin's shape had changed?

Some very nerdy scientists did a shape morpholgy 'relatedness' study some years back.

And, of course, I've done analysis of multiple example instruments from all the main makers and generations of clasical Cremona work.  So yes.  The outlines don't remain static at all.  But the basic recipes used and options presented changed very little and very slowly, or not at all for many recipes.  What did change however is combinations and preferences seen in the choices they make using the recipes.  These choices control the structure and shape.  And the patterns of their choices do develop and change over time.

Share this post


Link to post
Share on other sites
18 minutes ago, Michael Darnton said:

Marty, I have dozens--maybe hundreds--of violin outlines I have taken over the years. If you stack them on a window (budget light table) you can see that the curves and locations of principle spots are very consistent over the whole period. While overall dimensions vary a bit and make you think they are different, one doesn't realize how much the Cremonese have in common until you throw a Gagliano, or Venetian, or anything else on the stack.

Yes. Quite similar.  But there are differences.  They certainly did not just statical use one shape.

However, there are some.long stretches in Brothers Amati and Nicolo Amati where the recipe choices used get quite static for long stretches of instruments.

Share this post


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...
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.