|
Propeller Head Plaza Technical and scientific discussion of amps, cables and other topics. |
|
|
|
Propeller Head Plaza Technical and scientific discussion of amps, cables and other topics. |
|
In Reply to: Here's my favorite line: posted by clarkjohnsen on April 13, 2007 at 11:53:27:
"Science has not provided any convincing evidence for the existence . . . of any measurable property that would set the Cremonese instruments apart from the finest violins made by skilled craftsman today.""That's interesting Clark because I have measured Three such instruments with a simple microphone (but good one) into a 1/6th octave spectrum analyzer and LO and behold all three gave a different harmonic spectrum. So at least the output can be seen graphically to be different. Now what I was not able to do was to compare TWO Stradivari violins to see how similar their harmonic spectra are. Unfortunately, I had only one Strad, one Guarneri and one Guad (not from cremona but similar time period...early 1700s). We also didn't have a modern one handy but my guess is that it would also measure quite differently.
My approach to this problem would be as follows:
1)Assemble many examples of several types of violins (the bigger the database the better)
2)Have the same person play each one with a single note or two and collect a high resolution harmonic spectrum.
3) Repeat this with each violin several times to account for slight variations due to how the instrument is being played.
4) Analyze the spectra using PCA (principle component analysis), cluster analysis and/or a neural network algorithm. These are computational tools for taking complex data sets and finding distinctive features in them. An example of what these tools can do would be to determine what region in the world marijuana is grown based on GCMS (gas chromatography mass spectrometry) of extracts from various known plants and then classifying an unknown.
5) Classify instruments based on these results (ie. new, old, cremona etc.)
6) Bring in a new instrument and an old one that were not part of the "training" sets.
7) See if the computer can correctly determine if it is "old" or "new" or if it can perhaps determine the make (this would require a pretty large database).8) Use the database and computational tools in conjunction with violin makers to make modern violins that result in spectra that mimic classic old violins. Run the new spectra through the tools to see if the new violin matches "new" or "old". Repeat until satisfied.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Here's my favorite line: - morricab 02:27:03 04/17/07 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
