This was a spin off from the Freestyle Violin project. These tests took place at the Oberlin Violinmakers Workshop in 2023. Many of the violinmakers at the workshop had brought instruments with them to show their colleagues. Thinking about violin design, one of the Freestyle group, Boris Haug, wondered if we could predict a violin’s tone from its appearance.
First Test
We collected together 10 violins which we set out on a table and asked participants (all professional violinmakers) to predict the “brightness” of the tone of the instrument based on its design features and the materials used. Without hearing the instruments the participants assigned a number to each one, ranking them as follows:
- = Dark
- = Neutral
- = Bright
Once these predictions had been made, the instruments were played in random order and participants were asked to use the same number system to denote their actual perceived brightness. Those numbers were averaged and are shown below.
First test Results. For the most part the sound tests confirmed the participants predictions with the predicted bright fiddles being perceived to sound on the brighter side. There were three notable exceptions Violin C was predicted to sound the brightest of all the violins yet it the sound test it was perceived to be amongst the darker violins. The story was the opposite for fiddles G and H.
Violin C was the “violinabox” constructed from non-traditional materials and having a very heavy back and a very light top, These materials seemed to have overwhelmed the design features of the fiddle.
Second Test
Commenting on the First experiment, some people felt that the size of the room, which was very small and full of people, might have had a big impact on the way that the sound was perceived. So it was decided to repeat the playing part of the experiment, using the same instruments in a concert hall. In addition it was felt that “bright and dark” were rather ambiguous terms, so in addition to grading for brightness, participants were also asked to rate the “loudness” and the “richness” of the violins, using the same numbering system
Second test results. In general the larger hall seems to have mellowed the differences in perceived brightness. There were fewer extreme numbers. Violins C still switched from predicted bright to perceived dark but not as radically, and the story was the inverse for violin G. Violin H which had been perceived as the brightest sounding violin in the small room, mellowed to be the second darkest in the large hall.
In general the Loudness and Richness ratings tracked the Brightness ratings in the hall. The most intriguing instrument was violin H, the one that turned from bright in the small room to dark in the main hall. It was also voted the loudest violin.
| Instrument | predicted brightness | Perceived brightness – small room | Perceived brightness – Hall | Loudness – Hall | Richness – Hall |
| A | 1.95 | 2.17 | 2.04 | 2.04 | 2.26 |
| B | 1.47 | 1.66 | 2.09 | 2.09 | 2.17 |
| C (Violinabox) | 2.59 | 1.46 | 1.84 | 2.2 | 2.0 |
| D | 2.43 | 2.21 | 1.95 | 2.09 | 2.09 |
| E | 1.82 | 1.16 | 2.17 | 2.26 | 1.82 |
| F | 1.05 | 1.86 | 2.13 | 2.33 | 2.21 |
| G | 1.68 | 2.21 | 2.18 | 2.23 | 2.19 |
| H | 1.86 | 2.74 | 1.91 | 2.41 | 2.55 |
| I | 2.45 | 2.6 | 2.0 | 2.04 | 1.95 |
| J | 2.0 | 2.08 | 2.04 | 2.1 | 2.27 |
| Mean | 1.93 | 2.01 | 2.03 | 2.22 | 2.15 |
