01.08.2010 Feynman on piano tuning
This article* (subscription required), from my good buddy Khurram, is a presentation and analysis of a letter from great theoretical physicist Richard Feynman to his piano tuner.
(Read about Feynman here. He's on Bill Gates' "Who I'd Have Dinner With" list— and my list too— and his books Surely You're Joking, Mr. Feynman! and What Do You Care What Other People Think? make for the very finest sort of excited-about-science general reading.)
In it, he outlines problems he's intuited about tuning a piano strictly to equal-temperament. As the strings get higher and therefore shorter, the string becomes relatively thicker. This causes its stiffness to play more of a role in its overtone compositions than at longer string lengths. After some calculations— simple once they're worked out, but impressive considering they're in a casually tossed-off note**— he suggests that a piano tuned by ear might sound more "accurate" than one strictly tuned by comparing frequencies.
Like any good scientist, he states the assumptions on which he's basing this hypothesis and asks his tuner for more information about (a) what strings he tunes, in what order, so as to better know the relationships between them, and (b) whether he in fact deviates from the theory to "tune by ear" in order to create a more pleasing result.
(As it turns out, there are several physical phenomena not intuited by Feynman, which means that piano tuners do "stretch" the top and bottom pitches of the piano as our ears start to need proportionally faster and faster vibrations in order to perceive the same rise in pitch.)
*Article (requires subscription/sign-in)
**Feynman did make a few "slips of the pen" in his formulae, which are cleared up in the subsequent analysis.
Posted by charlie williams at January 8, 2010 03:36 PM
