First, a couple items of full disclosure: This post was not timed for the Grammies aftermath (congratulations, Adele), but was related to a fascinating Feb. 12 posting on Gizmag (http://www.gizmag.com/gilmore-self-tuning-piano-system/21425/).  Second, I freely admit to going to small, oddball electronic music concerts in seedy bars or art galleries, where the musician will haul out multiple circuit boards for sustain, reverb, and feedback effects.  As pretty as those boards may be to look at on stage, I’ve always wondered why more musicians don’t talk to the engineering community about employing advanced VLSI programming tricks in their show.  No, I’m not talking about pulling out an Apple Macbook for some software effects, of the type that Deadmau5 and Girl Talk use in dance numbers like the Grammies extravaganza.  I’m talking deep hardware programming in a musical act.

Kansas City mechanical engineer Don Gilmore may come from the classical music community, but he’s looking at new ideas for instrument tuning, that could revolutionize even the regular manipulation of sound in jazz, electronic, improvisational, and indie rock music.  The Gizmag article on Gilmore’s work mentioned an effort by Gibson to offer self-tuned guitars like Firebird X, which has been slow to win acceptance because of the number of guitarists that love their tuning pegs.  Gilmore has faced some of the same skeptics in the music industry – he had an advocate in the QRS Music company for a while, but is facing an uphill struggle in integrating thermal tuning based on a series of IR sensors.

Gilmore had been goofing around with software and a modified EBow (an electronic bow for guitar or violin), when he realized that a DC power supply could be used to change the pitch of an individual string.  That’s all well and good for a typical stringed instrument, but a piano has 88 keys controlling more than 200 strings.  Each of those strings requires its own sustainer module, based on the use of IR sensors, since the magnetic coils used in tuning smaller instruments would never work in a string atmosphere as dense as a grand piano.  (Are any of you picturing Terry Riley’s bowed piano experiments, or am I getting too weird here?)

Gilmore’s system uses a master microprocessor and a high-speed counter connected to a 100-MHz clock.  The processor compares the pitch of each string to the stored pitch of a perfectly tuned piano.  An FPGA is used to handle the massive amount of I/O involved in tuning all the strings, with each string having its own dedicated counter and evaluator circuit.  A piano can be fully tuned using this logic-driven thermal system, in less than a minute.  Gilmore mentioned that even FPGAs are pin-limited, so he developed a series of shift registers to shift a 219-bit word from a serial line in the FPGA, providing 219 independent pulse-width modulated duty cycles.

Now Gilmore is a musical traditionalist, so he is interested in an FPGA that is fully characterized for the proper tuning of a piano or other instrument.  But what if we start using the reprogrammability features of FPGAs to play devious games?  What if instruments can shift instantaneously to different tunings or de-tuned states?  Sure, many guitarists love their handheld capo devices that change tunings, but radical tuning shifts could have as profound an effect on modern music as the invention of the theremin (the instrument that provides the spacey sounds of the Star Trek theme or the Beach Boys’ ‘Good Vibrations’).

Many of you may shudder at the notion of a deliberately de-tuned instrument.  You’re probably the type of customer that might scream in a record store, “That’s not music, that’s noise!”  But noise can be big business in the music industry.  Just two weeks ago, experimental composer Pauline Oliveros won a $50,000 John Cage Award (http://www.newmusicbox.org/articles/pauline-oliveros-winner-of-50000-john-cage-award) for her ground-breaking work on de-tuned concertinas, banjos, and other instruments.  Can you imagine what she might do if she got her hands on a series of IR sensors controlled by FPGAs?  I’m not sure the Grammies will be ready for the results.