Sound Machine Groove

As reported in the NY Times, researchers at Lawrence Berkeley National Laboratory’s Physics Division are using technology originally designed to image sub-atomic particles to capture 3-D representations of the inscriptions etched on the surface of wax cylinders and shellac phonograph records; the researchers then model a virtual stylus as it travels through the grooves and are thus able to produce a digital proxy for the original sound recording. The results are remarkable, as the sample .wav files on the project’s home page attest. (Not incidentally, the resulting digital files are edited to eliminate scratches and noise: “’We thought these methods, which demand pattern recognition and noise suppression, could also analyze the grooved shapes in mechanical recordings,’ says [Carl] Haber.”) See also this press release.

Pretty amazing stuff that dovetails nicely with my own fixation on forms of digital inscription. Moreover, one of Kari’s students just recently built a freakin’ phonograph from scratch as a class project! And see also the Digital Needle for a homebrew effort along the same lines.

All of which reminds me of one of my favorite Nicholson Baker passages, from The Mezzanine:

If you made a negative of that image of my skate blade’s gorge, you would arrive at the magnified record groove—a hushed black river valley of asphaltic ripples soft enough to be impressed with the treads of your Vibram soles; an image cast from a master mold that was the result of a stylus forced to plow through wax as it negotiated complex mechanical compromises between all the various conceptually independent oscillations that stereophony demanded of it; ripples so interfingered and confused that only after a day with surveying equipment, pacing off distances and making calculations (your feet sparking static with each step) are you able to spray-paint “Bass Clarinet” with some confidence in orange on an intermittent flume of vinyl, as workers in Scotchgard vests spray-paint the road to indicate utility lines beneath. Cobblestone-sized particles of airborne dust, unlucky spores with rinds like coconuts, and big obsidian chunks of cigarette smoke are lodged here and there in the oddly echoless surface, and once in a while, a precious boulder of diamond, shorn somehow from the stylus by this softer surface, shines out from the slope, where it has been pounded deep into the material by later playings, sworn at by the listener as if it too were common dust. That was needle wear.

Back to the Lawrence Berkeley Labs project, the digital images themselves compell me too, for many of the same reasons I’ve writtten about here before. They’re just as arresting as we’ve come to expect from such visualizations—how I would love to here an art historian discuss the visual aesthetics of CGI!—but they also testify to the manner in which the epistemological divide between inference and direct observation grows always more porous by machines speaking to machines in tongues we can see but not hear.

Oh—anyone think they know where my title for this entry comes from?