Slashdot Mirror
Linux PCs Drive 74-Channel Pipe Organ
cyberman11 writes "According to the EE Times, Marshall & Ogletree LLC have created an electronic simulation of a classic Aeolian-Skinner pipe organ in the Trinity Church situated, just 600 feet from ground zero near the World Trade Center site in New York. The system consists of 10 Linux PCs that drive 74 Carver amplifiers and 74 Definitive Technology speakers, for a total of 15,000 watts."
The article raised more questions than it answered, for me. The part I don't understand is why so many channels are necessary - any loudspeaker will produce polyphony (when cell phones claim to have polyphonic ring tones, it has more to do with the tone generation circuit or software than with the speaker). Maybe someone who knows more about acoustics than I do can answer this one: what is meant by a "massive amount of polyphony"? More frequency content in the spectrum? And are the 74 Epiphany channels matched to 74 original pipes? Does each speaker play only one tone?
... but c'mon, my Helmut Walcha CD's don't sound THAT bad, do they?
Finally, does this also mean that recordings of organ music are poor substitutes for the real thing, since they will be played only on stereo speakers, which are presumably capable of "less" polyphony? I am sure that many organ zeolots have been saying all along that there is no substitute for live performance
eikimartinson.com
I *am* an organist... and I've played some very, very good electronic instruments, but none have exactly modeled the experience of a real instrument, and it's not because of any large lapse of sound quality or discrepency in the samples or production.
There are a combination of things that, added up, definitely detract from the unique experience of a well-built pipe organ. Often, the electronic instruments do not accurately model how a pipe speaks -- only the tone once a pipe is speaking. Also, there's a difference in the response/attack of reed pipes, flute pipes, principal pipes, etc. -- the electronic instrument often models the sound accurately, but doesn't capture the actual 'feel' of the sound, and the performer would overcompensate.
This makes it difficult both for the listener, who will notice a difference since the electronic instrument is probably not voiced in the same way as an acoustic instrument (which is specific to the room in which the instrument is built). Also, the performer may not be comfortable with playing his Bach on a non-mechincal (or electropneumatic) instrument, and this would contribute to the feeling of unnatural-ness. (Maybe we, as performers, just haven't found a good way to deal with the actual articulation/technique problems on electronic insturments.)