wjm@whuxj.UUCP (MITCHELL) (02/09/84)
I cannot agree too strongly with those recent articles posted here that emphasized the need to listen frequently to LIVE music. I try to get to live concerts (mostly at Carnegie Hall and Lincoln Center in NYC) several times a month (I'd go more often if it wasn't as much of a trip into the city from out in Jersey) to recalibrate my idea of what music should sound like. To me, the optimal high fidelity reproduction system would give the listener the impression of being about 2/3 of the way back in the orchestra in a hall like Carnegie Hall or Symphony Hall in Boston - but that's my personal preference, just as my tastes in music tend to favor the classical side of things. However, I'd like to be able to change the settings to reproduce a smaller room for small chamber ensembles (like string trios and quartets) since I also enjoy listening to chamber music. So much for the ultimate ( I can dream, can't I ?? :-) ). Now the question is, given a finite budget and the constraints of technology, how close to it can we come? As I read the hi-fi magazines each month, I'm both pleased and upset. I'm pleased by the advances in technology that make it posisible to come closer to this goal, but I'm upset by the bad things that are done (like noisy recordings and multi-miking). Since the goal of high-fidelity is to accurately reproduce a live performance, one must listen to LIVE performances OFTEN to KNOW what it SHOULD sound like. As I've said many times, as an engineer, I feel that any distortion from the live sound is due to quantifiable parameters (although we may not know what these parameters are at the present time), but one can have a "gut feel" that SOMETHING is wrong without having to make a large series of lab tests, This feeling is generally caused by knowing that recording X does NOT sound like a live performance, and one can then go into the lab and find out why. As I've said before, certain digital recordings (Denon's in particular) do not sound like live music. At the present time, I don't know what parameter of their recording process is causing the problem (although I suspect it is due to their 44.1 MB sampling rate lopping off some harmonics, and the noise introduced by their 14 bit encoding). On the other hand, Telarc's recordings sound quite realistic. Bill Mitchell CSO Whippany, NJ (whuxj!wjm)
pmr@drufl.UUCP (Rastocny) (02/10/84)
Bill, There was an excellent article in the "Computer Music Journal," Vol. 7, No. 3, Fall 1983, by F. Richard Moore that described a good model for what you are talking about. It relates two loudspeakers to two holes cut into the walls of a small room that is within a much larger room. Basically, the only perceptions that you have of what is happening in the outer room is via these two holes (speakers). The article is quite lengthy and deals with several interresting topics on loudspeaker placement, virtual acoustic space, radiation vectors, early echo patterns, and global reverberation all related to Cmusic but totally applicable to stereophony. Good reading, Phil Rastocny