[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [microsound] mp3recorder



>From: derek holzer <derek@xxxxxxx>
>looking at the spectrographic display of the frequencies, i got a pretty 
>good picture of where the MP3 encoding scheme drops frequencies... you see 
>almost nothing in the high end [the "air"], and then in lower, more 
>"audible" ranges you see that there are certain bands that it cuts, and 
>certain bands that it leaves in. these seem to be pretty logarithmically 
>related, leading me to beleive they had musical harmonics in mind, and not 
>"noisy" broadband sounds [like field recordings] or inharmonic frequencies 
>[like a lot of microsound].

Supposedly MP3 encoding is based on psychoacoustic models so it should be 
related to human hearing rather than an arbitrary selection of musical 
harmonics.  The principle is based on frequency masking and it's a dynamic 
process so just because certain bands appear to be cut on a particular 
recording doesn't mean those same bands would be cut on different source 
material.  Because of this you can't really tell anything about mp3 encoding 
by looking at a spectrograph

>conclusion: not the best choice if you do field recording, make inharmonic 
>music, or want a full range of the sonic spectrum to use for further 
>transformations. the artifacts of MP3 will show up very quickly once you 
>start playing with these sounds digitally.

I agree that a lossy format like mp3 or minidisc is never the best choice 
for ANY recording but it shouldn't have anything specific to do with the 
type of music or sound you're recording.  It's just bad because it's lossy.  
At the same time those formats are good when cost, size and convienience are 
the deciding factors.

_________________________________________________________________
MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*.  
http://join.msn.com/?page=features/virus

------------------------------