[thelist] audio hardware question

[Bill Haenel]

> > with the source audio, especially high-end distortion that isn't so
> > audible in the source, the encoding process for lower bandwidths
> > exposes the problems and magnifies them in the form of high-end
>
> not contradictory, no... i was talking more about flat audio and the
> like... for instance, your crispness is gonna disappear at lower
> sample rates and the like...
>
> however, i'm curious about your compression methods or
> algorithms... i've experienced what you describe, but it hasn't been
> as much of a factor for me, perhaps simply because of my
> expectations...

Well, at this time we're using audio that's compressed for on-air broadcast.
What we're finding is that our compression for on-air is NOT optimal for
webcast, as it's dynamic range is a bit too broad. We're shopping for
separate equipment to process our web audio, then we'll run that to our
encoder. In spite of all of this, our audio (we run two streams: 32k WM and
32k Real) is some of the best low-bandwidth stuff I've heard. One of the
secrets (shhh - don't tell anyone!) is encoding in mono - one channel, but
twice as much data for that one channel as you'd have for each of the
separate stereo channels (sort of). Of course, you don't get the stereo
effect, but overall the audio is much better. We're currently running our
Real stream in stereo as an experiment. But the way it's sounding, until
encoding techniques improve significantly, we'll probably run both of our
low-bandwidth streams in mono. Please feel free to have a listen (shameless
plug): http://www.publicbroadcasting.net/ncpr/ppr.pprmain.

BH