Marty Michener wrote:
> Raimund (and before Walt gets going again):
>
> We have discussed this almost to death previously on this group for over=
> two years.
>
> We who have been using ATRAC 4 to record, very often quiet sounds nearly=
> masked by loud sounds, have been later able to filter out the loud sounds=
> (like the usual rooster crowing near the mic) and have been pleased to fi=
nd
> our quiet sounds still there as if recorded by DAT or R-R tape. So we ke=
ep
> doing - using MD.
Actual recording, actual experience is what really counts. We do it
because it works.
> But, there is much evidence that human ears use arrival time differences=
> down to even a few microseconds (see my references previously to
> Batteau). Again, this is the "nth" go around for many of us on this
> group. I claim digital samp[ling in general does a lousy job at preservi=
ng
> arrival times accurately on sibilant sounds or sharp ringing sounds like=
> rain spattering or jingling keys. I disagree with Walt on this one, I
> think it IS worth using much higher rates, as you and Dr Kaernbach have
> suggested, especially for acoustical research.
I don't think we disagree as much as you think. Going to higher sampling
rates is sometimes necessary. I do think that one should try the regular
stuff first and not assume it won't work, which is all I'm saying. A lot
of acoustical research consumes far more money than it needs to if some
simple tests were done to find out just what's really needed.
Currently there are devices that can sample sound at rates as high as
750kHz, maybe higher. Should we use them? If needed, yes, if not, there
may be alternates that will save funds.
Science has so little funding nowadays I hate to see any money wasted.
The other part of it is if you do nothing but use the high rates this
can lead to conclusions that the ultra fine details are what mattered.
And this may not be the case at all. We see this in many areas nowadays.
We have developed techniques that can do extremely fine measurements.
Because we can make such fine measurements, we often give them
significance they don't deserve and use them excessively. In laws, in
medical diagnosis, in all sorts of ways this is costing all of us a lot,
often unnecessary cost, often damaging results. Everywhere we need to
back off and ask if this is really doing what we think it is.
Walt
________________________________________________________________________
________________________________________________________________________
|