Re: [ARSCLIST] A fundamental Flaw: Was Sampling Theory (was Fred Layn's post on the Studer list re: Quantegy)

[Mike Richter <mrichter@xxxxxxx>]

At 10:51 AM 1/21/2005 -0500, Steven Smolian wrote:
> Im I correct in recalling that the ear hears frequency distribution
> differently at different levels?  If so, does this affect this thread?

You're neither quite correct nor substantially incorrect.

There was a set of standard curves published many years ago on the matter,
showing substantial falloff in highs and lows as level decreased. However,
more recent work (sorry, I have no references) indicate that those are
substantially in error and that the actual response does not vary nearly as
much as believed.

The original work was done in 1933 by Fletcher and Munson (Journal of the
Acoustic Society of America, 5.2) and the Fletcher-Munson curves were the
standard, showing more than 60 db falloff at 25 Hz relative to 1 KHz at 0
dbm; at 110 dbm, sensitivity was essentially flat from 25-1000 Hz. While
they also showed variation from 1 KHz to 15 KHz, that was virtually
independent of amplitude; any 'error' would therefore be the same for live
and recorded sound.

The recent work has shown the low-end variation to be much less than
measured in 1933; details are no doubt available with research for which I
lack the time.

Getting to the question, any correction would be bass enhancement at lower
levels. IMHO, that is the sort of thing the user typically would do for her
own taste regardless of theory and measurement, so would be irrelevant here.

(Reference is the old standby: Radiotron Designer's Handbook, in this case
the fourth edition.)

Mike
--
mrichter@xxxxxxx
http://www.mrichter.com/