[net.audio] Let's get on with it...

wagner@uw-june (Dave Wagner) (08/28/85)

> It would be nice if someone, someday said that measuring parameter YYYY
> translates to sound ZZZZ.  A hypothetical listing of this translation follows:
>  
> Measured Parameter            Sonic result
> 
> Damping Factor                Higher factors reduce muddiness
> Slew Rate                     Higher slews improve crispness
> Signal-to-noise ratio         Higher ratios improve low-level resolution
> Supply Regulation             Better regulation improves imaging
> Supply Ripple @ full power    Lower ripple improves focus
> Supply Impedance vs. Freq.    Linear impedances improve inner detail
> Chassis Leakage @ full power  Lower leakage improves low-level resolution
> Phase vs. Freq.               Low shifts improve inner detail
> 
It would also be nice if someone, someday said what "focus" and "inner detail"
mean!

            Dave Wagner
            University of Washington Comp Sci Department
            wagner@washington.arpa
            {ihnp4,decvax,ucbvax}!uw-beaver!uw-wally!wagner

mohler@drune.UUCP (MohlerDS) (08/28/85)

YES, LETS GET ON WITH IT...THAT WAS MY FIRST POINT BEFORE THE WRATH OF THE 
NET WAS APPLIED TO ME CONCERNING MY WIDELY MISUNDERSTOOD ARTICLE.

A small additional request, it also would be nice to see an article on
component interaction, an area I think is often totally ignored! 

			David S. Mohler
			AT&T - ISL @ Denver
			drune!mohler

pmr@drutx.UUCP (Rastocny) (09/03/85)

[Is this an analog or digital bug?]

There has been much talk and slander about subjective evaluations on the net.
Here are some of my rambling thoughts about this subject.

There are few areas in audio where knowing what measures differently and knowing
what sounds different are unique, like the differences between dynamic and
electrostatic loudspeakers.  Measurable differences exist but to quote the
statistics that pinpoint the audible or subjective claims is difficult.
Agreement upon which statistic clearly explains what is being heard has not
yet been properly correlated.  But yet these differences can be easily
substantiated by all who hear the two types of loudspeakers regardless of
reference SPL and associated electronics/signal sources (cotton, of course,
removed from the ears :-).

Since horn loudspeakers inherently have low harmonic distortion from low
diaphragm movement, and since horns and electrostats do NOT sound alike, there
must be some other (complex) parameter(s) affecting these transducers.
It would be nice if someone, someday said that measuring parameter YYYY
translates to sound ZZZZ.  A hypothetical listing of this translation follows:

Measured Parameter		Sonic result

Damping Factor			Higher factors reduce muddiness
Slew Rate			Higher slews improve crispness
Signal-to-noise ratio		Higher ratios improve low-level resolution
Supply Regulation		Better regulation improves imaging
Supply Ripple @ full power	Lower ripple improves focus
Supply Impedance vs. Freq.	Linear impedances improve inner detail
Chassis Leakage @ full power	Lower leakage improves low-level resolution
Phase vs. Freq.			Low shifts improve inner detail

The list goes on, but I think you get the idea.  After all, what good is it to
know that a piece of equipment has 0.000001% THD @1KHz unless you know what the
audible effects of such minute distortion are.  On the other hand, is lower
necessarily better?  That is, can you hear the difference between a 10E-6% and
a 10E-7% THD circuit?  If the 10E-6 equipment sounds more like the real thing,
what other factors are involved in why equipment sounds different?  (Slew rates
come to mind where some low-slew amps sounded more accurate than high-slew amps
with the same THD spec.  It turned out that a slew-induced distortion parameter
(SMPTE) was involved but not currently being measured.  As soon as it was
discovered, it was remedied, but not until people started hearing strange
things from these high-slew amplifiers.)

If THD, the long-honored spec of an amplifier's performance, no longer tells us
anything about the amplifier's sound, why bother measuring it?  If phase
linearity and supply impedance tell more about an amplifier's sound, why NOT
measure and publish these specs instead?

The bottom line is that an amplifier should neither add nor subtract ANYTHING
from the signal (wire with gain), and that the effects of the loads (input or
output) not influence each other or the amplifier's performance.  But in reality
loads do alter an amplifier's performance and amplifiers do inherently add and
subtract things.  Determining what to measure to reduce the audible effects of
this signal modification should be of utmost concern to designers and engineers.
But, as the old saying goes, the squeaky wheel gets the grease and until people
start complaining about something, nothing will be done.  (SMPTE was complained
about and then measured.)

I for one cannot rely on the present meaningless measurements or the current
sales hype to explain why I hear (or don't hear) different things in different
amplifiers, CD players, phono cartridges, interconnect cables, loudspeakers,
etc.  I hope that some day someone completes a thorough and conclusive study
of the sonic signature of circuit topologies and publishes these results in
JAES.  Then, maybe, we could get away from this senseless bickering and get on
with understanding and improving the state of the art.

		Yours for higher fidelity,
		Phil Rastocny
		AT&T-ISL
		ihnp4!drutx!pmr