Re: [ARSCLIST] Listening Test - was: Sampling Theory (was Fred Layn's post on the Studer list re: Quantegy)

[Eric Jacobs <ericj@xxxxxxxxxxxxxxxxxxx>]

I think a worthy test that everyone should perform at least
once in their audio lifetime, but better if you can do it
with each jump in audio recording and playback technology:

A.  Get an analog function generator and connect it to your
    all-analog monitoring system and A-B different waveforms
    at different frequencies to see what you personally can
    and cannot hear.

B.  Repeat the same test, but pump it through your ADC-DAC
    chain instead at various sample rates and word lengths.
    At a minimum, try 16/44.1 versus 24/88.2.

I know an analog function generator does not produce music,
but it does produce a very repeatable signal for basic
listening tests, and as it is analog, it contains all the
harmonics that can affect timbre.

How the results of these tests with a function generator
extrapolate to actual music can be debated, but what you hear
in these tests with your ears, in your environment and with
your equipment is undebatable.

I just did the test this evening, dragging out my trusty
HP 3312A Function Generator, and then listening to sine,
triangle and square waves.  I validated the waveforms on
my Tektronix 2465A 350 MHz 4-channel oscilloscope.

-----------------------------------------------------------
For anyone interested, I took photographs of the analog
signals and the resulting ADC-DAC signals at different
sample rates.  The file is about 1.5 MB - so a fast email
connection is recommended.  But I'm getting off track...
-----------------------------------------------------------

I'll spare you my equipment list, suffice it to say that
everything is either professional studio gear or audiophile
monitoring components with very wide bandwidth and excellent
linearity.

So what did I hear?  (Again, you may hear something different
than I did - I only include the results of my test for those
who may be curious)

The test:  I had my wife help me with an A-B test.  First, she
switched between waveforms, telling me which I was listening to.
Then she switched randomly (blind to me) between waveforms and
I had to identify them.  I could not see the function generator,
and did not know my success/failure rate until the end of the
test.  I was blind tested with the three waveforms at 1, 2, 4,
8, 10, 12, and 16 kHz.

Just for the record - the limits of my hearing: I can hear an
18 kHz sine and triangle wave, but not the 18 kHz square wave.
I can clearly no longer hear 20 kHz (maybe if the room was
REALLY REALLY quiet...), at least with my electronics.

I was able to identify each waveform with 100% consistency up
to 8 kHz.  Each waveform had an unmistakable timbre.  I could
continue to identify waveforms accurately up to 12 kHz, but I
keyed onto different audible cues - not necessarily timbre.

At 10 kHz, the sine and triangle were not clearly differentiated
by their sound, but rather by their intensity or volume level.
The triangle being slightly louder than the sine.  The square
still had a distinct timbre to it that was different from the
sine and triangle.  The results were still the same at 12 kHz.
My batting average remained 100%.

At 16 kHz, I could no longer differentiate the sine and triangle,
but I could still consistently pick out the square wave by its
intensity - but not by its timbre.

To make things more interesting, we repeated the entire listening
exercise by monitoring the ADC-DAC output sampled at 88.2 kHz, and
the results were the same as the analog listening test!  There
may have been subtle differences in loudness and timbre between
analog and digital, but that wasn't the goal of this test.  I just
wanted to see if I could hear a difference between waveforms with
all else constant.

However, at 44.1 kHz I had much more difficulty differentiating the
square wave at 10 kHz - but I still consistently could.  I could not
tell the 10 kHz sine and triangle apart.  At 16 kHz I could no
longer differentiate the square wave from the triangle or the sine,
whereas I could at 88.2 kHz sampling.

The most interesting conclusion for me from this experiment is that
the waveform clearly impacts the EQ.  Different waveforms of the
same frequency and amplitude are audibly louder or have more energy,
even if the harmonics are ultrasonic and I can no longer differentiate
on the basis of timbre.  If a non-sinusoidal wave or impulse has
its waveform significantly altered in the ADC-DAC chain, then the
perceived loudness of those frequencies will change as well, and
hence the EQ!

And at least with my ears, so long as a square wave is "square" at
12 kHz via an ADC-DAC chain, I shouldn't be able to readily tell the
difference between high quality analog and high quality digital.  How
"square" does square have to be at 12 kHz?  That's another test for
another day.  But I am absolutely positive that I can hear a clear
and consistent difference between 16/44.1 and 24/88.2.

I have a bunch of analog drum tests with which I would also
like to repeat a similar test.  But again, that's for another day.

Whether you share your results with the ARSCList or not is not
important.  It's more important that you do a first-hand test and
convince yourself of the differences you can and cannot hear.

Have fun testing away!

Eric



-----Original Message-----
From: Association for Recorded Sound Discussion List
[mailto:ARSCLIST@xxxxxxx]On Behalf Of Dave Bradley
Sent: Monday, January 17, 2005 6:36 PM
To: ARSCLIST@xxxxxxxxxxxx
Subject: Re: [ARSCLIST] Sampling Theory (was Fred Layn's post on the
Studer list re: Quantegy)


>The history of recorded sound follows a path that never included the
>quality of sound as a prime criteria.  Think me off the deep end, then I
>strongly suggest you study the history a bit closer.
>
>What I believe you lack is a reference point.  Get out & hear a double
>reed instrument, a high hat, a steel drum, a double base, a timpani, a
>National steel string guitar or childrens' voices or G-d forbid, the
>Grambling Marching Band.

I'm quite familiar with all of those instruments in an acoustic setting,
having played trumpet in an orchestra (nothing too sophisticated as I don't
play well).  I've heard many an analog master tape as well, having worked
in the industry for more than 20 years. I've transferred my share of 30 ips
and 15 ips master tapes, as well as my share of consumer format reels at
slower speeds.  I have more than enough reference point. I've also got the
ears to tell me that analog isn't giving you more than digital, it's giving
you less. The problem is that most people are accustomed to hearing less
and find more to be a sound they aren't familiar with or comfortable
with.  Trust me, I have a reference point.

And I'm certainly not forcing anything on anyone or telling anyone that it
has to be their "standard".  However, I also won't sit by quietly while
people talk about how horrid digital is and listen to things about digital
that I know to be inaccurate or at the very least based on listening to an
inferior digital setup.  There are people who would spend thousands on a
turntable, carefully selecting which arm, cartridge, stylus, even proper
interconnects just to get the "great sound that analog is capable of" who
think that a $50 CD player that doesn't sound as nice is proof that digital
stinks.  I've heard high end analog. I love the sound it gives.  I've heard
high end digital. I love the sound it gives.  To hear someone say that
digital is incapable of what analog does it annoying because it's not true.
I take you back to the start of this whole thread when I responded to
someone who thought that because digital is based on individual samples,
that the output upon playback would be stepped on a scope, not a smooth
sine wave like it was trying to reproduce. That's total BS and nothing
could be further from the truth. Just because someone doesn't like digital,
and just because someone falsely believes that digital is incapable of
reproducing what analog can reproduce does not mean that garbage like the
stepped sine wave junk should be spread around as the gospel truth when it
couldn't be any further from the truth.  To that end, I stopped sitting
silently and took the time to speak up.  I no more desire to force a format
on anyone than I desire to destroy all the tape archives from the past 50
years. However, I'm also not going to sit by in a forum that is supposed to
be made up of professionals and hear someone say that the analog output of
a DAC would show a stepped wave form instead of a sinewave simply because
they've seem a digital editing setup show the individual steps based on the
sample values. That's not what comes out of the DAC, that's what it looks
like while in digital format, not after being converted back to analog.  To
say otherwise is to lie or grossly misunderstand the technology.

So tell me where I've forced anyone to accept digital as a standard, which
you seem to be claiming I've done with your statement that, "its absurd if
not simply arrogant to force such technology as a STANDARD, which it AIN'T!"


-----------------
Diamond Productions
Specializing in analog tape & film preservation / restoration in the
digital domain.
Dave Bradley   President