I'm about to compare interconnect cables blind. I'm trying to
establish for myself whether cable quality is an important factor, and
whether interconnects of similar construction can vary in subjective
sound.

I'd like to use two cables that are electrically similar. Right now I
own two cables that might be fit for comparison: a 2M Radio Shack
"gold" special, and a Transparent 2M Link 300. I thought I would
measure the capacitance of each cable using this rather cheap meter
that I own. Turns out the Radio Shack is 400 pF, and the Transparent
audio is 280 pF. Do these numbers sound reasonable? I'm not sure if
my meter is working or if I'm doing this correctly.

And my EE knowledge is rusty. I want to figure out how this
capacitance might affect the transfer function. This interconnect
runs between a CAL Sigma II DAC and an Audio Research SP-6. What is
the impedance of a capacitor again? Something like 1 / ( 2 * pi *
frequency * capacitance) ?

Since I suspect that cable quality is real and can be heard blind if
the test is carried out properly (my theory is that quick switching is
an improper way to seek out small changes in sound), I'm interested in
results that would challenge the scientific field of psychoacoustics.
How small a difference in transfer function is considered inaudible?

Regards,
Mike

A difference of less than 0.3dB may be considered "pretty inaudible",
especially at the top end of the audio range. The capacitance values you
quote are a tad on the high side, unless the interconnects are very long.
Nevertheless, they're unlikely to amount to a noticeable HF reduction.

I'm not saying you won't hear a difference, but if you do, the capacitance
and its attendant HF reduction per se isn't going to be the cause.

"Michael Mossey" wrote in message
...
I'm about to compare interconnect cables blind. I'm trying to
establish for myself whether cable quality is an important factor, and
whether interconnects of similar construction can vary in subjective
sound.

I'd like to use two cables that are electrically similar. Right now I
own two cables that might be fit for comparison: a 2M Radio Shack
"gold" special, and a Transparent 2M Link 300. I thought I would
measure the capacitance of each cable using this rather cheap meter
that I own. Turns out the Radio Shack is 400 pF, and the Transparent
audio is 280 pF. Do these numbers sound reasonable? I'm not sure if
my meter is working or if I'm doing this correctly.

And my EE knowledge is rusty. I want to figure out how this
capacitance might affect the transfer function. This interconnect
runs between a CAL Sigma II DAC and an Audio Research SP-6. What is
the impedance of a capacitor again? Something like 1 / ( 2 * pi *
frequency * capacitance) ?

Since I suspect that cable quality is real and can be heard blind if
the test is carried out properly (my theory is that quick switching is
an improper way to seek out small changes in sound), I'm interested in
results that would challenge the scientific field of psychoacoustics.
How small a difference in transfer function is considered inaudible?

Regards,
Mike

On 28 Nov 2003 15:28:51 GMT, (Michael Mossey)
wrote:

I'm about to compare interconnect cables blind. I'm trying to
establish for myself whether cable quality is an important factor, and
whether interconnects of similar construction can vary in subjective
sound.

I'd like to use two cables that are electrically similar. Right now I
own two cables that might be fit for comparison: a 2M Radio Shack
"gold" special, and a Transparent 2M Link 300. I thought I would
measure the capacitance of each cable using this rather cheap meter
that I own. Turns out the Radio Shack is 400 pF, and the Transparent
audio is 280 pF. Do these numbers sound reasonable? I'm not sure if
my meter is working or if I'm doing this correctly.

These are reasonable figures for 2 metre cables.

And my EE knowledge is rusty. I want to figure out how this
capacitance might affect the transfer function. This interconnect
runs between a CAL Sigma II DAC and an Audio Research SP-6. What is
the impedance of a capacitor again? Something like 1 / ( 2 * pi *
frequency * capacitance) ?

The important thing is the output resistance of the DAC, which forms a
low-pass filter in conjunction with the cable capacitance. If it's
less than 300 ohms, there should be no audible treble droop with
either cable.

Since I suspect that cable quality is real and can be heard blind if
the test is carried out properly (my theory is that quick switching is
an improper way to seek out small changes in sound), I'm interested in
results that would challenge the scientific field of psychoacoustics.

If that's your theory, then you are indeed challenging a century or
more of psychoacoustics! :-)

How small a difference in transfer function is considered inaudible?

Typically, 0.5 to 1dB level difference is audible. Freequency droops
are less obvious, you might need as much as 2-3dB at 20kHz to be
audible.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

(Stewart Pinkerton) wrote:

On 28 Nov 2003 15:28:51 GMT, (Michael Mossey)
wrote:

I'm about to compare interconnect cables blind. I'm trying to
establish for myself whether cable quality is an important factor, and
whether interconnects of similar construction can vary in subjective
sound.

I'd like to use two cables that are electrically similar. Right now I
own two cables that might be fit for comparison: a 2M Radio Shack
"gold" special, and a Transparent 2M Link 300. I thought I would
measure the capacitance of each cable using this rather cheap meter
that I own. Turns out the Radio Shack is 400 pF, and the Transparent
audio is 280 pF. Do these numbers sound reasonable? I'm not sure if
my meter is working or if I'm doing this correctly.

These are reasonable figures for 2 metre cables.

And my EE knowledge is rusty. I want to figure out how this
capacitance might affect the transfer function. This interconnect
runs between a CAL Sigma II DAC and an Audio Research SP-6. What is
the impedance of a capacitor again? Something like 1 / ( 2 * pi *
frequency * capacitance) ?

The important thing is the output resistance of the DAC, which forms a
low-pass filter in conjunction with the cable capacitance. If it's
less than 300 ohms, there should be no audible treble droop with
either cable.

Since I suspect that cable quality is real and can be heard blind if
the test is carried out properly (my theory is that quick switching is
an improper way to seek out small changes in sound), I'm interested in
results that would challenge the scientific field of psychoacoustics.

If that's your theory, then you are indeed challenging a century or
more of psychoacoustics! :-)

How small a difference in transfer function is considered inaudible?

Typically, 0.5 to 1dB level difference is audible. Freequency droops
are less obvious, you might need as much as 2-3dB at 20kHz to be
audible.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

Interestingly my experience tells me that 2-dB is right on the 'threshold' of
audibility with most music programs. With super-sensitive sources such as full
band pink noise 0.2 dB (+/- 0.1 dB 100-10,000 Hz) seems to at be the edge of
audbility.

With 50-year old males -3 db at 20 kHz is still a wildly optimistic guess.

On Tue, 02 Dec 2003 06:37:51 GMT, (Nousaine) wrote:

(Stewart Pinkerton) wrote:

On 28 Nov 2003 15:28:51 GMT, (Michael Mossey)
wrote:

Since I suspect that cable quality is real and can be heard blind if
the test is carried out properly (my theory is that quick switching is
an improper way to seek out small changes in sound), I'm interested in
results that would challenge the scientific field of psychoacoustics.

If that's your theory, then you are indeed challenging a century or
more of psychoacoustics! :-)

How small a difference in transfer function is considered inaudible?

Typically, 0.5 to 1dB level difference is audible. Freequency droops
are less obvious, you might need as much as 2-3dB at 20kHz to be
audible.

Interestingly my experience tells me that 2-dB is right on the 'threshold' of
audibility with most music programs. With super-sensitive sources such as full
band pink noise 0.2 dB (+/- 0.1 dB 100-10,000 Hz) seems to at be the edge of
audbility.

With 50-year old males -3 db at 20 kHz is still a wildly optimistic guess.

Sadly, I suspect that this is true! :-(

Of course, we must assume young and undamaged ears if we are not to be
accused of excessive 'lowest common denominator' thinking.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

(Stewart Pinkerton) wrote in message ...
On 28 Nov 2003 15:28:51 GMT, (Michael Mossey)
wrote:

I'm about to compare interconnect cables blind. I'm trying to
establish for myself whether cable quality is an important factor, and
whether interconnects of similar construction can vary in subjective
sound.

I'd like to use two cables that are electrically similar. Right now I
own two cables that might be fit for comparison: a 2M Radio Shack
"gold" special, and a Transparent 2M Link 300. I thought I would
measure the capacitance of each cable using this rather cheap meter
that I own. Turns out the Radio Shack is 400 pF, and the Transparent
audio is 280 pF. Do these numbers sound reasonable? I'm not sure if
my meter is working or if I'm doing this correctly.

These are reasonable figures for 2 metre cables.

And my EE knowledge is rusty. I want to figure out how this
capacitance might affect the transfer function. This interconnect
runs between a CAL Sigma II DAC and an Audio Research SP-6. What is
the impedance of a capacitor again? Something like 1 / ( 2 * pi *
frequency * capacitance) ?

The important thing is the output resistance of the DAC, which forms a
low-pass filter in conjunction with the cable capacitance. If it's
less than 300 ohms, there should be no audible treble droop with
either cable.

Since I suspect that cable quality is real and can be heard blind if
the test is carried out properly (my theory is that quick switching is
an improper way to seek out small changes in sound), I'm interested in
results that would challenge the scientific field of psychoacoustics.

If that's your theory, then you are indeed challenging a century or
more of psychoacoustics! :-)

...and he would be the perfect person to conduct this test! If he
really beleives in his theory, he will turn every stone to find the
difference, let us just hope that he does not forget what "blind"
means. :-)


How small a difference in transfer function is considered inaudible?

Typically, 0.5 to 1dB level difference is audible. Freequency droops
are less obvious, you might need as much as 2-3dB at 20kHz to be
audible.

I have performed blind tests with music, and I could hear a first
order 47 kHz lowpass filter. Yes, I had a random generator connecting
the filter for me, and seven correct responses out of seven tries,
giving me 99% confidence. Need I say that the difference was not
large? Note that this does not not mean that i hear frequencies at 47
kHz, sadly the limit is somewhere around 15kHz nowadays, but a 47 kHz
first order lowpass filter has a drop of about 0.42 dB at 15 kHz and I
am pretty convinced that this is what I heard.

So, if we play with the thought that the output resistance and cable
capacitance form a first order lowpass filter, 400 pF and 47 kHz means
that the output resistance of the driving amplifier should be 8.5
kohms. Probably the output resistance is 100 ohms or less, so the
cutoff frequency should be higher.

Svante wrote:

How small a difference in transfer function is considered inaudible?

Typically, 0.5 to 1dB level difference is audible. Freequency droops
are less obvious, you might need as much as 2-3dB at 20kHz to be
audible.

I have performed blind tests with music, and I could hear a first
order 47 kHz lowpass filter. Yes, I had a random generator connecting
the filter for me, and seven correct responses out of seven tries,
giving me 99% confidence. Need I say that the difference was not
large? Note that this does not not mean that i hear frequencies at 47
kHz, sadly the limit is somewhere around 15kHz nowadays, but a 47 kHz
first order lowpass filter has a drop of about 0.42 dB at 15 kHz and I
am pretty convinced that this is what I heard.


What test material were you using? Music or noise?



--

-S.

"They've got God on their side. All we've got is science and reason."
-- Dawn Hulsey, Talent Director

(Svante) wrote:

I have performed blind tests with music, and I could hear a first
order 47 kHz lowpass filter. Yes, I had a random generator connecting
the filter for me, and seven correct responses out of seven tries,
giving me 99% confidence. Need I say that the difference was not
large?

Most likely you heard changes due to phase shift which start to get
noticable at 1/10 of the frequency of your filter, in this case at
about 5 kHz. You may repeat the test but use pink noise rather than
music and you should no longer be able to detect the presence of the
filter.

Note that this does not not mean that i hear frequencies at 47
kHz, sadly the limit is somewhere around 15kHz nowadays, but a 47 kHz
first order lowpass filter has a drop of about 0.42 dB at 15 kHz and I
am pretty convinced that this is what I heard.

So I think that you didn't hear the 0.42 dB drop at 15 kHz but the
phase shift. OTOH, at 7.5 kHz the drop ist 0.2 dB which may not be
audible as loss of trebble but as a slight veil on the music.

Norbert