"Menno" wrote in message
...
Eiron wrote:
Shock result - interconnects sound the same!

As an interconnect test I soldered up a short adapter (4cm) with a
male phono, two female phonos and a jack plug.
This allowed me to take a mono source and feed it directly to the
left channel of a sound card, while also sending it through the
interconnect under test to the right channel of the sound card.

The source was Donna Summer's 'Bad Reputation', which has a lot of
treble.
I would have used Abba's 'The Day Before You Came' but I get a bit
emotional listening to that which might spoil the results.

Using Goldwave and comparing the source directly with the source
passed through the oldest, cheapest interconnect in my junkbox
revealed that the difference peaked at -57.9dB. After taking the
difference and maximising it, it sounded like the original, with more
noise and a grainy effect to the treble.

If anyone can hear anything at 58dB below the music level he must have
better ears than me.

I haven't yet calibrated the sound card, i.e. balanced the two
channels
but these preliminary results suggest that the squirrel is nuts.



What differences did you see? Did you also look at phase plots or just
amplitude? My experience is that the audible difference in interlinks is
more in the spatial information than the attenuation of high frequenies.
So
with better interlinks you get more of the stereo image embedded in the
recording. More depth and better precision in where sounds come from.
Obviously you would need a recording that has that in it in the first
place.
Pop/rock music with a lot of electronic instruments is usually not the
best
for this.

How would you measure depth and precision in a way that is repeatable and
verifiable.

On the theory side, many people seem to overlook the fact that the energy
of
an electrical signal is stored in the electromagnetic field generated by
the
signal (Maxwell's law from the 19th century if I remember well). Most of
the
EM field is located OUTSIDE the conductor. For example in an interlink,
most
of the EM field is between the inner conductor and the shield, so in the
insulation material between them. In my first year in university
(electrical
engineering), I learned that various materials have different effects on
the
EM field. Polair molecules respond different to changing EM fields than
a-polair molecules. So a coax with a teflon/PTFE (a-polair) dielectircum
has
different characteristics than one with a PVC (polair) dielectricum. Air
as
dielectricum is quite nice. Think of helix or foams cables. Capacitors
come
with various dielectricums for various applications. So why not cables.

However, the Electromagnetic field in a cable carrying normal signal
currents is so small, that variations in the dielectric constant and
dielectric losses are vanishingly small. No-one suggests that cables don't
measure different, but that these differences cause such smal variation in
level, phase shift and frequency response as to be very well below any
threshold of audibility.

Then there's the magnetic influence of various metals. Iron
(ferromagnetic)
wire sounds different that copper (non ferromagnetic) wire. Silver sound
different than copper. Many years ago I did a listening test of interlinks
for a Dutch audio magazine together with a student. We did used nice
stereo equipment for this but nothing over the top. Both of us were
shocked
by the difference between
the silver (Siltech) cable and the copper cables. The difference between
the
silver and copper cables was much larger than that between the various
copper
cables. A guy in Japan, at the time working for Mitsubishu (I think),
experimented with various metals like tin, (OFC) copper, iron, aluminium,
silver,
nickel. He published the results in the french magazine l"Audiophile in
the
80's. I read a dutch translation and the outcome was that basically every
metal had its own sound. I recall that the autor thought that silver
sounded
overly brilliant.

Again this is nonsense when the differences are expressed in units of
measurement. . Remember also that l'audiophile is a a subjective-driven
magazine, and I would not expect them to know the science behind what they
are publishing.

When dealing with larger currents (speaker cables), the mechanical
characteristics start to play a role. Parallel conducters push/pull on
each
other due to magnetic (Lorentz) forces thereby creating a signal dependant
inductance and capacitance which modulates the signal going through the
cable.

The mechanical movement will be infinitesimal, when considering the current
flowing and the magnitude of the earth's magnetic field, consequently the
change in capacitance and inductance will be similarly tiny. Even if it
wasn't, the change in voltage across the 'speaker considering the source
impedance from the amplifier is very low would be again vanishingly small.

Looking into studies about hearing will probably reveal that the human
brain
can detect much more in sounds than the avarage person thinks. The ears
are
just the sensor (say the A/D converter in your soundcard) while our brain
does the real thing (like the processor and software that you used in your
experiment). The ultimate goal of good sounding cables is to please our
listening experience, not to produce FFT plots. So use your ears initially
and then try to find out what you hear with measurements.

Thresholds of hearing are very well documented and understood. Although the
effects you describe above are real effects, what you seem to ignore is that
the magnitude of these effects is orders of magnitude below what is
detectable.

I agree that there's a lot of BS around cables. But to simply dismiss the
whole matter seems a bit short sighted. Specially since there's a lot of
scientific knowledge to back up certain theories. You just need think and
search outside the audio box. And always remember that the whole chain is
as
strong as the weakest link. The weakest link may not be the interlink or
the
speaker cable.

Menno

What I would like to see is more research on thresholds of audibility. If
there are genuinely people out there who's hearing is so sensitive compared
to the rest of us, we should know about it and use this knowledge to learn
more about how hearing works. Sadly, those who claim to hear these effects
seem unwilling to be tested.

S.

On Tue, 7 Mar 2006 22:35:27 -0000, "Serge Auckland"
wrote:

"Menno" wrote in message
. ..

snipping to the chase

Thresholds of hearing are very well documented and understood. Although the
effects you describe above are real effects, what you seem to ignore is that
the magnitude of these effects is orders of magnitude below what is
detectable.

I agree that there's a lot of BS around cables. But to simply dismiss the
whole matter seems a bit short sighted. Specially since there's a lot of
scientific knowledge to back up certain theories. You just need think and
search outside the audio box. And always remember that the whole chain is as
strong as the weakest link. The weakest link may not be the interlink or the
speaker cable.

Menno

What I would like to see is more research on thresholds of audibility. If
there are genuinely people out there who's hearing is so sensitive compared
to the rest of us, we should know about it and use this knowledge to learn
more about how hearing works. Sadly, those who claim to hear these effects
seem unwilling to be tested.

More to the point, if you make a few calculations on *quantity*, all
of the perfectly genuine physical effects mentioned by Menno give rise
to disturbances in the soundfield at the subatomic level, IOW they are
hundreds or thousands of times below any possibility of being detected
by human ears.

I've worked on submarine systems, which are capable of recognising the
sound signature of vessels a *thousand miles* away, and there was
never any question of using special cables. I've also looked at
nonlinearity of cables in the design of military test equipment, and
there was nothing to be found down to less than -150dBm.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

In article , Stewart
Pinkerton
wrote:


What I would like to see is more research on thresholds of audibility.
If there are genuinely people out there who's hearing is so sensitive
compared to the rest of us, we should know about it and use this
knowledge to learn more about how hearing works. Sadly, those who
claim to hear these effects seem unwilling to be tested.

More to the point, if you make a few calculations on *quantity*, all of
the perfectly genuine physical effects mentioned by Menno give rise to
disturbances in the soundfield at the subatomic level, IOW they are
hundreds or thousands of times below any possibility of being detected
by human ears.

Similarly, when measurements are done on cables, the effects on the signals
of the kinds of physical effects described then to be in the region between
'very very tiny' down to 'undetectable' or 'nil'.

I've worked on submarine systems, which are capable of recognising the
sound signature of vessels a *thousand miles* away, and there was never
any question of using special cables. I've also looked at nonlinearity
of cables in the design of military test equipment, and there was
nothing to be found down to less than -150dBm.

FWIW I was for many years involved in designing ultra-high precision test
equipment for the NPL. The kinds of effects Menno described can generally
be ignored in such contexts... :-)

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

Menno wrote:
Eiron wrote:

Shock result - interconnects sound the same!

As an interconnect test I soldered up a short adapter (4cm) with a
male phono, two female phonos and a jack plug.
This allowed me to take a mono source and feed it directly to the
left channel of a sound card, while also sending it through the
interconnect under test to the right channel of the sound card.

The source was Donna Summer's 'Bad Reputation', which has a lot of
treble.
I would have used Abba's 'The Day Before You Came' but I get a bit
emotional listening to that which might spoil the results.

Using Goldwave and comparing the source directly with the source
passed through the oldest, cheapest interconnect in my junkbox
revealed that the difference peaked at -57.9dB. After taking the
difference and maximising it, it sounded like the original, with more
noise and a grainy effect to the treble.

If anyone can hear anything at 58dB below the music level he must have
better ears than me.

I haven't yet calibrated the sound card, i.e. balanced the two
channels
but these preliminary results suggest that the squirrel is nuts.




What differences did you see? Did you also look at phase plots or just
amplitude? My experience is that the audible difference in interlinks is
more in the spatial information than the attenuation of high frequenies. So
with better interlinks you get more of the stereo image embedded in the
recording. More depth and better precision in where sounds come from.

Why don't you try it yourself and see?
On a second look at the results, the signal through the crappy cable was
down by 0.1% or 0.01dB. So by using Goldwave and boosting the level a bit,
the difference was down to -65dB, with noise at about -70dB, and the voice
barely distinguishable.
Which still shows that a crappy 1 metre interconnect makes no audible difference.

--
Eiron

There's something scary about stupidity made coherent - Tom Stoppard.

Menno wrote:

And always remember that the whole chain is as strong as the weakest
link. The weakest link may not be the interlink or the speaker cable.

And it would appear that the original poster used a PC sound card as his
test bed platform... so he's probably right, the analogue stage on most
PC sound cards is pretty appalling.

So it's no wonder the OP didn't hear a difference, as the weakest link
in the chain was the sound card.

Now, repeat the experiment using a sound card with SPDIF out, one that
*doesn't* resample everything to 48kHz (cos that makes the sound fuzzy
the way most sound cards do it), and hook it up to a high quality
offboard DAC. Such as an Arcam Black Box 50.

Then re-post with your results.

--
Glenn Richards Tel: (01453) 845735
Squirrel Solutions http://www.squirrelsolutions.co.uk/

IT consultancy, hardware and software support, broadband installation

Eiron wrote:

On a second look at the results, the signal through the crappy cable was
down by 0.1% or 0.01dB.

So there is a difference then.

and the voice barely distinguishable.
^^^^^^^^^^^^^^^^^^^^^^
"barely", meaning "there was a difference, but only a very tiny one".

Which still shows that a crappy 1 metre interconnect makes no audible
difference.

You just contradicted yourself there. Either there was a difference or
there wasn't. You've said there was. There was a 0.01dB loss in the
crappy cable - that's still a loss.

The human ear is far more sensitive than most test equipment. If you can
measure it, you can certainly hear it.

--
Glenn Richards Tel: (01453) 845735
Squirrel Solutions http://www.squirrelsolutions.co.uk/

IT consultancy, hardware and software support, broadband installation

On Wed, 08 Mar 2006 11:48:23 +0000, Glenn Richards
wrote:

Eiron wrote:

On a second look at the results, the signal through the crappy cable was
down by 0.1% or 0.01dB.

So there is a difference then.

and the voice barely distinguishable.
^^^^^^^^^^^^^^^^^^^^^^
"barely", meaning "there was a difference, but only a very tiny one".


The very method has small uncertainties in it that would result in
small differences, even with the same cable.

Which still shows that a crappy 1 metre interconnect makes no audible
difference.

You just contradicted yourself there. Either there was a difference or
there wasn't. You've said there was. There was a 0.01dB loss in the
crappy cable - that's still a loss.


Try and read what people say, not what you would like them to say. He
didn't contradict himself. So look again, and see if you can find the
important adjective you ignored the first time.

The human ear is far more sensitive than most test equipment. If you can
measure it, you can certainly hear it.

Quite the reverse is the case. Anything that makes an audible
difference can be measured with the greatest ease, because it will be
gross. The opposite is most definitely not the case.

d

Pearce Consulting
http://www.pearce.uk.com

Glenn Richards wrote:

Eiron wrote:

On a second look at the results, the signal through the crappy cable was
down by 0.1% or 0.01dB.


So there is a difference then.

and the voice barely distinguishable.

^^^^^^^^^^^^^^^^^^^^^^
"barely", meaning "there was a difference, but only a very tiny one".

Which still shows that a crappy 1 metre interconnect makes no audible
difference.


You just contradicted yourself there. Either there was a difference or
there wasn't. You've said there was. There was a 0.01dB loss in the
crappy cable - that's still a loss.

The human ear is far more sensitive than most test equipment. If you can
measure it, you can certainly hear it.

Read what I wrote, not what you would like to hear.

Now, it is not true that "If you can measure it, you can certainly hear it"
but the opposite is the case: If there is a difference to be heard,
you can certainly measure it.

If you can't accept that, there is no hope for you.
If you can, then we can devise some test that will measure those differences.

--
Eiron

There's something scary about stupidity made coherent - Tom Stoppard.

On 2006-03-08, Glenn Richards wrote:
Eiron wrote:

On a second look at the results, the signal through the crappy cable was
down by 0.1% or 0.01dB.

So there is a difference then.

and the voice barely distinguishable.
^^^^^^^^^^^^^^^^^^^^^^
"barely", meaning "there was a difference, but only a very tiny one".

There are always differences that you can measure between cables.
Their RLC parameters vary with physical construction and the materials
used so no two are exactly the same. The usual interaction between cable,
source and load causes differences in frequency response (amplitude
and phase). The issue is whether a difference is audible.

...
The human ear is far more sensitive than most test equipment. If you can
measure it, you can certainly hear it.

The professionally published literature on audible differences seems to
disagree with this conclusion. (Well, the literature that I am aware
of - and I do admit I am aware of only a few results so if someone can
point to relevant papers please do so.)

IIRC, something like a 0.3 dB wide band audio level change is just about
audible as far as I recall from a paper I saw in JAES. Other results I
have seen show that for level differences over narrow frequency bands
the audibility threshold is greater.

It is trivially possible to measure far smaller level differences
than this.

--
John Phillips

On 2006-03-07, Serge Auckland wrote:

"Menno" wrote in message
...
When dealing with larger currents (speaker cables), the mechanical
characteristics start to play a role. Parallel conducters push/pull on
each
other due to magnetic (Lorentz) forces thereby creating a signal dependant
inductance and capacitance which modulates the signal going through the
cable.

The mechanical movement will be infinitesimal, when considering the current
flowing and the magnitude of the earth's magnetic field, consequently the
change in capacitance and inductance will be similarly tiny. Even if it
wasn't, the change in voltage across the 'speaker considering the source
impedance from the amplifier is very low would be again vanishingly small.

I was considering this. The standard formula for the attractive force
(F) between two parallel conductors of length l and separation d is:

F = k * I1 * I2 * l/d

The constant k is defined to be 2*10^-7 N/Amp^2 and I1 = -I2 for a
speaker cable (i.e. the same current returns in the opposite direction).

So, for a 1A current, conductors 2mm apart repel each other with a force
of 10^-4 Newtons per metre or about 0.01 grams force per metre.

I haven't looked up the elastic properties of dielectrics yet to determine
the mechanical movement (any references for PVC or similar?). However 10
milligrams per metre of length isn't a big force.

Even then the intermodulation produced by the varying capacitance will
be rather small as the amplitude and phase response of a loudspeaker
cable is primarily determined by the inductance not the capacitance.

--
John Phillips