Stewart Pinkerton wrote:

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

No, it's true.

There are cases when the notion that non-blind tests are "inherently
unreliable" just doesn't wash. I'm sure that, if "Callas" was fed a spoonful
of ****e, and a spoonful of sugar, s/he/it would be able to establish which
is the ****e and which s the sugar, and could do so regardless of whether or
not they were blindfolded. Indeed, even if one were to conduct a hundred
such tests 'eyes-on' and a hundred blind, I bet the result would be
consistent and reliable.


Not that I'm saying the 203 isn't a good DAC, you
understand! :-)

I know you're not saying that. Even if you were, I'd disagree. :-)


--
Wally
www.art-gallery.myby.co.uk
On webcam: Black Cat In Coal Cellar

wrote:
Stewart Pinkerton wrote:

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

No, it's true.

There are cases when the notion that non-blind tests are "inherently
unreliable" just doesn't wash. I'm sure that, if "Callas" was fed a spoonful
of ****e, and a spoonful of sugar, s/he/it would be able to establish which
is the ****e and which s the sugar, and could do so regardless of whether or
not they were blindfolded. Indeed, even if one were to conduct a hundred
such tests 'eyes-on' and a hundred blind, I bet the result would be
consistent and reliable.

Your example and reasoning are poor.

To begin with, with audio equipment, the listener has paid for and so
wishes to perceive improvements in the sound to which he is listening.

There is no such parallel in your example; you are in no way considering
the desire on the part of the subject to perceive that his exciting new
piece of equipment makes a difference.

Next, the example given describes very strongly different flavours. You
are correct to state that I would certainly perceive the differences.
The sensory input from the two stimuli are *so* different there is no
mistaking them.

The problem with audio comparisons is far subtler. The differences
sensory input can be small - and indeed, non-existant. As differences
become smaller, it takes a increasingly skilled listener to perceive
them reliably. Lesser listeners will fail outright in the first place,
or find that their subjective desire to perceive improve generates a
fake difference which is in fact greater than the real difference.

I have to say also that given your emotive response in this matter, I
suspect you're being dogmatic; this is a subject about which you have
Opinions. Have you tried blind testing yourself?

--
Callas

wrote:
Stewart Pinkerton wrote:

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

No, it's true.

There are cases when the notion that non-blind tests are "inherently
unreliable" just doesn't wash. I'm sure that, if "Callas" was fed a spoonful
of ****e, and a spoonful of sugar, s/he/it would be able to establish which
is the ****e and which s the sugar, and could do so regardless of whether or
not they were blindfolded. Indeed, even if one were to conduct a hundred
such tests 'eyes-on' and a hundred blind, I bet the result would be
consistent and reliable.

Your example and reasoning are poor.

To begin with, with audio equipment, the listener has paid for and so
wishes to perceive improvements in the sound to which he is listening.

There is no such parallel in your example; you are in no way considering
the desire on the part of the subject to perceive that his exciting new
piece of equipment makes a difference.

Next, the example given describes very strongly different flavours. You
are correct to state that I would certainly perceive the differences.
The sensory input from the two stimuli are *so* different there is no
mistaking them.

The problem with audio comparisons is far subtler. The differences
sensory input can be small - and indeed, non-existant. As differences
become smaller, it takes a increasingly skilled listener to perceive
them reliably. Lesser listeners will fail outright in the first place,
or find that their subjective desire to perceive improve generates a
fake difference which is in fact greater than the real difference.

I have to say also that given your emotive response in this matter, I
suspect you're being dogmatic; this is a subject about which you have
Opinions. Have you tried blind testing yourself?

--
Callas

"Wally" wrote in message

Stewart Pinkerton wrote:

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

No, it's true.

There are cases when the notion that non-blind tests are "inherently
unreliable" just doesn't wash. I'm sure that, if "Callas" was fed a
spoonful of ****e, and a spoonful of sugar, s/he/it would be able to
establish which is the ****e and which s the sugar, and could do so
regardless of whether or not they were blindfolded. Indeed, even if
one were to conduct a hundred such tests 'eyes-on' and a hundred
blind, I bet the result would be consistent and reliable.

That's not the same as comparing reasonably good DACs.

Do try to make your comments relevant to the topic described in the title
line, eh?

"Wally" wrote in message

Stewart Pinkerton wrote:

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

No, it's true.

There are cases when the notion that non-blind tests are "inherently
unreliable" just doesn't wash. I'm sure that, if "Callas" was fed a
spoonful of ****e, and a spoonful of sugar, s/he/it would be able to
establish which is the ****e and which s the sugar, and could do so
regardless of whether or not they were blindfolded. Indeed, even if
one were to conduct a hundred such tests 'eyes-on' and a hundred
blind, I bet the result would be consistent and reliable.

That's not the same as comparing reasonably good DACs.

Do try to make your comments relevant to the topic described in the title
line, eh?

Stewart Pinkerton wrote:

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

No, it's true.

There are cases when the notion that non-blind tests are "inherently
unreliable" just doesn't wash. I'm sure that, if "Callas" was fed a spoonful
of ****e, and a spoonful of sugar, s/he/it would be able to establish which
is the ****e and which s the sugar, and could do so regardless of whether or
not they were blindfolded. Indeed, even if one were to conduct a hundred
such tests 'eyes-on' and a hundred blind, I bet the result would be
consistent and reliable.


Not that I'm saying the 203 isn't a good DAC, you
understand! :-)

I know you're not saying that. Even if you were, I'd disagree. :-)


--
Wally
www.art-gallery.myby.co.uk
On webcam: Black Cat In Coal Cellar

On Sun, 4 Jan 2004 14:56:56 -0000, "Wally"
wrote:

Callas wrote:
wrote:
Following the recent thread about adding a DAC to my Schneider DVD
player, I went ahead and purchased a used Meridian 203 when it was
offered by a subscriber here. It arrived this morning and it's been
playing all day...

Sorry to be negative, but I think only blind tests are valid.

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

No, it's true. Not that I'm saying the 203 isn't a good DAC, you
understand! :-)
--

Stewart Pinkerton | Music is Art - Audio is Engineering

On Sun, 4 Jan 2004 14:56:56 -0000
"Wally" wrote:

With something as subjective as the perception of sound, non-blind
tests are inherently unreliable.

Bull****.

I suspect what you meant to say was that a non-blind test isnt necessarily invalid in terms of testing gross changes ;-)

For tiny changes on the threshold of perception, you must use blind testing.

In your case it was 'crap DVD player' vs 'decent DAC' which isnt exactly a subtle difference ;-)

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

Ian Molton wrote:

Bull****.

I suspect what you meant to say was that a non-blind test isnt
necessarily invalid in terms of testing gross changes ;-)

Thank you for translating. :-)


For tiny changes on the threshold of perception, you must use blind
testing.

It seems to me that the idea of blind testing is that it assumes the
listener is unable to separate what he hears from what he thinks he hears -
that the piece of kit that is considered to be 'better' actually does sound
better to him, and that blind testing eliminates such preconceptions or
biases and permits a more impartial comparison.

While I realise that a biased opinion can occur in a non-blind test, I don't
think it's neccessarily so. If there is a detectable difference between,
say, the sound of a ride cymbal played on two different bits of kit, then
that difference must be a detectable one, regardless of whether or the
listener knows which bit of kit is being used. Were it otherwise, then the
difference wouldn't be detectable in either kind of test.

In other words, the problem isn't some inherent drawback in the nature of a
given test, but in the approach or attitude of the tester. That
notwithstanding, it's my view that, by the time the differences are so small
that blind testing is deemed to be neccessary, then the differences are so
small that they're neither here nor there.


In your case it was 'crap DVD player' vs 'decent DAC' which isnt
exactly a subtle difference ;-)

I refer the honourable gentleman to my later posting, in which I bring ****e
and sugar into the discussion. :-)


--
Wally
www.art-gallery.myby.co.uk
On webcam: Black Cat In Coal Cellar

"Wally" wrote in message


It seems to me that the idea of blind testing is that it assumes the
listener is unable to separate what he hears from what he thinks he
hears - that the piece of kit that is considered to be 'better'
actually does sound better to him, and that blind testing eliminates
such preconceptions or biases and permits a more impartial comparison.

That is a well known fact in the cases where the difference is subtle or
controversial.

While I realise that a biased opinion can occur in a non-blind test,
I don't think it's neccessarily so.

Depends on the size of the difference.

If there is a detectable
difference between, say, the sound of a ride cymbal played on two
different bits of kit, then that difference must be a detectable one,
regardless of whether or the listener knows which bit of kit is being
used. Were it otherwise, then the difference wouldn't be detectable
in either kind of test.

Again, you are running way wide of the topic at hand.

In other words, the problem isn't some inherent drawback in the
nature of a given test, but in the approach or attitude of the
tester.

The problem primarily depends on the size of the difference. There are many
cases where people have estimated low when they estimated the size of the
difference, and were later on highly embarassed or at least mislead.

That notwithstanding, it's my view that, by the time the
differences are so small that blind testing is deemed to be
neccessary, then the differences are so small that they're neither
here nor there.

There is quite a bit of middle ground. If you want to experience it for
yourself, please visit www.pcabx.com.

In your case it was 'crap DVD player' vs 'decent DAC' which isnt
exactly a subtle difference ;-)

I refer the honourable gentleman to my later posting, in which I
bring ****e and sugar into the discussion. :-)

Most just ****e.

It's clear that you're speaking quite hypothetically about blind tests - you
don't have any relevant audio experiences. I've just told you how to get
wised up for free and in complete privacy.