Don Pearce wrote:

A good way to measure transparency would be to make multiple passes
through the recording chain to see what gets nasty. Given a CD you can
make copy after copy, through hundreds of generations if you wish,
without a single change in the sound that is produced at the far end.
You can't do that with analogue tape or vinyl - in a very few
generations the thing will have degenerated to unusability. Now THAT
is non-transparent. If you want a glass analogy, imagine stacking many
sheets together and finding how many it takes to render the view
invisible, with the analogue glass, it is a very small stack. For the
digital glass, you can just go on stacking.

d


Thats true only if you stay within the digital domain, a more
representive test would be to connect a CDP to a CDR via its analog
connection, and see how the recording changes with a number of generations.

--
Nick

On Thu, 31 Aug 2006 11:17:52 +0100, Nick Gorham
wrote:

Don Pearce wrote:

A good way to measure transparency would be to make multiple passes
through the recording chain to see what gets nasty. Given a CD you can
make copy after copy, through hundreds of generations if you wish,
without a single change in the sound that is produced at the far end.
You can't do that with analogue tape or vinyl - in a very few
generations the thing will have degenerated to unusability. Now THAT
is non-transparent. If you want a glass analogy, imagine stacking many
sheets together and finding how many it takes to render the view
invisible, with the analogue glass, it is a very small stack. For the
digital glass, you can just go on stacking.

d


Thats true only if you stay within the digital domain, a more
representive test would be to connect a CDP to a CDR via its analog
connection, and see how the recording changes with a number of generations.

Fair enough, I would be happy with that. The result would not be
hugely different even in this case.

d

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

On Thu, 31 Aug 2006 09:39:22 GMT, (Don Pearce)
wrote:

The question of transparency of the CD goes in another direction too.
That would be how accurately you can get back what you record on it.
From that point of view it has to be said that it can be entirely
transparent. If you use a programme like Exact Audio Copy it will tell
you how many errors it has found, and how it has dealt with them. All
my CDs return zero errors, so they are transparent.


Red herring. If the conversion to numbers and back to sound is
non-transparent, it's irrelevant whether you can make perfect copies
of the media.

On Thu, 31 Aug 2006 11:55:02 +0100, Laurence Payne
lpayne1NOSPAM@dslDOTpipexDOTcom wrote:

On Thu, 31 Aug 2006 09:39:22 GMT, (Don Pearce)
wrote:

The question of transparency of the CD goes in another direction too.
That would be how accurately you can get back what you record on it.
From that point of view it has to be said that it can be entirely
transparent. If you use a programme like Exact Audio Copy it will tell
you how many errors it has found, and how it has dealt with them. All
my CDs return zero errors, so they are transparent.


Red herring. If the conversion to numbers and back to sound is
non-transparent, it's irrelevant whether you can make perfect copies
of the media.

Did you stop reading at that point? I went on to explain in the second
paragraph.

d

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

On Thu, 31 Aug 2006 10:20:36 GMT, (Don Pearce)
wrote:

On Thu, 31 Aug 2006 11:17:52 +0100, Nick Gorham
wrote:

Don Pearce wrote:

A good way to measure transparency would be to make multiple passes
through the recording chain to see what gets nasty. Given a CD you can
make copy after copy, through hundreds of generations if you wish,
without a single change in the sound that is produced at the far end.
You can't do that with analogue tape or vinyl - in a very few
generations the thing will have degenerated to unusability. Now THAT
is non-transparent. If you want a glass analogy, imagine stacking many
sheets together and finding how many it takes to render the view
invisible, with the analogue glass, it is a very small stack. For the
digital glass, you can just go on stacking.

d


Thats true only if you stay within the digital domain, a more
representive test would be to connect a CDP to a CDR via its analog
connection, and see how the recording changes with a number of generations.

Fair enough, I would be happy with that. The result would not be
hugely different even in this case.

Changed my mind. No it is not reasonable to go back in and out of the
analogue domain for this test. The question is the transparency of the
CD as a medium, not any coding and decoding processes that may be
associated with it - that is another question entirely.

d

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

"Don Pearce" wrote in message
...
On Thu, 31 Aug 2006 10:20:36 GMT, (Don Pearce)
wrote:

On Thu, 31 Aug 2006 11:17:52 +0100, Nick Gorham
wrote:

Don Pearce wrote:

A good way to measure transparency would be to make multiple passes
through the recording chain to see what gets nasty. Given a CD you can
make copy after copy, through hundreds of generations if you wish,
without a single change in the sound that is produced at the far end.
You can't do that with analogue tape or vinyl - in a very few
generations the thing will have degenerated to unusability. Now THAT
is non-transparent. If you want a glass analogy, imagine stacking many
sheets together and finding how many it takes to render the view
invisible, with the analogue glass, it is a very small stack. For the
digital glass, you can just go on stacking.

d


Thats true only if you stay within the digital domain, a more
representive test would be to connect a CDP to a CDR via its analog
connection, and see how the recording changes with a number of
generations.

Fair enough, I would be happy with that. The result would not be
hugely different even in this case.

Changed my mind. No it is not reasonable to go back in and out of the
analogue domain for this test. The question is the transparency of the
CD as a medium, not any coding and decoding processes that may be
associated with it - that is another question entirely.



OK, glad you said that - not only would it represent more a test of DAC
quality than the medium or coding processes but I do believe (no, really...)
the question of *cable quality* (interconnects) would enter into the
equation.....

On Thu, 31 Aug 2006 12:21:47 +0100, "Keith G"
wrote:


"Don Pearce" wrote in message
...
On Thu, 31 Aug 2006 10:20:36 GMT, (Don Pearce)
wrote:

On Thu, 31 Aug 2006 11:17:52 +0100, Nick Gorham
wrote:

Don Pearce wrote:

A good way to measure transparency would be to make multiple passes
through the recording chain to see what gets nasty. Given a CD you can
make copy after copy, through hundreds of generations if you wish,
without a single change in the sound that is produced at the far end.
You can't do that with analogue tape or vinyl - in a very few
generations the thing will have degenerated to unusability. Now THAT
is non-transparent. If you want a glass analogy, imagine stacking many
sheets together and finding how many it takes to render the view
invisible, with the analogue glass, it is a very small stack. For the
digital glass, you can just go on stacking.

d


Thats true only if you stay within the digital domain, a more
representive test would be to connect a CDP to a CDR via its analog
connection, and see how the recording changes with a number of
generations.

Fair enough, I would be happy with that. The result would not be
hugely different even in this case.

Changed my mind. No it is not reasonable to go back in and out of the
analogue domain for this test. The question is the transparency of the
CD as a medium, not any coding and decoding processes that may be
associated with it - that is another question entirely.



OK, glad you said that - not only would it represent more a test of DAC
quality than the medium or coding processes but I do believe (no, really...)
the question of *cable quality* (interconnects) would enter into the
equation.....

I agree. If you make 100 serial copies with a couple of metres of
cable each time, you have 200 metres of cable in the loop. You also
have 200 connectors instead of two, and as you say 100 DACs and ADCs.

But fair's fair. For the analogue half of this gedankenexperiment I
wouldn't ask for the analogue signals to be converted to digital and
back either. They can also remain in their native form as they are
written to the vinyl.

d

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

Don Pearce wrote:


Thats true only if you stay within the digital domain, a more
representive test would be to connect a CDP to a CDR via its analog
connection, and see how the recording changes with a number of generations.

Fair enough, I would be happy with that. The result would not be
hugely different even in this case.


Changed my mind. No it is not reasonable to go back in and out of the
analogue domain for this test. The question is the transparency of the
CD as a medium, not any coding and decoding processes that may be
associated with it - that is another question entirely.

d


Ok, I can see your point, if the test is as you say, but in reality, it
is a entirly meaningless test in that case. I may as well, say that a
musical score is a entirly transparrent medium,

It can be copied many times without error. Its only when it transfered
to the analog domain (in this case performed) that differences become
important.

--
Nick

In article ,
Nick Gorham wrote:
Changed my mind. No it is not reasonable to go back in and out of the
analogue domain for this test. The question is the transparency of the
CD as a medium, not any coding and decoding processes that may be
associated with it - that is another question entirely.

d


Ok, I can see your point, if the test is as you say, but in reality, it
is a entirly meaningless test in that case. I may as well, say that a
musical score is a entirly transparrent medium,

It can be copied many times without error. Its only when it transfered
to the analog domain (in this case performed) that differences become
important.

Some time ago I had to do some fairly heavy editing on location using only
a pair of domestic MiniDiscs. Now MiniDisc will only allow certain edits,
and the only way round this was to copy via analogue, and edit that copy.
I think I got up to about 10 generations. It still sounded ok - analogue
would have been unusable. And of course MiniDisc is a data reduced system.

--
*Even a blind pig stumbles across an acorn now and again *

Dave Plowman London SW
To e-mail, change noise into sound.

On Thu, 31 Aug 2006 12:45:24 +0100, Nick Gorham
wrote:

Don Pearce wrote:


Thats true only if you stay within the digital domain, a more
representive test would be to connect a CDP to a CDR via its analog
connection, and see how the recording changes with a number of generations.

Fair enough, I would be happy with that. The result would not be
hugely different even in this case.


Changed my mind. No it is not reasonable to go back in and out of the
analogue domain for this test. The question is the transparency of the
CD as a medium, not any coding and decoding processes that may be
associated with it - that is another question entirely.

d


Ok, I can see your point, if the test is as you say, but in reality, it
is a entirly meaningless test in that case. I may as well, say that a
musical score is a entirly transparrent medium,

Not perfectly - I would guess that very few scores were ever
transcribed without error. Of course if you digitize them........ ;-)

It can be copied many times without error. Its only when it transfered
to the analog domain (in this case performed) that differences become
important.

Can be, but is it? That is the question.

d

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