"Super-CD" ... soon ?
j...@gnv.ifas.ufl.edu
less-than-perfect sound reproduced by digital methods (CDs in partic)
I have been hearing of a new device that might put an end to the
unhappiness. Apparently, a blue(ish) semiconductor laser has been
made lately. Most CD players use an infrared laser which outputs at
around 1000 nanometers. This new device hits the blue-green area of
the spectrum, around 500 nm. What this will allow is for a CD pit of
half the current physical size to be resolved, immediately doubling
the information capacity of a CD or CD-ROM.
Higher resolution should also allow the inter-track distances to be
tightened-up. We can assume that tracking mechanisms/methods will
constantly improve so it should be no prob to cope with this.
What all this makes possible is the "SUPER-CD". Such a unit offers three
major capabilities. Firstly, it should be able to play existing disks,
which would be no problem. Secondly, it could play disks recorded at
the current sampling rate/depth, but which contain twice (or more)
program material. Thirdly, an 'audiophile' pressing could be made that
uses the extra capacity to store music recorded at a much higher
sampling rate & depth than the current standard. This extra resolution
should negate any difference in sound quality between the beloved
vinyl disks of old and digital.
Given that downwards compatibility should be easy ... I see no reason
that "Super CD" could not take-off the moment the new lasers are
availible in bulk. Given that we would also hear the end of the eternal
snobby complaining about inferior digital sound - the effort would
be well worth it. Sell the first 50,000 units at $2500 each to the
dedicated and well-financed audiophiles - and then cut the price to
$199.95 for the mass market :-)
-- Jim Mason
-- Analog must die !
Tim Channon
>less-than-perfect sound reproduced by digital methods (CDs in partic)
>I have been hearing of a new device that might put an end to the
If 'at least 5 yrs' is soon!
We have not hit the limit of 16 bit linear PCM yet. The entire record and
replay chain is nowhere near that yet.
Do you have any idea of what >90dB dynamic range really means?
Taking the upper limit as SPL of 110dB at which you will suffer permanent
hearing damage if sustained and subtracting 90, you get SPL of 20dB.
Anyone care to give examples of SPL 20dB?
What is the dynamic range of a large orchestra?
TC.
E-mail: tcha...@black.demon.co.uk or tcha...@cix.compulink.co.uk
Martin Krueger
Channon) writes:
>
>Do you have any idea of what >90dB dynamic range really means?
>
I've once seen a chart explaining the usable dynamic range of a compact disc in
professional use (with 16 bit depht). What I remember is :
1) On the upper end of the theoretical 90dB range they have to cut off 10dB as
a clipping reserve, since digital clipping would immediatly destroy a whole
recording.
2) You cannot really use the lower 30-40dB of the 90dB range, since distorsion
becomes too high (up to 40% in the first 10db range, as far as I remember). The
reason for this is the decreasing resolution of the sampling device at lower
input levels.
Thus the reamining usable dynamic range of a compact disc doesn't exceed
40-50dB, that is still a lot but can be outperformed by analog recordings.
M A R T I N
Joern Yngve Dahl-Stamnes
I think it depend of what kind of music that is recorded. Hard rock music
may need the upper 10 dB for clipping reserve, while classic music don't
need that much. My opinion is that there is more than 40-50 dB dynamic
range on a classic CD.
I got one classic CD with church organ music and I can hear the birds
singing outside the church while in the organ is silent. And on another CD
I can here papers sheets beeing moved. I have never heard it on LP's.
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+--+-+-+-+-+-+-+
* I went back to my mother, * Joern Yngve Dahl-Stamnes *
* I said, "I'm crazy ma, help me!" * University of Trondheim *
* She said, "I know how it feels son, * The Norwegian Institute of Tech. *
* 'cause it runs in the family." * Division of Physical Electronics *
* - The Who, Quadrophenia * e-mail: da...@fhydra.dnet.unit.no *
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+--+-+-+-+-+-+-+
Tim Channon
>>
>
>I've once seen a chart explaining the usable dynamic range of a compact disc> in
>professional use (with 16 bit depht). What I remember is :
>
>1) On the upper end of the theoretical 90dB range they have to cut off 10dB> as
>a clipping reserve, since digital clipping would immediatly destroy a whole
>recording.
Sorry to disagree but to illustrate: I asked a Polygram engineer during a
live recording session how it was possible to record to apparently <1dB of
clip. He shrugged and just replied "It's ok if they play in tune" 8-)
That was a few years ago when the recording was onto 1610. Since then
recordings at > 16 bit are possible if not common, but your point is valid
for unpredictable sound sources.
A long time ago we designed a limiter for use before A/D -- it has never sold
for the purpose.
>2) You cannot really use the lower 30-40dB of the 90dB range, since> distorsion
>becomes too high (up to 40% in the first 10db range, as far as I remember).> The
>reason for this is the decreasing resolution of the sampling device at lower
>input levels.
Sorry. Fade a signal down to -80dB and listen. Just some added hiss and _no_
noticable distortion. This is what dither is all about. I have a digital
fader (re-dithered) which sounds ok when the music sound is at the same level
as the noise. Switch the dither off and there is your distortion (or more
precisely quantisation noise which sounds a bit like crossover distortion to
me).
Even if there is distortion, it would have to be very bad to exceed the noise
floor.
All this does of course assume that the CD DAC is good, but that is not the
subject of this thread.
>
>Thus the reamining usable dynamic range of a compact disc doesn't exceed
>40-50dB, that is still a lot but can be outperformed by analog recordings.
>
Perhaps I should pass you the fader handheld when it's faded also to zero but
the monitor gain is right up.
What is 80dBs above 1 watt per channel?
Jim Cathey
>1) On the upper end of the theoretical 90dB range they have to cut off 10dB
>as a clipping reserve, since digital clipping would immediatly destroy a
>whole recording.
Who says that the original recording has to be done with (only) 16 bits?
Using 17 would make the problem go away --- just post-process the recording
to do a 'nice' clip and/or scale the recording to fit the intended output
medium.
Don't mistake what may be current practice for an inherent limitation.
--
+----------------+
! II CCCCCC ! Jim Cathey
! II SSSSCC ! ISC-Bunker Ramo
! II CC ! TAF-C8; Spokane, WA 99220
! IISSSS CC ! UUCP: uunet!isc-br!jimc (ji...@isc-br.isc-br.com)
! II CCCCCC ! (509) 927-5757
+----------------+
"PC's --- the junk bonds of the computer industry"
System admin
>I got one classic CD with church organ music and I can hear the birds
>singing outside the church while in the organ is silent. And on another CD
>I can here papers sheets beeing moved. I have never heard it on LP's.
On my second-ever CD (Jean-Michelle Jarre, The Essential), on whichever
track it is that starts with the thunderclap you can very clearly hear
the rain against the window before the bang by turning up the volume.
I never even knew it was there before because it had disappeared into
the tape/vinyl hiss.
Damon
Martin Krueger
Channon) writes:
>Sorry to disagree but to illustrate: I asked a Polygram engineer during a
>live recording session how it was possible to record to apparently <1dB of
>clip. He shrugged and just replied "It's ok if they play in tune" 8-)
And if not they use the master-tape instead...
>Sorry. Fade a signal down to -80dB and listen. Just some added hiss and _no_
>noticable distortion. This is what dither is all about. I have a digital
>fader (re-dithered) which sounds ok when the music sound is at the same level
>as the noise. Switch the dither off and there is your distortion (or more
>precisely quantisation noise which sounds a bit like crossover distortion to
>me).
Adding HISS ? I was always told the very big advantage of digital recordings is
the absence of hiss. Just another way of restrecting the dynamic range on the
lower end.
>What is 80dBs above 1 watt per channel?
A lot, but what is 40dB's above 1 watt?
Greetings
MARTIN
Dave Martindale
>the absence of hiss. Just another way of restrecting the dynamic range on the
>lower end.
Yes, you add white noise (hiss) at a level equivalent to the amplitude of
the least significant bit in the 16-bit word. This is at -96 dB, completely
inaudible under most music.
This is far below the level of hiss from analog magnetic tape - thus the
claim of "absence of hiss". It isn't absolutely absent, but it's at a
far lower level.