HDCD re-encoding
 

In article , Adrian C
wrote:
Jim Lesurf wrote:
That would not prevent, say, an SPDIF output from giving you the
(encoded) series of 16bit values as they are present on the CD. Ditto
for reading the CD with something like a computer. Neither of these
processes would 'decode' the HDCD data, just copy it with the bits in
their 'encoded' pattern.

I can't recall seeing a full and detailed technical description of the
HDCD system, so for all I know, though, the non-audio data on the disc
may contain some 'extra' info that tells the HDCD decoder 'this is an
HDCD disc' in such a way that a simple 'clone' of the audio data would
not duplicate.

Copying CDs on a computer does work, the HDCD is intact.

I did a test, recorded rear panel SPDIF from my HDCD Rotel RCD-971
playing a HDCD disc, to a computer (SPDIF interfaced, true 44.1KHz no
soundcard internal digital bus resampling). The Wav file has LSB info
(did hex dump), but on burning it to a CD and replaying it - the HDCD
decoder doesn't kick in. It could be my SPDIF to Wav process is up the
spout though (cheap soundcard). Be interested if anyone else has had
success this way, or has my CD player indeed knobbled the SPDIF output?

OK. That is quite interesting as it implies the decoder also wants some of
the non-audio info on the disc to 'flag' that it is HDCD encoded and enable
the decoder.

Anyway, the HDCD patent 5479168 says this,

"The command codes and other auxiliary data are encrypted with a
pseudo-random noise and inserted into the least significant bit of the
main signal digital words in a serial fashion, one bit per word. The LSB
of the audio is replaced by a "random" noise for the duration of the
control insertion. (Of course, more than one bit could be "borrowed" for
this purpose, but more of the main program would be lost.) The system
is set up so that when the control channel is not needed, the LSB
carries the normal audio signal.

Thanks for finding the patent and quoting the number. (Is it the USA patent
number?) The above is consistent what I've read in general descriptions in
the past.

Since the digital to analog converters in most of the current generation
of digital audio products are not accurate to 16 bits,

Interesting assertion. :-)

the loss of the 16th bit will not be audible during undecoded playback,
as long as the information inserted there has noise-like properties.
Even in high quality systems which do resolve all 16 bits, the insertion
is not normally audible because the LSB of most programs already has
very noise-like properties. The low level gain compression and dynamic
dither described previously raise the level of the program during very
quiet periods and help hide the code insertions during those program
conditions under which they might be noticeable. In typical classical
music programming, the control signal would be inserted for intervals of
about a millisecond each occurring several times per second at most. The
loss of full program resolution for these brief intervals is not
noticeable."

Thanks for the above. In some ways it makes it sound akin to 'NICAM' in
terms of being a quasi-floating-point approach.

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

HDCD re-encoding
 

In article , Arny Krueger
wrote:


BTW this manual gives additional information about HDCD processing:

http://www.euphonix.com/support/manu...op_man_301.pdf

Thanks for the above. I'll get a copy. :-)

Particularly note page 35-37 of the document, the paragraph entitled:
"Low Level Extension"


Peak Extension & HDCD Limiting

"Peak Extension is a restorable (with HDCD decoding) soft peak limiter
that allows peak levels up to 6 dB above standard full scale level (+6
dBfs) on HDCD 16-bit recordings without generating overs". The limiter
has a carefully crafted easy-over" curve, designed to mimic the sound
of analog tape saturation that operates over an input signal level
range of -3 dBfs to +6 dBfs, in effect squeezing the top 9 dB of the
input signals range into the top 3 dB of the 16-bit recording.

Not encountered that before in descriptions of HDCD. Not particularly a
feature I'd welcome as it may encourage those making recordings to
over-record and cause needless level compression.

Also, if the implication is that the HDCD decoder 'expands' this
compression back out again, I am puzzled since the above implies that the
CD player would either always have to have its output scaled down to allow
for this not the clip the output stages, or the outputs must be able to
deliver levels well above the 0dB level (typically 2V for normal players)
without clipping. Can't recall any reviews that mention this. Curious.

"During HDCD 16-bit decoded playback, Peak Extension peak limiting is
undone by the HDCD decoder using a precisely mapped inverse of the
limiting curve controlled by the hidden LSB code, and the dynamics of
the original material are restored up to +6 dBfs, thus extending
dynamic range.

Ah, that implies the above would be the case. Wonder why I've never seen a
review of an HDCD player mention this being measured in practice... Perhaps
they never use it.

Low Level Extension

"Low Level Extension is an average signal level based low level
compression / expansion system used on HDCD 16-bit amplitude encoded
recordings which very gradually raises gain a preset amount when the
average signal level drops below a preset threshold. During HDCD 16-bit
decoded playback the compression curve is expanded back to linear gain
by the HDCD decoder using a precisely mapped inverse of the compression
curve controlled by a hidden code, producing a dynamic range and
resolution floor beyond 16-bit. During undecoded playback low level
information normally lost by standard 16-bit players is preserved,
providing more accurate timbral and spatial reproduction.

Which seems a little like a quasi-floating-point arrangemen with the top 15
bits being the mantissa and the pattern of the LSB being the exponent,
operated in blocks.

"There are two modes of Low Level Extension, Normal" and Special".
Normal mode begins to affect the input signal 45 dB below peak level,
gradually raising the gain 4 dB as the level drops over an 18 dB range.
Special mode begins to affect the input signal 39 dB below peak level,
and gradually raises the gain 7.5 dB over a 26 dB range.

I must admit that this doesn't seem particularly impressive to me. Using a
sliding gain like this to get a nominal increase in range of only 7 - 12 dB
does not seem a lot when noise-shaping could drop the audible noise floor
and effective resolution by similar (or greater) amounts with no need for a
'special decoder' or the risk that players with no decoder will be
adversely affected. Perhaps this is why the idea never really caught on?



Normal mode is
optimized to provide the best combination of decoded dynamic range and
resolution and undecoded compatibility. Special mode is designed to
provide the best possible decoded dynamic range and resolution at some
potential expense of undecoded compatibility. To access Special mode,
from the Operating Menu select (SETUP/OUTPUT/HDCD_16/LOWLVL/ SPECIAL).
Typically, Special mode is used only for HDCD 16-bit master tracking
with the assumption that the recording will be decoded by the Model Two
to a 24-bit or 20-bit word length for digital post production before
being re-encoded to HDCD 16-bit using Normal mode to produce a release
master.

Thanks for the above info! :-)

However I suspect that I'll spend some time (when I can) trying to
understand HDCD, only to put it in the 'dead-on-its-feet' category like
SACD and DVD-A! :-)

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

HDCD re-encoding
 

In article , Glenn Richards
wrote:
Jim Lesurf wrote:

It is an interesting question. However for myself, the main 'bar' to
giving any answer is the absence of detailed technical information on
HDCD that would let me determine what *actual* effect the process
has... I've read all sorts of claims and general descriptions over the
years that 'describe' HDCD, but none that would enable any analysis,
etc. (Indeed, the descriptions I've seen at different times
contradicted one another!)

The effect seems to be one of a compander. Another post mentioned
something about soft peak limiting, which would resolve with the
observed effects. Many HDCD discs sound squashed dynamically when played
on a non-HDCD player.

If so, I doubt that the people at 'HDCD inc' would welcome us knowing this!
It implies that most of (who don't have players with HDCD players) should
avoid such encoded CDs as they will exhibit audible compression which could
be avoided with correctly recorded non-HDCD discs! Not exactly an advert
for people to buy HDCDs...

According to the manual SBM uses "a form of noise shaping to encode an
effective resolution of 24 bits into the 16-bit medium". Whatever the
jargon, it works - analogue recordings made from HDCD do sound more
detailed.

Again, if so, this is 'bad news' for the people at 'HDCD Inc'. It is quite
easy for people making professional recordings to employ noise shaping.
Indeed, I'd expect this to be quite common. The Sony SBM is essentially
just one proprietary version of this. Hence it indicates that there is no
need for anyone making professional CD recordings to use HDCD - and by
doing so have to pay fees, and degrade the results on most (non-HDCD)
players.

I have had my doubts about HDCD being worthwhile. What you say leads me to
feel I should avoid any HDCD discs like the plague. :-) Certainly, if I
were a professional CD producer I would do so, given what you say...

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

HDCD re-encoding
 

Jim Lesurf wrote:
Thanks for finding the patent and quoting the number. (Is it the USA patent
number?) The above is consistent what I've read in general descriptions in
the past.

No problem Jim. Yes, it's the USA patent number.

Go to the following URL and enter 5479168 in the Query box.
http://patft1.uspto.gov/netahtml/PTO/srchnum.htm

--
Adrian C

HDCD re-encoding
 

Jim Lesurf wrote:

The effect seems to be one of a compander. Another post mentioned
something about soft peak limiting, which would resolve with the
observed effects. Many HDCD discs sound squashed dynamically when
played on a non-HDCD player.
If so, I doubt that the people at 'HDCD inc' would welcome us knowing
this! It implies that most of (who don't have players with HDCD
players) should avoid such encoded CDs as they will exhibit audible
compression which could be avoided with correctly recorded non-HDCD
discs! Not exactly an advert for people to buy HDCDs...

Perhaps. But I think the idea behind it (just playing devil's advocate
for a moment) is to allow the "loudness war" to take place outside the
realms of the HDCD encoded signal.

Or to put it another way, a non-HDCD disc will exhibit audible
compression as collateral damage from the loudness war.

An HDCD disc, on the other hand, will exhibit the same degree of
compression as collateral damage - until you play it back on an
HDCD-equipped CD player. At which point the compression is reversed and
the dynamics are restored to the music.

Of course none of this should be necessary, if it wasn't for a bunch of
cloth-eared twits in the music industry that insist on making everything
as loud as possible.

Just my £0.02 worth, anyway.

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

IT consultancy, hardware and software support, broadband installation

HDCD re-encoding
 

In article , Glenn Richards
wrote:
Jim Lesurf wrote:

The effect seems to be one of a compander. Another post mentioned
something about soft peak limiting, which would resolve with the
observed effects. Many HDCD discs sound squashed dynamically when
played on a non-HDCD player.
If so, I doubt that the people at 'HDCD inc' would welcome us knowing
this! It implies that most of (who don't have players with HDCD
players) should avoid such encoded CDs as they will exhibit audible
compression which could be avoided with correctly recorded non-HDCD
discs! Not exactly an advert for people to buy HDCDs...

Perhaps. But I think the idea behind it (just playing devil's advocate
for a moment) is to allow the "loudness war" to take place outside the
realms of the HDCD encoded signal.

Hadn't thought of that. :-) Yes, you may be right.

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

HDCD re-encoding
 

In article , Adrian C
wrote:
Jim Lesurf wrote:
Thanks for finding the patent and quoting the number. (Is it the USA
patent number?) The above is consistent what I've read in general
descriptions in the past.

No problem Jim. Yes, it's the USA patent number.

Go to the following URL and enter 5479168 in the Query box.
http://patft1.uspto.gov/netahtml/PTO/srchnum.htm

Thanks for the above. :-)

I've been pondering writing sometime a "quasi-dead digital formats" article
sometime to make use of all the analysis I did of SACD ages ago. The info
that has come out of this discussion has set me thinking about that again,
and the idea of adding in HDCD...

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

HDCD re-encoding
 

Jim Lesurf wrote:

Perhaps. But I think the idea behind it (just playing devil's
advocate for a moment) is to allow the "loudness war" to take place
outside the realms of the HDCD encoded signal.
Hadn't thought of that. :-) Yes, you may be right.

:-)

Or maybe I'm just too cynical...

Anyway, have a look at this. These are taken from Mark Chesnutt's 1997
album "Thank God For Believers". This is the final track on the album,
"It's Not Over (If I'm Not Over You)", featuring Vince Gill and Alison
Krauss.

Yes, in case you're wondering from that title, it covers both kinds of
music - country AND western! But anyway...

This is a screenshot from Nero Wave Editor of the MP3 ripped straight
from the original CD with no HDCD decoding. Ripped using cdparanoia and
encoded using LAME as VBR, then processed with MP3Gain for volume
levelling, with the target volume at 89dB:

http://intranet.squirrel-net.co.uk/s...ack10-cdda.png

The following screenshot is the same track, this time copied onto a
CD-RW from the analogue outputs of an Arcam DV-79 (which has HDCD
decoding), recorded onto a Sony CD recorder with Super Bit Mapping
enabled. Ripping and encoding was done using the same software
(cdparanoia/LAME) with identical settings, then volume levelled to 89dB
using MP3Gain. Note the increased dynamic range on this version:

http://intranet.squirrel-net.co.uk/s...ack10-hdcd.png

And just for comparison, here's a screen grab from Patty Loveless' 1988
album "Honky Tonk Angel", this is track number 9, "Timber I'm Falling In
Love":

http://intranet.squirrel-net.co.uk/s...ots/timber.png

Again, this is volume levelled to 89dB... those were the days.

And from 2005, Kelly Clarkson's "Breakaway":

http://intranet.squirrel-net.co.uk/s.../breakaway.png

Also volume levelled to 89dB...

Anyone else think that the perpetrators of the loudness war should be
taken outside and shot?

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

IT consultancy, hardware and software support, broadband installation

HDCD re-encoding
 

Glenn Richards wrote:

Or maybe I'm just too cynical...

Far from it. You are quite the opposite of cynical.

None of my HDCDs seem to use the optional "Headroom Extension"
but it is difficult to tell by looking at the waveform as it only
applies when the signal is between 0.7 and 1 (or between -1 and -0.7).

It seems that if you use Total Recorder Professional between
Windows Media Player and a 24-bit soundcard you can capture the 24-bit
digitally decoded signal - at least it has something in the lowest 8 bits.

Anyone know which HDCDs, if any, actually use "Headroom Extension"?

--
Eiron

No good deed ever goes unpunished.

HDCD re-encoding
 

Eiron wrote:

Or maybe I'm just too cynical...
Far from it. You are quite the opposite of cynical.

Oh, trust me, on certain things I'm very cynical. Did you look at the
screenshots that I posted the URLs to?

If you want real cynicism, look at the last URL I posted (the 2005 pop
recording). No dynamics whatsoever. Oh, yay.

It seems that if you use Total Recorder Professional between Windows
Media Player and a 24-bit soundcard you can capture the 24-bit
digitally decoded signal - at least it has something in the lowest 8
bits.

Not sure if my soundcard is 24-bit, I don't think it is. Can't even
remember what's in this PC atm, I vaguely recall it's a Creative SB
Live, which I don't think is 24-bit.

Might be worth upgrading at some point. But then again, it might not. I
can now capture the decoded HDCD signal by making an analogue copy onto
a standalone CD recorder... which does sound noticeably better (less
compressed).

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

IT consultancy, hardware and software support, broadband installation