USB turntable
 

On 2006-12-17, Laurence Payne lpayne1NOSPAM@dslDOTpipexDOTcom wrote:
What does Super Bit Mapping claim to do?

SBM seems to be Sony's name for their proprietary noise-shaped dither.

The noise shaping allows lower-level signals on a CD to be audible,
because the noise floor is shaped to put most of the noise energy at
frequencies where the ear is not very sensitive.

--
John Phillips

USB turntable
 

On Sun, 17 Dec 2006 12:19:48 +0000, Laurence Payne
lpayne1NOSPAM@dslDOTpipexDOTcom wrote:

On Sun, 17 Dec 2006 11:12:34 +0000, Glenn Richards
wrote:

What does Super Bit Mapping claim to do?

http://www.swee****er.com/publicatio...nterSN_05.html

Um. The description is typical snake-oil pseudo-science. Were you
having a problem with "digital quantisation noise"? Whatever that
is?

Well, it certainly has nothing whatever to do with bit mapping. All I
could see there was a description of dither noise shaping - a totally
standard technique as far as I am aware. Of course this is unlikely to
be needed at 16 bits as the noise is so low anyway - and that goes in
Spades if you are transcribing from vinyl.

d




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

USB turntable
 

On 17 Dec 2006 12:27:42 GMT, John Phillips
wrote:

What does Super Bit Mapping claim to do?

SBM seems to be Sony's name for their proprietary noise-shaped dither.

The noise shaping allows lower-level signals on a CD to be audible,
because the noise floor is shaped to put most of the noise energy at
frequencies where the ear is not very sensitive.

Hardly necessary for 16-bit recording, surely? Especially with vinyl
as the source :-) Glen - fancy posting some wavs of with and without
samples? If it's doing something you hear as an improvement, I wonder
if they're sneaking in a smile-curve eq ?

USB turntable
 

On 2006-12-17, Laurence Payne lpayne1NOSPAM@dslDOTpipexDOTcom wrote:
On 17 Dec 2006 12:27:42 GMT, John Phillips
wrote:

What does Super Bit Mapping claim to do?

SBM seems to be Sony's name for their proprietary noise-shaped dither.

The noise shaping allows lower-level signals on a CD to be audible,
because the noise floor is shaped to put most of the noise energy at
frequencies where the ear is not very sensitive.

Hardly necessary for 16-bit recording, surely? Especially with vinyl
as the source :-) ...

I was thinking about that. Normally a source with its own high-enough
noise floor, as long as the noise is not correlated with the signal,
should see no benefits from turning on noise shaping for the dither.

Indeed it should not even theoretically benefit from dither of any form
in the recorder as the source is, in effect, pre-dithered.

--
John Phillips

USB turntable
 

In article , Laurence Payne
lpayne1NOSPAM@dslDOTpipexDOTcom wrote:

I looked at that Sony CD recorder you mentioned. Retails at well under
£200. It's not going to have anything particularly wonderful in the ADC
department at that price I'm afraid. And, for real-time recording,
you'll have more and more problems finding media that will burn at 1X.

FWIW I use a Pioneer audio recorder in much the same way as Glenn describes
and it works quite well. (Alas, they seem to have abandoned making such
recorders.) I haven't had any problem as yet finding suitable discs for it.
Although in general I just use the same set of TDK audio CDRW's to record,
then blank them again once I've edited and written the results onto a CDR
on my computer.

My main reasons for doing this are that I don't use the computer in the
same room as the main audio system, and since it isn't either 'IBM
architecture' nor a Mac, I haven't seen any sound cards for it that work
with a level of ADC performance above 'dire'. The recorder is easy to use,
and allows me to treat it just like a would a cassette or reel recorder as
part of the audio system. It is also quite a decent CD drive.

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

USB turntable
 

John Phillips wrote:

I was thinking about that. Normally a source with its own high-enough
noise floor, as long as the noise is not correlated with the signal,
should see no benefits from turning on noise shaping for the dither.

Interesting theory, borne out by something I discovered a few years ago...

In the early days of MP3 I didn't have a CD drive capable of digital
audio extraction. So I hooked my CD player up to the analogue input of
my sound card (Audio Dynamics DMI-50XGS - Jim should know about that
one). I then MP3 encoded the resulting WAV file at 128Kbit.

I later got a CD writer on that system which did support digital audio
extraction (a 6x4x24 speed SCSI drive). Extracted tracks digitally and
encoded, and compared.

The analogue copies sounded better. Far fewer MP3 compression artefacts.

Which, I can only assume, was because of the higher noise floor on those
tracks causing a noise shaping effect. Therefore covering up the
compression artefacts.

This was using BladeEnc in 1999, long before LAME. Nowadays of course
we've got LAME, VBR, noise shaping built in to LAME etc etc.

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

IT consultancy, hardware and software support, broadband installation

USB turntable
 

On Sun, 17 Dec 2006 17:16:15 +0000, Glenn Richards
wrote:

John Phillips wrote:

I was thinking about that. Normally a source with its own high-enough
noise floor, as long as the noise is not correlated with the signal,
should see no benefits from turning on noise shaping for the dither.

Interesting theory, borne out by something I discovered a few years ago...

In the early days of MP3 I didn't have a CD drive capable of digital
audio extraction. So I hooked my CD player up to the analogue input of
my sound card (Audio Dynamics DMI-50XGS - Jim should know about that
one). I then MP3 encoded the resulting WAV file at 128Kbit.

I later got a CD writer on that system which did support digital audio
extraction (a 6x4x24 speed SCSI drive). Extracted tracks digitally and
encoded, and compared.

The analogue copies sounded better. Far fewer MP3 compression artefacts.

Which, I can only assume, was because of the higher noise floor on those
tracks causing a noise shaping effect. Therefore covering up the
compression artefacts.

This was using BladeEnc in 1999, long before LAME. Nowadays of course
we've got LAME, VBR, noise shaping built in to LAME etc etc.

Sorry Glenn, but this is gibberish - you are confusing two totally
separate effects. If the noise floor of a piece of material was loud
enough to make a difference, it would make the resultant compression
vastly worse because a) noise can't be compressed and b) the encoder's
attempts to compress it would steal vital data bits from the wanted
music.

This effect can make itself felt hen you try to create an MP3 from bad
vinyl. It will carefully encode the ticks, pops and hiss at the
expense of poor encoding of the music - definitely a bad move.

So no, the noise floor of either is low enough that it would be well
into the masking zone, and neither encoder would be bothering to try
to do anything with it, or high enough to be comfortably audible.

And of course analogue noise doesn't have any noise shaping effect 0-
it simply is what it is, typically white-ish.

Of course the fact is that for most sources, the background noise is
much higher than any deliberately applied dither signal, so it
dominates and controls the dithering process. The result is that the
noise coming out sounds essentially identical to the analogue noise
going in.

d

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

USB turntable
 

Laurence Payne wrote:

Hardly necessary for 16-bit recording, surely? Especially with vinyl
as the source :-) Glen - fancy posting some wavs of with and without
samples? If it's doing something you hear as an improvement, I wonder
if they're sneaking in a smile-curve eq ?

I've tested it with copying an HDCD-encoded disc to get the decoded
signal on a normal CD and there is an improvement. Haven't tried "with"
and "without" tests with vinyl, might give that a go when I get home.
(At a friend's atm so no vinyl playback.)

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

IT consultancy, hardware and software support, broadband installation

USB turntable
 

Don Pearce wrote:

This was using BladeEnc in 1999, long before LAME. Nowadays of course
we've got LAME, VBR, noise shaping built in to LAME etc etc.
Sorry Glenn, but this is gibberish - you are confusing two totally
separate effects. If the noise floor of a piece of material was loud
enough to make a difference, it would make the resultant compression
vastly worse because a) noise can't be compressed and b) the encoder's
attempts to compress it would steal vital data bits from the wanted
music.

I just read my last posting... about as clear as mud... that'll teach me
to post when I'm on the verge of falling asleep! (Yesterday was a long day.)

But the effect was that stuff captured by analogue transfer from CD to
sound card was largely free of compression artefacts, whereas a
digitally extracted version was "artefact city". Not a problem nowadays
though with modern MP3 encoders though.

This effect can make itself felt hen you try to create an MP3 from bad
vinyl. It will carefully encode the ticks, pops and hiss at the
expense of poor encoding of the music - definitely a bad move.

Ah yes. Heard this a few times, dodgy MP3s I've downloaded that were
obviously ripped from vinyl. Use the de-clicker function in most audio
software before encoding and it does a far far better job.

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

IT consultancy, hardware and software support, broadband installation

USB turntable
 

Dave Plowman (News) wrote:

I dunno the relative manufacturing costs of MM versus ceramic. This
sort of thing has shifted dramatically with CAD machinery.

CAM - the machinery bit is Computer Aided Manufacture. CAD is the Design bit
wot is used to produce data that gets fed to the CAM bit.


--
Wally
www.wally.myby.co.uk
You're unique - just like everybody else.