Sign in with
Sign up | Sign in
Your question

Comparing quality on vinyl with Digital

Last response: in Home Audio
Share
Anonymous
July 27, 2004 8:17:22 AM

Archived from groups: rec.audio.high-end (More info?)

Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
off to a good start......

http://www.stevehoffman.tv/forums/showthread.php?t=3711...

________
Thom
July 28, 2004 3:35:07 AM

Archived from groups: rec.audio.high-end (More info?)

thomh wrote:
> Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
> off to a good start......
>
> http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>
> ________
> Thom
>

Wow, very impressive.

Didn't someone ask what a Hoffmanite is? Here is a definitive example of
a post from a Hoffmanite, IMO of course:

"For instance, Steve's Elvis 24KT Hits or Nat King Cole DCC CD's. Has
there been another companies Vinyl, SACD, or CD that has even come close
to Steve's redbook work on these 2 artists? The answer is a definite no.

"Steve has stated in the past that it doesn't matter what format is
being used that he can still get that quality of sound and I believe him."
Anonymous
July 28, 2004 10:21:18 AM

Archived from groups: rec.audio.high-end (More info?)

On Tue, 27 Jul 2004 04:17:22 GMT, "thomh" <thom@paperless.no> wrote:

>Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
>off to a good start......
>
>http://www.stevehoffman.tv/forums/showthread.php?t=3711...

Absolutely hilarious! The real answer to the question - what digital
resolution is closest to vinyl - is not 24/192, 24/96 or 16/44, it's
actually about 12-13 bits on the best day of vinyl's life, with an
effective bandwidth which could be captured by 32k sampling. In the
real world, 16/44 is gross overkill for a vinyl comparison, although
you could argue that 16/96 or 16/192 would make the filtering a lot
easier. It remains the case that more than 16 bits is utterly
unnecessary in *any* playback medium.

Some of the supposed 'engineers' posting on Hoffman's forum clearly
have no understanding of resolution, or of how digital audio actually
works. The classic blunders of 'analogue has infinite resolution' and
'digital has stairsteps' are alive and well in Hoffman's little world!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Related resources
Anonymous
July 28, 2004 10:21:51 AM

Archived from groups: rec.audio.high-end (More info?)

thomh <thom@paperless.no> wrote:
> Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
> off to a good start......

> http://www.stevehoffman.tv/forums/showthread.php?t=3711...

It never ends, does it?

"With 24 bits you are still limited (as with any n-number of bits) to a
finite amount of values to represent the signal within a given period of
time (given by the sampling rate)- in this case 16777216 (2^24). Whereas
with analog you don't have 'steps' (small as they might be). In the
micro-micro-level, there's most likely stuff that still 'slips thru the
cracks' (so to speak) on digital that is properly represented on analog,
dynamics-wise. "

--

-S.
"We started to see evidence of the professional groupie in the early 80's.
Alarmingly, these girls bore a striking resemblance to Motley Crue." --
David Lee Roth
Anonymous
July 28, 2004 10:24:30 AM

Archived from groups: rec.audio.high-end (More info?)

From: chung chunglau@covad.net
>Date: 7/27/2004 4:35 PM Pacific Standard Time
>Message-id: <ce6orb06td@news4.newsguy.com>
>
>thomh wrote:
>> Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
>> off to a good start......
>>
>> http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>>
>> ________
>> Thom
>>
>
>Wow, very impressive.
>
>Didn't someone ask what a Hoffmanite is? Here is a definitive example of
>a post from a Hoffmanite, IMO of course:
>
>"For instance, Steve's Elvis 24KT Hits or Nat King Cole DCC CD's. Has
>there been another companies Vinyl, SACD, or CD that has even come close
>to Steve's redbook work on these 2 artists? The answer is a definite no.
>
>"Steve has stated in the past that it doesn't matter what format is
>being used that he can still get that quality of sound and I believe him."
>
>
>
>
>
>

So you think a Hoffmanite is someone who really likes Hoffman's work? Hmmm I
wonder whether that makes Steve Sullivan a Hoffmanite. I will stae for the
record that Steve Hoffman's vinyl mastering of Elvis 24KT Hits is so much
better than any other version I have heard that it is on the verge of wierd. I
suggest you compare his mastering of this title with any other version. You
might be a Hoffmanite and you don't know it yet.
Anonymous
July 29, 2004 3:35:10 AM

Archived from groups: rec.audio.high-end (More info?)

"S888Wheel" <s888wheel@aol.com> wrote in message
news:yEHNc.201444$XM6.63631@attbi_s53...
> From: chung chunglau@covad.net
> >Date: 7/27/2004 4:35 PM Pacific Standard Time
> >Message-id: <ce6orb06td@news4.newsguy.com>
> >
> >thomh wrote:
> >> Nothing like "audiophiles" discussing vinyl vs. digital to get the
morning
> >> off to a good start......
> >>
> >> http://www.stevehoffman.tv/forums/showthread.php?t=3711...
> >>
> >> ________
> >> Thom
> >>
> >
> >Wow, very impressive.
> >
> >Didn't someone ask what a Hoffmanite is? Here is a definitive example of
> >a post from a Hoffmanite, IMO of course:
> >
> >"For instance, Steve's Elvis 24KT Hits or Nat King Cole DCC CD's. Has
> >there been another companies Vinyl, SACD, or CD that has even come close
> >to Steve's redbook work on these 2 artists? The answer is a definite no.
> >
> >"Steve has stated in the past that it doesn't matter what format is
> >being used that he can still get that quality of sound and I believe
him."
> >
> >
> >
> >
> >
> >
>
> So you think a Hoffmanite is someone who really likes Hoffman's work? Hmmm
I
> wonder whether that makes Steve Sullivan a Hoffmanite. I will stae for the
> record that Steve Hoffman's vinyl mastering of Elvis 24KT Hits is so much
> better than any other version I have heard that it is on the verge of
wierd. I
> suggest you compare his mastering of this title with any other version.
You
> might be a Hoffmanite and you don't know it yet.
>

Well, I do like some of his work but I do *not* consider myself a
Hoffmanite.

Remember that

"There is a big difference between kneeling down and bending over."--
Frank Zappa

_______
Thom
Anonymous
July 29, 2004 3:35:51 AM

Archived from groups: rec.audio.high-end (More info?)

Stewart Pinkerton wrote:
> On Tue, 27 Jul 2004 04:17:22 GMT, "thomh" <thom@paperless.no> wrote:
>
>
>>Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
>>off to a good start......
>>
>>http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>
>
> Absolutely hilarious! The real answer to the question - what digital
> resolution is closest to vinyl - is not 24/192, 24/96 or 16/44, it's
> actually about 12-13 bits on the best day of vinyl's life, with an

hope this isn't a silly question, but how does vinyl have about 12-13 bits of
resolution?


> The classic blunders of 'analogue has infinite resolution' and
> 'digital has stairsteps' are alive and well in Hoffman's little world!

On another topic, when we try to digitise an input signal using a fixed bit
resolution, wouldn't having more bits mean that the increments between levels
become smaller, i.e. more fine grain?

Is 16-bits enough because the human ear can't detect the difference between
1 level out of 65536? So given a 2V peak to peak signal, going up the next level
means an increase of 30 micro volts?
Anonymous
July 29, 2004 3:42:04 AM

Archived from groups: rec.audio.high-end (More info?)

thomh wrote:

> Nothing like "audiophiles" discussing vinyl vs. digital to get the
> morning off to a good start......
>
> http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>
=======================================

And here, from yesterday (27JUL04) is an audiophile on a photography
forum, relating digital vs. analog in the two different fields.

Don't shoot the messenger, it's a quote (and excuse the errors, he is
not primarily an English-speaker):


> I'm sorry to tell, but your ears are for sure contaminated with too
> much digital quantisized music. I had (and have again time to time in
> busy times, since i work in a semi-digital studio) the same problem.
>
> An instrument with few cents out of tune in orchestra can hear even a
> non-trained hear. Doing difference between almost identical
> istruments is just you play them yourself. As a bass player, i do the
> difference between the same instrument but different magnetic heads
> indeed. Can even pick up higher than a third harmonics doing
> difference in tube microphones and condenser type, same with
> transistor amplifyers, mos amps and tube amplifyers. If you hear a
> choir lowering or uppering the original note tone few tenths of
> semitone per 5 minutes, then you got a good hear.
>
> And i sure can tell you - the vinyl sounds better almost in every
> aspect. The quality reduction number 7 for tape and 17 for CD-vinyl
> is from scientific article, that calculated it based on average human
> psychoacoustical capabilities. So there's about plus/minus 5-10 error
> depending on individual level of different trained or not trained
> ears.
>
> Sine you got a Lynn Axis turntable, what needle cartige you have and
> what amplifyer/speakers? From the turntable rest matters the most.
> Well yeah, seismic-feedback turntables definately ARE the best (if
> you have spare 10 000USD to spend), but mosty the cartige matters
> primarly. Please try to do so: have a team of accoustic specialists
> to reconfigure your listening room and making a listening hot-spot
> for you in the room. Buy at least 500+USD hi-fi needle cartige. Use
> A-class (thermodynamically stabilized ones recommended) tube pre and
> post amplifyer, professional quality cables and recommended speakers
> by hear-geniuses. Train your ears with various db levels suited for
> you few weeks with analogue sound only (large tape or LPs), no
> digital please! And i'll guarantee you, you'll make a lot difference
> between the CD (even with a pro 128X oversampling and smoothening DA
> converter) and vinyl on the same system after that.
>
> The human senses need training and have quite long delays on
> adapting, and when contaminating them with lower quality information
> carriers, it needs time to get the levels up again, because when
> being on the same environment for too long, it all seems normal for
> you (same with the digital/analogue photography). For me, i work on a
> studio with 24bit dynamic depth and 96kHz samplerates a typical
> working environment, and even those resolutions contaminate my ears
> when working long time. I need at least one day rest and retraining
> with analogue to adapt my original capabilities again. (Too loud [and
> musically non-professional indeed] music on clubs and partyes is the
> most violent ear-killers indeed. People go with the flow, i use
> professional ear-drops if i really have to be on some of those nasty
> events). When i finish the work in digital, i play final mix via high
> end system and rerecord via selected air room with directional tube
> microphone on the large tape recorder. And that's the result
> satisfying me - removing the gaps between the samples with
> electromagnetic interference. And then puting it into the vinyl for
> respected quality recordcompanyes that still prefer 12"/10"/7" vinyls
> more than anything else.
>
> Convert all the text into the photography, and hope i can deliver my
> point of view on this topic too.
>
> So please understand, film isn't dead, just like 12" are not. Not for
> another 10 year for sure. And those, who say the digital is better in
> quality, i get very mad on. Every thing has it's place - digital too.
> But i'm fully certain that the analogue's place is TOO too
> underestimated nowadays... (btw: i'm a young guy indeed in these
> words, wide open to all the innovative technologys, not
> oldie-oriented)
Anonymous
July 29, 2004 8:34:14 AM

Archived from groups: rec.audio.high-end (More info?)

I seem to remember a link to Christine Tham's analysis of CD vs DVD-A vs
SACD here a few months ago. Anyway, here is her Spectral and Dynamics
Comparisons of LPs vs digital formats.

http://www.audioholics.com/techtips/specsformats/LPsvsC...

________
Thom
July 29, 2004 8:35:51 AM

Archived from groups: rec.audio.high-end (More info?)

S888Wheel wrote:

> From: chung chunglau@covad.net
>>Date: 7/27/2004 4:35 PM Pacific Standard Time
>>Message-id: <ce6orb06td@news4.newsguy.com>
>>
>>thomh wrote:
>>> Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
>>> off to a good start......
>>>
>>> http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>>>
>>> ________
>>> Thom
>>>
>>
>>Wow, very impressive.
>>
>>Didn't someone ask what a Hoffmanite is? Here is a definitive example of
>>a post from a Hoffmanite, IMO of course:
>>
>>"For instance, Steve's Elvis 24KT Hits or Nat King Cole DCC CD's. Has
>>there been another companies Vinyl, SACD, or CD that has even come close
>>to Steve's redbook work on these 2 artists? The answer is a definite no.
>>
>>"Steve has stated in the past that it doesn't matter what format is
>>being used that he can still get that quality of sound and I believe him."
>>
>>
>>
>>
>>
>>
>
> So you think a Hoffmanite is someone who really likes Hoffman's work? Hmmm I
> wonder whether that makes Steve Sullivan a Hoffmanite. I will stae for the
> record that Steve Hoffman's vinyl mastering of Elvis 24KT Hits is so much
> better than any other version I have heard that it is on the verge of wierd. I
> suggest you compare his mastering of this title with any other version. You
> might be a Hoffmanite and you don't know it yet.
>

I believe that a Hoffmanite, the way Steven Sullivan uses that term, is
a very devoted fan of Steve Hoffman, and one who agrees with everything
Hoffman says. I would also say that a Hoffmanite believes that Hoffman
is the ultimate authority on all audio matters, but certainly Mr.
Sullivan may or may not agree with that qualification.

No thanks, I don't think I would ever be a Hoffmanite, since I don't
agree with his views. You are probably a good candidate, though.
Anonymous
July 30, 2004 3:43:28 AM

Archived from groups: rec.audio.high-end (More info?)

On 28 Jul 2004 23:35:51 GMT, Tat Chan <le_king_num_7@hotmail.com>
wrote:

>Stewart Pinkerton wrote:
>> On Tue, 27 Jul 2004 04:17:22 GMT, "thomh" <thom@paperless.no> wrote:
>>
>>
>>>Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
>>>off to a good start......
>>>
>>>http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>>
>>
>> Absolutely hilarious! The real answer to the question - what digital
>> resolution is closest to vinyl - is not 24/192, 24/96 or 16/44, it's
>> actually about 12-13 bits on the best day of vinyl's life, with an
>
>hope this isn't a silly question, but how does vinyl have about 12-13 bits of
>resolution?

13 bits is the equivalent of 78dB dynamic range, which is more than
you'll find on even the best direct-cut virgin vinyl. Resolution is
implicitly tied to dynamic range, system resolution being taken as the
value of the noise floor and commonly expressed as xdB below peak
level.

> > The classic blunders of 'analogue has infinite resolution' and
> > 'digital has stairsteps' are alive and well in Hoffman's little world!
>
>On another topic, when we try to digitise an input signal using a fixed bit
>resolution, wouldn't having more bits mean that the increments between levels
>become smaller, i.e. more fine grain?

Yes. Consider however that there exist no *master* tapes with a
dynamic range of more than 85dB, and you'll see that the 93dB range of
properly dithered 16 bit digital is all you'll ever need. High
sampling rates are another and more complex argument, having nothing
to do with actual signal bandwidth, for which 25kHz will always be
more than enough.

>Is 16-bits enough because the human ear can't detect the difference between
>1 level out of 65536? So given a 2V peak to peak signal, going up the next level
>means an increase of 30 micro volts?

You need to understand that dither makes this distinction untrue, as
it randomises the absolute levels, in the same way that the noise
floor randomises low-level analog signals. As a result, you can hear
tones well below the LSB value in digital, just as you can in analog
below the noise floor. I have recorded and replayed a 3kHz signal at
-105dB on a CD-R, and it was perfectly audible below the smooth -93dB
noise floor that is characteristic of properly made CD. The 'stair
steps' claim is an intuitive urban myth, with no existence in reality.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
July 30, 2004 4:02:24 AM

Archived from groups: rec.audio.high-end (More info?)

On 28 Jul 2004 23:42:04 GMT, Gene Poon <sheehans@ap.net> wrote:

>thomh wrote:
>
>> Nothing like "audiophiles" discussing vinyl vs. digital to get the
>> morning off to a good start......
>>
>> http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>>
>=======================================
>
>And here, from yesterday (27JUL04) is an audiophile on a photography
>forum, relating digital vs. analog in the two different fields.
>
>Don't shoot the messenger, it's a quote (and excuse the errors, he is
>not primarily an English-speaker):

There is however one vital difference here: the resolution of
fine-grain film still far exceeds that possible with even the best
digital cameras. Basic 'Red Book' CD however far exceeds the
resolution of even the best analogue *master* tapes, never mind the
crippled child that is LP. Apples and oranges.

As a case in point, listen to any JVC XRCD. These are all made from
top-quality analogue master tapes, and if you turn up the volume in
the quieter passages, you can quite clearly hear the distinctive
'whoosh' of the analogue tape noise floor. Try that on an LP of the
same recording!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
July 30, 2004 4:04:47 AM

Archived from groups: rec.audio.high-end (More info?)

"thomh" <thom@paperless.no> wrote in message
news:a7%Nc.45975$8_6.7528@attbi_s04...
> I seem to remember a link to Christine Tham's analysis of CD vs DVD-A vs
> SACD here a few months ago. Anyway, here is her Spectral and Dynamics
> Comparisons of LPs vs digital formats.
>
> http://www.audioholics.com/techtips/specsformats/LPsvsC...
>

I, along with a lot of other people, feel that our own CD-Rs made from LPs
sound as good or even better than does the LP source itself. If this be
true, is "red book" CD lacking in any "audible" way?
Anonymous
July 30, 2004 8:13:02 AM

Archived from groups: rec.audio.high-end (More info?)

On 7/28/04 7:35 PM, in article ce9d8n0heg@news4.newsguy.com, "Tat Chan"
<le_king_num_7@hotmail.com> wrote:

> Stewart Pinkerton wrote:
>> On Tue, 27 Jul 2004 04:17:22 GMT, "thomh" <thom@paperless.no> wrote:
>>
>>
>>> Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
>>> off to a good start......
>>>
>>> http://www.stevehoffman.tv/forums/showthread.php?t=3711...
>>
>>
>> Absolutely hilarious! The real answer to the question - what digital
>> resolution is closest to vinyl - is not 24/192, 24/96 or 16/44, it's
>> actually about 12-13 bits on the best day of vinyl's life, with an
>
> hope this isn't a silly question, but how does vinyl have about 12-13 bits of
> resolution?

I think they are talking about dynamic range and the old 6dB per bit rule.
There is a bit more to it IMHO, but who knows!

>
>
>> The classic blunders of 'analogue has infinite resolution' and
>> 'digital has stairsteps' are alive and well in Hoffman's little world!
>
> On another topic, when we try to digitise an input signal using a fixed bit
> resolution, wouldn't having more bits mean that the increments between levels
> become smaller, i.e. more fine grain?
>
> Is 16-bits enough because the human ear can't detect the difference between
> 1 level out of 65536? So given a 2V peak to peak signal, going up the next
> level
> means an increase of 30 micro volts?
Anonymous
July 30, 2004 8:15:57 AM

Archived from groups: rec.audio.high-end (More info?)

On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

> There is however one vital difference here: the resolution of
> fine-grain film still far exceeds that possible with even the best
> digital cameras.

Actually, that is not true. The current 4+ megapixel cameras equal to
exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
range, it approaches or equals medium format film.

The only problem with the digital cameras is color reproduction and
contrast. And that is right now equal to most common films.

>Basic 'Red Book' CD however far exceeds the
> resolution of even the best analogue *master* tapes, never mind the
> crippled child that is LP. Apples and oranges.

Editorial comments aside - using "resolution" as your only metric is not
measuring it in all ways if you are to use your digital camera example.
Anonymous
July 30, 2004 9:33:47 PM

Archived from groups: rec.audio.high-end (More info?)

Stewart Pinkerton <patent3@dircon.co.uk> wrote:
> On 28 Jul 2004 23:35:51 GMT, Tat Chan <le_king_num_7@hotmail.com>
> wrote:

> >Stewart Pinkerton wrote:
> >> On Tue, 27 Jul 2004 04:17:22 GMT, "thomh" <thom@paperless.no> wrote:
> >>
> >>
> >>>Nothing like "audiophiles" discussing vinyl vs. digital to get the morning
> >>>off to a good start......
> >>>
> >>>http://www.stevehoffman.tv/forums/showthread.php?t=3711...
> >>
> >>
> >> Absolutely hilarious! The real answer to the question - what digital
> >> resolution is closest to vinyl - is not 24/192, 24/96 or 16/44, it's
> >> actually about 12-13 bits on the best day of vinyl's life, with an
> >
> >hope this isn't a silly question, but how does vinyl have about 12-13 bits of
> >resolution?

> 13 bits is the equivalent of 78dB dynamic range, which is more than
> you'll find on even the best direct-cut virgin vinyl. Resolution is
> implicitly tied to dynamic range, system resolution being taken as the
> value of the noise floor and commonly expressed as xdB below peak
> level.

and sure enbouhg, in Christine Tham's comparison of LPs to various
digital versions, the LPs never even come *close* to 75 dB dynamic range
whereas the digital versions do so handily. She goes to great lengths,
however, to find ways to argue that LP still bests CD, concluding that LP
'consistenly have higher relative dynamics over digital formats'. After
reading the page, I'm *still* unclear as to what 'relative dynamics'
really means to her, other than a form a special pleading for her
preferred format.



> >> >On another topic, when we try to digitise an input signal using a fixed bit
> >resolution, wouldn't having more bits mean that the increments between levels
> >become smaller, i.e. more fine grain?




--

-S.
"We started to see evidence of the professional groupie in the early 80's.
Alarmingly, these girls bore a striking resemblance to Motley Crue." --
David Lee Roth
Anonymous
July 30, 2004 9:44:48 PM

Archived from groups: rec.audio.high-end (More info?)

On Fri, 30 Jul 2004 04:15:57 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>> There is however one vital difference here: the resolution of
>> fine-grain film still far exceeds that possible with even the best
>> digital cameras.
>
>Actually, that is not true. The current 4+ megapixel cameras equal to
>exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>range, it approaches or equals medium format film.

That's utter rubbish. Most of my film archive is shot on Kodachrome
25, and even a scan at 2400dpi does not capture all the detail. That's
the equivalent of an 8.64 megapixel camera. To capture the grain
structure of that film (or Ektar 25), I need to scan at 4800 dpi, the
equivalent of 35 megapixels!

>The only problem with the digital cameras is color reproduction and
>contrast. And that is right now equal to most common films.

I find no problem with contrast in my digital images, and the colour
reproduction seems to be as good as most film.

>>Basic 'Red Book' CD however far exceeds the
>> resolution of even the best analogue *master* tapes, never mind the
>> crippled child that is LP. Apples and oranges.
>
>Editorial comments aside - using "resolution" as your only metric is not
>measuring it in all ways if you are to use your digital camera example.

I'm not the one who was attempting to use resolution as a comparison,
I simply pointed out that it was inappropriate.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
July 30, 2004 9:49:04 PM

Archived from groups: rec.audio.high-end (More info?)

Tat Chan <le_king_num_7@hotmail.com> wrote:

>Stewart Pinkerton wrote:

>> Absolutely hilarious! The real answer to the question - what digital
>> resolution is closest to vinyl - is not 24/192, 24/96 or 16/44, it's
>> actually about 12-13 bits on the best day of vinyl's life, with an
>
>hope this isn't a silly question, but how does vinyl have about 12-13 bits of
>resolution?

Aside from doing some math you can easily do some tests on your own if
you have a quality soundcard in your PC. Record a LP with music of your
choice at the highest quality your soundcard is able to. You are free to
select the sample rate. Save the recording to a file. Next get some good
audio editor - some of them are freeware - and reduce the word length to
15 bits using dithering. Save the result to another file. Re-read the
original file, reduce the word length to 14 bits. Save again. You may
proceed to less bits per word. Then compare the results.

To help you find out the differences you may look at Arny Krueger' site
http://www.pcabx.com/

Norbert
Anonymous
July 31, 2004 7:31:51 AM

Archived from groups: rec.audio.high-end (More info?)

thanks for the informative reply. Comments embedded below.

Stewart Pinkerton wrote:

> On 28 Jul 2004 23:35:51 GMT, Tat Chan <le_king_num_7@hotmail.com>
> wrote:
>
>

>>hope this isn't a silly question, but how does vinyl have about 12-13 bits of
>>resolution?
>
>
> 13 bits is the equivalent of 78dB dynamic range, which is more than
> you'll find on even the best direct-cut virgin vinyl. Resolution is
> implicitly tied to dynamic range,
> system resolution being taken as the
> value of the noise floor and commonly expressed as xdB below peak
> level.

Ah, I wasn't considering dynamic range. It has been a while since I
looked at a signal processing text book, but I don't recall dynamic
range and dither being mentioned when ADCs/DACs were discussed (perhaps
they are application specific?). Ah, Flash ADCs and counter ADCs ...
those were the days!

>
> High sampling rates are another and more complex argument, having nothing
> to do with actual signal bandwidth, for which 25kHz will always be
> more than enough.
>

not wanting to open a can of worms here, but I thought sampling rates
that are at least twice the highest frequency in the input signal would
always be good enough to reproduce the signal in its entirety?

> You need to understand that dither makes this distinction untrue, as
> it randomises the absolute levels, in the same way that the noise
> floor randomises low-level analog signals. As a result, you can hear
> tones well below the LSB value in digital, just as you can in analog
> below the noise floor. I have recorded and replayed a 3kHz signal at
> -105dB on a CD-R, and it was perfectly audible below the smooth -93dB
> noise floor that is characteristic of properly made CD.

I need to read up on dither. I'm sure they didn't teach this in class!

>The 'stair steps' claim is an intuitive urban myth, with no existence
in reality.

well, you tend to get those diagrams in signal pro text books. Now,
where did I put mine ...
Anonymous
July 31, 2004 7:59:56 AM

Archived from groups: rec.audio.high-end (More info?)

On 7/30/04 1:44 PM, in article cee1eg018pf@news1.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

> On Fri, 30 Jul 2004 04:15:57 GMT, B&D <bromo@ix.netcom.com> wrote:
>
>> On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>> Pinkerton" <patent3@dircon.co.uk> wrote:
>>
>>> There is however one vital difference here: the resolution of
>>> fine-grain film still far exceeds that possible with even the best
>>> digital cameras.
>>
>> Actually, that is not true. The current 4+ megapixel cameras equal to
>> exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>> range, it approaches or equals medium format film.
>
> That's utter rubbish. Most of my film archive is shot on Kodachrome
> 25, and even a scan at 2400dpi does not capture all the detail. That's
> the equivalent of an 8.64 megapixel camera. To capture the grain
> structure of that film (or Ektar 25), I need to scan at 4800 dpi, the
> equivalent of 35 megapixels!

Here is a useful website - parity is achieved at 4-6 MP for most films.

Mind you that is PRINTS, not slide film.

http://www.clarkvision.com/imagedetail/film.vs.digital....

It is a pretty good site overall - but the native resolution, scanning and
so one are addressed.

For 35mm film, only some very slow films exceed the 4-8MP top tier
commercial cameras.

>
>> The only problem with the digital cameras is color reproduction and
>> contrast. And that is right now equal to most common films.
>
> I find no problem with contrast in my digital images, and the colour
> reproduction seems to be as good as most film.

Color reproduction is precisely where digital has the most trouble -
contrast and saturation. Resolution is not the issue. Most film play
tricks or heavy expansion/compression in order to "make up for" the loss of
color shading. Kind of like expansion and compression with a fixed number
of bits.

This is a surmountable problems as 99% of the issues can be "fixed" with a
good program like Photoshop.
Anonymous
July 31, 2004 6:53:25 PM

Archived from groups: rec.audio.high-end (More info?)

On 7/30/04 1:44 PM, in article cee1eg018pf@news1.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

>>> Basic 'Red Book' CD however far exceeds the
>>> resolution of even the best analogue *master* tapes, never mind the
>>> crippled child that is LP. Apples and oranges.
>>
>> Editorial comments aside - using "resolution" as your only metric is not
>> measuring it in all ways if you are to use your digital camera example.
>
> I'm not the one who was attempting to use resolution as a comparison,
> I simply pointed out that it was inappropriate.

Well the above interchange illustrates that the senses fool - the absolute
accuracy of digital is more with resolution than color accuracy and/or color
depth. Yet the senses might be much more forgiving of the color depth than
resoution.

It is possible that your film is being saved as a JPG - which will have
artifacting. Using a lossless storage such as a TIFF might get you better
results.
July 31, 2004 6:57:35 PM

Archived from groups: rec.audio.high-end (More info?)

Tat Chan wrote:
>> You need to understand that dither makes this distinction untrue, as
>> it randomises the absolute levels, in the same way that the noise
>> floor randomises low-level analog signals. As a result, you can hear
>> tones well below the LSB value in digital, just as you can in analog
>> below the noise floor. I have recorded and replayed a 3kHz signal at
>> -105dB on a CD-R, and it was perfectly audible below the smooth -93dB
>> noise floor that is characteristic of properly made CD.
>
> I need to read up on dither. I'm sure they didn't teach this in class!
>
>> The 'stair steps' claim is an intuitive urban myth, with no existence
> in reality.
>
> well, you tend to get those diagrams in signal pro text books. Now,
> where did I put mine ...

All these books show only the principle of the D/A stage, not the analog
output to the amplifier. There is at least a second order lowpass in
between, which completely eliminates these steps. In fact the sampling
theorem tells us the original input and output are *identical* if we use the
proper filters before the A/D and after the D/A converters. Part of the
input filtering can be done digitally if we oversample. We can also correct
certain erroneous filter characteristics (like gain peaking).
If you have the program Audition or any other wave editor, you can zoom in
and look at the actual waveforms and the sampling points, and there is no
step but a smooth curve that connects them. That curve is amazingly detailed
despite the few sampling points.
There are certain rules, of what order and characteristic those anti
aliasing filters have to be and when you observe this, the output waveform
is identical to the input. But this input waveform is not the original feed,
it is band limited.
So all comes down to the filters used for this process. With oversampling
the filters have become more simple and the phase accuracy for high
frequencies has increased. There will always be a measurable difference, but
our abilities to store the analog waveform are worse, we have only analog
tape machines, which BTW have at the end also a stepped waveform, because
the magnetisation is also quantized as single gross magnetic particles in
the tape are toggled. This is the reason that from a certain point say
48kHz/18bit the digital storage is actually more precise and
undistinguishable from a life feed.

--
ciao Ban
Bordighera, Italy
Anonymous
July 31, 2004 6:58:20 PM

Archived from groups: rec.audio.high-end (More info?)

On Sat, 31 Jul 2004 03:31:51 GMT, Tat Chan <le_king_num_7@hotmail.com>
wrote:

>Stewart Pinkerton wrote:

>> High sampling rates are another and more complex argument, having nothing
>> to do with actual signal bandwidth, for which 25kHz will always be
>> more than enough.
>>
>not wanting to open a can of worms here, but I thought sampling rates
>that are at least twice the highest frequency in the input signal would
>always be good enough to reproduce the signal in its entirety?

In a perfect world, that would be true, but there are issues with the
deleterious effects of real-world 'brick wall' filtering, which are
greatly eased by maintaining the same 20-25kHz bandwidth but
increasing the sampling rate.

>> You need to understand that dither makes this distinction untrue, as
>> it randomises the absolute levels, in the same way that the noise
>> floor randomises low-level analog signals. As a result, you can hear
>> tones well below the LSB value in digital, just as you can in analog
>> below the noise floor. I have recorded and replayed a 3kHz signal at
>> -105dB on a CD-R, and it was perfectly audible below the smooth -93dB
>> noise floor that is characteristic of properly made CD.
>
>I need to read up on dither. I'm sure they didn't teach this in class!

They should have! You'll find it (and lots more!) in Ken Pohlmann's
seminal text on the matter - Principles of Digital Audio

> >The 'stair steps' claim is an intuitive urban myth, with no existence
>in reality.
>
>well, you tend to get those diagrams in signal pro text books. Now,
>where did I put mine ...

Yep, that diagram goes right before the reconstruction filter, which
removes the HF components which cause the stair steps............

At the output of a properly designed DAC (I put this rider because
Peter Qvortrup simply didn't include a reconstruction filter in his
ludicrous Audio Note devices!), the output is a pure analogue signal
band-limited to 22kHz, with no sign of stairsteps.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
July 31, 2004 6:59:10 PM

Archived from groups: rec.audio.high-end (More info?)

On Sat, 31 Jul 2004 03:59:56 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 7/30/04 1:44 PM, in article cee1eg018pf@news1.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>> On Fri, 30 Jul 2004 04:15:57 GMT, B&D <bromo@ix.netcom.com> wrote:
>>
>>> On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>>> Pinkerton" <patent3@dircon.co.uk> wrote:
>>>
>>>> There is however one vital difference here: the resolution of
>>>> fine-grain film still far exceeds that possible with even the best
>>>> digital cameras.
>>>
>>> Actually, that is not true. The current 4+ megapixel cameras equal to
>>> exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>>> range, it approaches or equals medium format film.
>>
>> That's utter rubbish. Most of my film archive is shot on Kodachrome
>> 25, and even a scan at 2400dpi does not capture all the detail. That's
>> the equivalent of an 8.64 megapixel camera. To capture the grain
>> structure of that film (or Ektar 25), I need to scan at 4800 dpi, the
>> equivalent of 35 megapixels!
>
>Here is a useful website - parity is achieved at 4-6 MP for most films.

Not 'most' films, only 200 ASA and up - and note what happens to a
digital camera image when you up the effective speed.

>Mind you that is PRINTS, not slide film.
>
>http://www.clarkvision.com/imagedetail/film.vs.digital....
>
>It is a pretty good site overall - but the native resolution, scanning and
>so one are addressed.
>
>For 35mm film, only some very slow films exceed the 4-8MP top tier
>commercial cameras.

You seem to have very selective vision here - Kodachrome 64 is not a
slow film by many photographers standards, and is certainly not a
'very slow' film by any standard. More than 90% of my 7,000 or so film
archive are either Kodachrome 25 or Ektar 25 - *those* are slow films,
and still greatly exceed the capability of even the Canon 1DS.

>>> The only problem with the digital cameras is color reproduction and
>>> contrast. And that is right now equal to most common films.
>>
>> I find no problem with contrast in my digital images, and the colour
>> reproduction seems to be as good as most film.
>
>Color reproduction is precisely where digital has the most trouble -
>contrast and saturation. Resolution is not the issue.

Resolution remains the primary issue if you are a serious
photographer, and need to produce immaculate 20x16s (or A2) on a
regular basis.

> Most film play
>tricks or heavy expansion/compression in order to "make up for" the loss of
>color shading. Kind of like expansion and compression with a fixed number
>of bits.
>
>This is a surmountable problems as 99% of the issues can be "fixed" with a
>good program like Photoshop.

Quite so. Photoshop has made life a *lot* easier for technically
competent photographers - although it's still no substitute for
getting the exposure right in the first place!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
July 31, 2004 6:59:50 PM

Archived from groups: rec.audio.high-end (More info?)

On 7/30/04 11:31 PM, in article HoEOc.61650$eM2.28670@attbi_s51, "Tat Chan"
<le_king_num_7@hotmail.com> wrote:

>> You need to understand that dither makes this distinction untrue, as
>> it randomises the absolute levels, in the same way that the noise
>> floor randomises low-level analog signals. As a result, you can hear
>> tones well below the LSB value in digital, just as you can in analog
>> below the noise floor. I have recorded and replayed a 3kHz signal at
>> -105dB on a CD-R, and it was perfectly audible below the smooth -93dB
>> noise floor that is characteristic of properly made CD.
>
> I need to read up on dither. I'm sure they didn't teach this in class!

Periodic signals can be noticeable below noise floor - I am specifically
thinking about audibility of morse code below the noise floor in a
communications application. I don't know how far below the noise floor it
is audible, but I do know below 0dB SNR is audible. From this, I will
speculate that noise of this nature might be audible or nearly so in a quiet
passage in very good playback gear with a low inherent noise floor.

I always wondered how some jazz singers use the brush on the cybals when
they sing, but I am wondering if it is to "mask" background noise?
Anonymous
August 1, 2004 1:23:31 AM

Archived from groups: rec.audio.high-end (More info?)

On 7/31/04 10:59 AM, in article cegc3u02oen@news3.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

> On Sat, 31 Jul 2004 03:59:56 GMT, B&D <bromo@ix.netcom.com> wrote:
>
>> On 7/30/04 1:44 PM, in article cee1eg018pf@news1.newsguy.com, "Stewart
>> Pinkerton" <patent3@dircon.co.uk> wrote:
>>
>>> On Fri, 30 Jul 2004 04:15:57 GMT, B&D <bromo@ix.netcom.com> wrote:
>>>
>>>> On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>>>> Pinkerton" <patent3@dircon.co.uk> wrote:
>>>>
>>>>> There is however one vital difference here: the resolution of
>>>>> fine-grain film still far exceeds that possible with even the best
>>>>> digital cameras.
>>>>
>>>> Actually, that is not true. The current 4+ megapixel cameras equal to
>>>> exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>>>> range, it approaches or equals medium format film.
>>>
>>> That's utter rubbish. Most of my film archive is shot on Kodachrome
>>> 25, and even a scan at 2400dpi does not capture all the detail. That's
>>> the equivalent of an 8.64 megapixel camera. To capture the grain
>>> structure of that film (or Ektar 25), I need to scan at 4800 dpi, the
>>> equivalent of 35 megapixels!
>>
>> Here is a useful website - parity is achieved at 4-6 MP for most films.
>
> Not 'most' films, only 200 ASA and up - and note what happens to a
> digital camera image when you up the effective speed.

That is the majority of films sold today (100, 200, 400 ASA with mostly 200
and 400). However, we have agreement - on slower films the MP rating goes
up nearly exponentially - depending upon the type of film, of course.

8MP is available on top rated consumer (not even pro-sumer) gear now - price
point ~US$1k or so. That is about the same as the Kodachrome 64 in
resolution according to the charts. Keep in mind that the processing of
Kodachrome is rare enough that most have to resort to mailing the film to be
processed. It is a great film (being a resident of Rochester, I sure wish
film were so much better than digital, actually, the gap has been closed,
though for just about every common film!). I believe that most consumers
buy 100, 200, 400 ASA films, and most 100 ASA films are not nearly as
resolved as Kodachrome. Also you have to keep in mind that Kodachrome does
not have a neutral color balance either - it is reversal film as well so it
is harder to get prints from it.

Digital cameras, though, don't capture images nearly as fast as film - I
didn't mention this before because it was not the direct topic, but that is
one area where film is better still, and unquestionably so.

>> Mind you that is PRINTS, not slide film.
>>
>> http://www.clarkvision.com/imagedetail/film.vs.digital....
>>
>> It is a pretty good site overall - but the native resolution, scanning and
>> so one are addressed.
>>
>> For 35mm film, only some very slow films exceed the 4-8MP top tier
>> commercial cameras.
>
> You seem to have very selective vision here - Kodachrome 64 is not a
> slow film by many photographers standards,

Yes it is! ASA 100 is a slow film by every important satndard - ASA
standards for instance. Do you have a standard such as ISO that would make
100 ASA Kodachrome a "fast" film? (It *is* faster than Daguerrotype I will
admit! :-) )

> and is certainly not a
> 'very slow' film by any standard.

No, not VERY slow, but SLOW.

> More than 90% of my 7,000 or so film
> archive are either Kodachrome 25 or Ektar 25 - *those* are slow films,
> and still greatly exceed the capability of even the Canon 1DS.

You bet - there are a lot of films that are slower and have higher
resolution than commonly available digital cameras.

I figure, from your film selection, that you do a lot of outdoor
photography? Kodachrome and Ektachrome would require a lot of large lights
indoors, that is for sure!

>> Color reproduction is precisely where digital has the most trouble -
>> contrast and saturation. Resolution is not the issue.
>
> Resolution remains the primary issue if you are a serious
> photographer, and need to produce immaculate 20x16s (or A2) on a
> regular basis.

It sure it a primary issue, but the point I was making is that the weakness
of digital is more in color balance and depth than in resolution per-se.

I would hone your photoshop skills, though, as we have found that a lot of
the shops that developed our Medium format films have shuttered or only do
digital.

The price curve is coming down, and 25MP camera backs are not as expensive
as a small car anymore - you might consider one of those eventually as I
believe that traditional film's days are numbered, at least for
professionals.

>> This is a surmountable problems as 99% of the issues can be "fixed" with a
>> good program like Photoshop.
>
> Quite so. Photoshop has made life a *lot* easier for technically
> competent photographers - although it's still no substitute for
> getting the exposure right in the first place!

Agreed - nothing like good source materials! Everytime you "touch"
something in photoshop, you are losing some of the original information for
a better overall effect (one hopes!)

I think this brings us right back to audiophile stuff -
Anonymous
August 1, 2004 6:43:47 PM

Archived from groups: rec.audio.high-end (More info?)

On 31 Jul 2004 21:23:31 GMT, B&D <bromo@ix.netcom.com> wrote:

>8MP is available on top rated consumer (not even pro-sumer) gear now - price
>point ~US$1k or so. That is about the same as the Kodachrome 64 in
>resolution according to the charts. Keep in mind that the processing of
>Kodachrome is rare enough that most have to resort to mailing the film to be
>processed. It is a great film (being a resident of Rochester, I sure wish
>film were so much better than digital, actually, the gap has been closed,
>though for just about every common film!). I believe that most consumers
>buy 100, 200, 400 ASA films, and most 100 ASA films are not nearly as
>resolved as Kodachrome.

Since 'most consumers' don't give a rat's about exhibition quality
prints, the comparison is hardly relevant. I concede that a good
digital camera is better than a happy snapper using 200 ASA film.

> Also you have to keep in mind that Kodachrome does
>not have a neutral color balance either

I find it very good - but I never used K64, which does *not* have the
same balance as K25.

> - it is reversal film as well so it
>is harder to get prints from it.

Not with Photoshop!

>Digital cameras, though, don't capture images nearly as fast as film - I
>didn't mention this before because it was not the direct topic, but that is
>one area where film is better still, and unquestionably so.
>
>>> Mind you that is PRINTS, not slide film.
>>>
>>> http://www.clarkvision.com/imagedetail/film.vs.digital....
>>>
>>> It is a pretty good site overall - but the native resolution, scanning and
>>> so one are addressed.
>>>
>>> For 35mm film, only some very slow films exceed the 4-8MP top tier
>>> commercial cameras.
>>
>> You seem to have very selective vision here - Kodachrome 64 is not a
>> slow film by many photographers standards,
>
>Yes it is! ASA 100 is a slow film by every important satndard - ASA
>standards for instance. Do you have a standard such as ISO that would make
>100 ASA Kodachrome a "fast" film? (It *is* faster than Daguerrotype I will
>admit! :-) )

I have always considered 100 ASA to be the norm for a 'general
purpose' film (and so, it appears, does the ASA), hence 25 ASA is slow
and 400 ASA is fast. Do you know anyone who does *not* consider 400
ASA to be a fast film?

>> and is certainly not a
>> 'very slow' film by any standard.
>
>No, not VERY slow, but SLOW.
>
>> More than 90% of my 7,000 or so film
>> archive are either Kodachrome 25 or Ektar 25 - *those* are slow films,
>> and still greatly exceed the capability of even the Canon 1DS.
>
>You bet - there are a lot of films that are slower and have higher
>resolution than commonly available digital cameras.
>
>I figure, from your film selection, that you do a lot of outdoor
>photography? Kodachrome and Ektachrome would require a lot of large lights
>indoors, that is for sure!

They do, but they're still perfectly useable, even with tungsten
rather than flash lighting. Flash is better for female portraits of
course, due to maintaining pupil dilation - although there's always
good ol' Photoshop for that wide-eyed look! What I don't do a lot of
is hand-held photography - for sports work I do always shoot on
200-400 ASA film, and for that I concede that a Canon 1DS would give
equal or superior results.

>>> Color reproduction is precisely where digital has the most trouble -
>>> contrast and saturation. Resolution is not the issue.
>>
>> Resolution remains the primary issue if you are a serious
>> photographer, and need to produce immaculate 20x16s (or A2) on a
>> regular basis.
>
>It sure it a primary issue, but the point I was making is that the weakness
>of digital is more in color balance and depth than in resolution per-se.

Only because you're trying to skate away from your initial poor
analogy between digital audio and digital photography.

>I would hone your photoshop skills, though, as we have found that a lot of
>the shops that developed our Medium format films have shuttered or only do
>digital.

Sadly, this is true in the UK also, I believe there are only a couple
of top-class professional printers left that do 'wet work'.

>The price curve is coming down, and 25MP camera backs are not as expensive
>as a small car anymore - you might consider one of those eventually as I
>believe that traditional film's days are numbered, at least for
>professionals.

Nah, I'll keep waiting for Nikon to make a decent 16MP body - those
Nikkors are too expensive to chop in! OTOH, you have a point. For the
kind of photography I mostly do, something like a Hasselblad would be
just as convenient, if I had the cash! I won't argue that a 25MP
sensor would give me all the image quality I'll ever need.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 1, 2004 6:47:54 PM

Archived from groups: rec.audio.high-end (More info?)

On 31 Jul 2004 14:53:25 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 7/30/04 1:44 PM, in article cee1eg018pf@news1.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>>>> Basic 'Red Book' CD however far exceeds the
>>>> resolution of even the best analogue *master* tapes, never mind the
>>>> crippled child that is LP. Apples and oranges.
>>>
>>> Editorial comments aside - using "resolution" as your only metric is not
>>> measuring it in all ways if you are to use your digital camera example.
>>
>> I'm not the one who was attempting to use resolution as a comparison,
>> I simply pointed out that it was inappropriate.
>
>Well the above interchange illustrates that the senses fool - the absolute
>accuracy of digital is more with resolution than color accuracy and/or color
>depth. Yet the senses might be much more forgiving of the color depth than
>resoution.

I have no idea what the above is supposed to mean. Do you?
>
>It is possible that your film is being saved as a JPG - which will have
>artifacting. Using a lossless storage such as a TIFF might get you better
>results.

I get excellent results - at 4800 dpi from Kodachrome 25, which is
*way* above what any digital camera can currently provide. Naturally,
I don't save as jpegs.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 1, 2004 8:52:21 PM

Archived from groups: rec.audio.high-end (More info?)

On 8/1/04 10:43 AM, in article ceivj302t6b@news2.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

>> It sure it a primary issue, but the point I was making is that the weakness
>> of digital is more in color balance and depth than in resolution per-se.
>
> Only because you're trying to skate away from your initial poor
> analogy between digital audio and digital photography.

Not at all - actually I felt the same about you trying to widen it to a
discussion of pros and cons of digital and film.

Actually - you had said that resolution was the achilles heel of digital
photography and that the color balance and depth was better. It is the
other way around in reality - especially in color depth.

My initial analogy would be that since you got it wrong-way-round regarding
digital vs. film -- such can be true of audiophile related items. I didn't
skate out of anything, nor did you offer anything other than an insistance
that you were somehow correct, even if it meant changing your tack.
Anonymous
August 1, 2004 9:53:17 PM

Archived from groups: rec.audio.high-end (More info?)

On 8/1/04 10:43 AM, in article ceivj302t6b@news2.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

>> I would hone your photoshop skills, though, as we have found that a lot of
>> the shops that developed our Medium format films have shuttered or only do
>> digital.
>
> Sadly, this is true in the UK also, I believe there are only a couple
> of top-class professional printers left that do 'wet work'.

I fear that that is more of a trend - even wedding photographers are moving
away from the Mamiya medium format standard these days - especially since
making the bride look "perfect" is so much easier in digital mediums.

Here is where you and I will agree 100% about audiophile stuff - Digital
sound is much, much, MUCH easier to "clean up" than anything analog - even
if the resolution is lost a bit. And while the capabilities are a bit
different (there is no substitute for talented performances in a decent
acoustic space simply miked and recorded) it is clear that the sound
engineer's work is to piece together a number of takes to make the
performance as "perfect" as possible.

I find it funny that I am an old technology curmudgeon with sound, and while
I use digital am painfully aware of its weaknesses, and in photography we
are the other way around! Ha! :-)

[But if consumer sound ever got developed the way photogrpahy is right now -
instead of being "stuck" with Redbook with other high rez formats dying on
the vine due to IP issues I think - we would be of a like mind on audio, I
think...]

>
>> The price curve is coming down, and 25MP camera backs are not as expensive
>> as a small car anymore - you might consider one of those eventually as I
>> believe that traditional film's days are numbered, at least for
>> professionals.
>
> Nah, I'll keep waiting for Nikon to make a decent 16MP body - those
> Nikkors are too expensive to chop in! OTOH, you have a point. For the
> kind of photography I mostly do, something like a Hasselblad would be
> just as convenient, if I had the cash! I won't argue that a 25MP
> sensor would give me all the image quality I'll ever need.

I would consider it - but be really careful as the technology is about to go
through another round as the CCD arrays have now been developed that have
inherently better color depth and balance. Right now you have to have 3
pixels to catch each of the primary colors - and now they have figured out
the way to have 1 pixel capture all 3 colors.

I figure that about the time that 16MP is available (I think Kodak has a
14MP that doesn't break the bank right now - compatible with Nikon lenses,
too) - but stay tuned - it might still be OK to wait.

And we all will have large format to fall back on if we prefer film! :-P

[Also, how the heck do you deal with the color balance, reciprocity and
other issue using Kodachrome indoors with Tungsten lighting?]
Anonymous
August 1, 2004 9:54:01 PM

Archived from groups: rec.audio.high-end (More info?)

On 8/1/04 10:47 AM, in article ceivqq02tc2@news2.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

>>> I'm not the one who was attempting to use resolution as a comparison,
>>> I simply pointed out that it was inappropriate.
>>
>> Well the above interchange illustrates that the senses fool - the absolute
>> accuracy of digital is more with resolution than color accuracy and/or color
>> depth. Yet the senses might be much more forgiving of the color depth than
>> resoution.
>
> I have no idea what the above is supposed to mean. Do you?

Sorry - I meant to say that your eyes might think that the picture is fine
since our color vision is not as sharp as our black and white vision - and
therefore "fool" us into thinking that the color depth is OK.

>> It is possible that your film is being saved as a JPG - which will have
>> artifacting. Using a lossless storage such as a TIFF might get you better
>> results.
>
> I get excellent results - at 4800 dpi from Kodachrome 25, which is
> *way* above what any digital camera can currently provide. Naturally,
> I don't save as jpegs.

The film itself is the limiting factor - so you can scan it at just about
any DPI you want - you are effectively "over and up sampling" the print.
You should also make sure that the 4800 dpi is true resolution and not
interpolated, though if you are getting results you feel are fine, you
should be OK either way.

6-8MP is the current consumer SOTA.
8-16MP is the current professional SOTA for 35mm film backs
25-32MP is the pro bleeding edge to replace Medium format.
Anonymous
August 2, 2004 1:28:19 AM

Archived from groups: rec.audio.high-end (More info?)

Hi,

In message <1YjOc.178583$a24.9390@attbi_s03>, B&D <bromo@ix.netcom.com>
writes
>On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>> There is however one vital difference here: the resolution of
>> fine-grain film still far exceeds that possible with even the best
>> digital cameras.
>
>Actually, that is not true. The current 4+ megapixel cameras equal to
>exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>range, it approaches or equals medium format film.

I think Stewart is correct. It's common practice to scan 35mm film to
4000 lines with telecine, which puts it way higher than 6 megapixels.
The guys at Quantel reckon on upwards of 25 million samples per frame,
at a minimum of 36 bits of colour information, more for processing. The
dynamic range of film is pretty awesome too, especially near black.

I've been involved on the periphery of some digital cinema developments
lately that are using 4x1080p24 systems (i.e. four times the vertical
resolution of 1080 line HDTV, for over 4000 lines). We showed these guys
a playout display system running at 3840*2400, and they had to scale the
images down significantly to display them. You need ultra-wide SCSI 320
just to play this stuff back at full frame rate.

By those measures, even 10+ megapixel SLRs have a way to go.

--
Regards,
Glenn Booth
Anonymous
August 2, 2004 2:18:32 AM

Archived from groups: rec.audio.high-end (More info?)

Hi,

In message <cejamd02562@news1.newsguy.com>, B&D <bromo@ix.netcom.com>
writes

>I find it funny that I am an old technology curmudgeon with sound, and while
>I use digital am painfully aware of its weaknesses, and in photography we
>are the other way around! Ha! :-)

We seem to be at opposite ends of the spectrum. I'm firmly in the
digital camp with audio (though I have a half-decent vinyl setup), but
when it comes to my day job (video and graphics) I'll still take a good
ole' big Eizo CRT over any of the digital displays I have seen, even
counting the 9.2 Megapixel Viewsonic 2290. The gamma curves on the
current crop of high end DVI panels still don't match a good analogue
display, IME, but it's getting close.

There are downsides to the CRTs too though. The DVI displays don't need
recalibrating every time you breathe on them (just like a turntable now
I think about it, and for similar reasons). You also don't need a
massive desk for a 22 inch DVI panel, and you won't get a hernia moving
one around.

--
Regards,
Glenn Booth
Anonymous
August 2, 2004 2:20:35 AM

Archived from groups: rec.audio.high-end (More info?)

On 8/1/04 5:28 PM, in article TfdPc.229386$XM6.147515@attbi_s53, "Glenn
Booth" <glenn@qtlg.demon.co.uk> wrote:

> Hi,
>
> In message <1YjOc.178583$a24.9390@attbi_s03>, B&D <bromo@ix.netcom.com>
> writes
>> On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>> Pinkerton" <patent3@dircon.co.uk> wrote:
>>
>>> There is however one vital difference here: the resolution of
>>> fine-grain film still far exceeds that possible with even the best
>>> digital cameras.
>>
>> Actually, that is not true. The current 4+ megapixel cameras equal to
>> exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>> range, it approaches or equals medium format film.
>
> I think Stewart is correct. It's common practice to scan 35mm film to
> 4000 lines with telecine, which puts it way higher than 6 megapixels.
> The guys at Quantel reckon on upwards of 25 million samples per frame,
> at a minimum of 36 bits of colour information, more for processing. The
> dynamic range of film is pretty awesome too, especially near black.

Film has greater contrast and color depth than digital - no argument form me
since it is absolutely true and verifiable. 6-8MP has every bit as much
resolution as most 35mm films - though you would have to get closer to 10-12
to have all but the most exotic and slow films if resolution was your only
goal (and according the Stewart, this is his main goal and metric).

Scanning film - you have to achieve nyquist and all kinds of other
considerations. You certainly would want to scan at much higher the raw
film rate to make sure everything is there - but also a lot of scanners
interpolate so its native resolution may not be the resolution you claim.

> I've been involved on the periphery of some digital cinema developments
> lately that are using 4x1080p24 systems (i.e. four times the vertical
> resolution of 1080 line HDTV, for over 4000 lines). We showed these guys
> a playout display system running at 3840*2400, and they had to scale the
> images down significantly to display them. You need ultra-wide SCSI 320
> just to play this stuff back at full frame rate.
>
> By those measures, even 10+ megapixel SLRs have a way to go.

Except the measures are not correct. If resolution is your only metric,
35mm film has been matched in almost every film commonly used. The slower
ebd of the spectrum is not yet matched - but it is close.

Here are some websites that are worht looking at:

http://www.dlcphotography.net/Digital%20vs%20Film.htm
http://www.extremetech.com/article2/0,1558,4351,00.asp

Anmd it isn't completely black-and-white either. But if CD is a good
reproduction of sound - then digital is a good reproduction of vision.
Anonymous
August 2, 2004 4:10:32 AM

Archived from groups: rec.audio.high-end (More info?)

On 8/1/04 6:18 PM, in article cejq7o06da@news3.newsguy.com, "Glenn Booth"
<glenn@qtlg.demon.co.uk> wrote:

> Hi,
>
> In message <cejamd02562@news1.newsguy.com>, B&D <bromo@ix.netcom.com>
> writes
>
>> I find it funny that I am an old technology curmudgeon with sound, and while
>> I use digital am painfully aware of its weaknesses, and in photography we
>> are the other way around! Ha! :-)
>
> We seem to be at opposite ends of the spectrum. I'm firmly in the
> digital camp with audio (though I have a half-decent vinyl setup), but
> when it comes to my day job (video and graphics) I'll still take a good
> ole' big Eizo CRT over any of the digital displays I have seen, even
> counting the 9.2 Megapixel Viewsonic 2290. The gamma curves on the
> current crop of high end DVI panels still don't match a good analogue
> display, IME, but it's getting close.

For graphics work - the contrast of a CRT can't be beat! I don't own a
turntable, but have a friend or two with impressive setups - I am totally
unwilling to spend the money required to get good vinyl sound that will
exceed the quality of the CD player I currently own! Still, I am a bit of a
curmudgeon!

> There are downsides to the CRTs too though. The DVI displays don't need
> recalibrating every time you breathe on them (just like a turntable now
> I think about it, and for similar reasons). You also don't need a
> massive desk for a 22 inch DVI panel, and you won't get a hernia moving
> one around.

True - We recently got a new computer and needed a new screen as well -
ended up with a Viewsonic G90fb (CRT) and it is a really nice one and not
very expensive.

Right now the really good LCD's are many multiples of the current CRT
technology.
Anonymous
August 3, 2004 2:48:51 AM

Archived from groups: rec.audio.high-end (More info?)

On 1 Aug 2004 16:52:21 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 8/1/04 10:43 AM, in article ceivj302t6b@news2.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>>> It sure it a primary issue, but the point I was making is that the weakness
>>> of digital is more in color balance and depth than in resolution per-se.
>>
>> Only because you're trying to skate away from your initial poor
>> analogy between digital audio and digital photography.
>
>Not at all - actually I felt the same about you trying to widen it to a
>discussion of pros and cons of digital and film.
>
>Actually - you had said that resolution was the achilles heel of digital
>photography and that the color balance and depth was better.

I said no such thing - you were the one who brought up colour balance.
I did say that digital photography is currently inferior in resolution
to film, which of course it is, despite your attempts to cloud the
issue by only talking about fast film.

>My initial analogy would be that since you got it wrong-way-round regarding
>digital vs. film -- such can be true of audiophile related items.

Rubbish - film still has greatly superior resolutioon to digital
photography, whether 35mm or medium format. Even basic 16/55 digital
audio, OTOH, has superior resolution to analogue *master* tapes, never
mind vinyl.

> I didn't
>skate out of anything, nor did you offer anything other than an insistance
>that you were somehow correct, even if it meant changing your tack.

You are looking in the mirror, not at me..............
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 3, 2004 2:49:55 AM

Archived from groups: rec.audio.high-end (More info?)

On 1 Aug 2004 17:53:17 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 8/1/04 10:43 AM, in article ceivj302t6b@news2.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>>> I would hone your photoshop skills, though, as we have found that a lot of
>>> the shops that developed our Medium format films have shuttered or only do
>>> digital.
>>
>> Sadly, this is true in the UK also, I believe there are only a couple
>> of top-class professional printers left that do 'wet work'.
>
>I fear that that is more of a trend - even wedding photographers are moving
>away from the Mamiya medium format standard these days - especially since
>making the bride look "perfect" is so much easier in digital mediums.

Agreed.

>Here is where you and I will agree 100% about audiophile stuff - Digital
>sound is much, much, MUCH easier to "clean up" than anything analog - even
>if the resolution is lost a bit.

No, we don't agree at all. Why will you not understand that the
resolution of digtal audio - even basic 16/44 - *vastly* exceeds the
resolution of analogue?

> And while the capabilities are a bit
>different (there is no substitute for talented performances in a decent
>acoustic space simply miked and recorded)

Excuse me? What on earth has this to do with analogue/digital?

> it is clear that the sound
>engineer's work is to piece together a number of takes to make the
>performance as "perfect" as possible.

Well, that's the way it's usually done, but I'm often convinced that a
truly 'live' performance has more musical integrity.

>I would consider it - but be really careful as the technology is about to go
>through another round as the CCD arrays have now been developed that have
>inherently better color depth and balance. Right now you have to have 3
>pixels to catch each of the primary colors - and now they have figured out
>the way to have 1 pixel capture all 3 colors.

Foveon? That is far from fully developed.

>[Also, how the heck do you deal with the color balance, reciprocity and
>other issue using Kodachrome indoors with Tungsten lighting?]

I find no reciprocity 'issues', I just give an extra 1/3 stop indoors,
and a wide range of colour correction filters has been around for a
*long* time! Of course, you do need to be careful that you don't mix
your lighting, as coloured shadows are impossible to balance!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 3, 2004 2:51:01 AM

Archived from groups: rec.audio.high-end (More info?)

On 1 Aug 2004 17:54:01 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 8/1/04 10:47 AM, in article ceivqq02tc2@news2.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:

>> I get excellent results - at 4800 dpi from Kodachrome 25, which is
>> *way* above what any digital camera can currently provide. Naturally,
>> I don't save as jpegs.
>
>The film itself is the limiting factor

No, it isn't. You simply refuse to believe what's written even in the
article you cited yourself.

> - so you can scan it at just about
>any DPI you want - you are effectively "over and up sampling" the print.

Utter rubbish for fine grain film.

>You should also make sure that the 4800 dpi is true resolution and not
>interpolated, though if you are getting results you feel are fine, you
>should be OK either way.

I am talking about true optical 4800 dpi, as 2400 is *not* adequate
for scanning K25.

>6-8MP is the current consumer SOTA.

Agreed. although many would argue that it's really 5, since the
current 8MP sensors have some significant flaws.

>8-16MP is the current professional SOTA for 35mm film backs

Actually, it's 14, and that camera is a dog. The real SOTA is the 11MP
of the Canon 1DS.

>25-32MP is the pro bleeding edge to replace Medium format.

And has even more inferior resolution to medium-format film (remember,
6x6 has more than 4 times the area of 24x36) - although adequate for
most magazine photography, and certainly all wedding photography, so
since this is overwhelmingly a pro-photography sector, film cameras
almost certainly will disappear from this market before long.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 3, 2004 2:53:03 AM

Archived from groups: rec.audio.high-end (More info?)

On 1 Aug 2004 22:20:35 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 8/1/04 5:28 PM, in article TfdPc.229386$XM6.147515@attbi_s53, "Glenn
>Booth" <glenn@qtlg.demon.co.uk> wrote:

>> In message <1YjOc.178583$a24.9390@attbi_s03>, B&D <bromo@ix.netcom.com>
>> writes
>>> On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>>> Pinkerton" <patent3@dircon.co.uk> wrote:
>>>
>>>> There is however one vital difference here: the resolution of
>>>> fine-grain film still far exceeds that possible with even the best
>>>> digital cameras.
>>>
>>> Actually, that is not true. The current 4+ megapixel cameras equal to
>>> exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>>> range, it approaches or equals medium format film.
>>
>> I think Stewart is correct. It's common practice to scan 35mm film to
>> 4000 lines with telecine, which puts it way higher than 6 megapixels.
>> The guys at Quantel reckon on upwards of 25 million samples per frame,
>> at a minimum of 36 bits of colour information, more for processing. The
>> dynamic range of film is pretty awesome too, especially near black.
>
>Film has greater contrast and color depth than digital - no argument form me
>since it is absolutely true and verifiable. 6-8MP has every bit as much
>resolution as most 35mm films - though you would have to get closer to 10-12
>to have all but the most exotic and slow films if resolution was your only
>goal (and according the Stewart, this is his main goal and metric).

Please stop putting words in my mouth. *You* brought up resolution as
a comparison between audio and photography, I simply pointed out that
it was a poor analogy, since the best digital photography has poorer
resolutuin than the best film, whereas even basic 16/44 audio exceeds
the resolution of the very best analogue.

>Scanning film - you have to achieve nyquist and all kinds of other
>considerations. You certainly would want to scan at much higher the raw
>film rate to make sure everything is there - but also a lot of scanners
>interpolate so its native resolution may not be the resolution you claim.

It's always a bad idea to use terms you don't understand. Scanning a
film is not the same as sampling audio, and you don't need to scan at
more than twice the film resolution to capture all the detail. Also,
don't try to duck the issue by talking about interpolation, when there
are several perfectly good scanners with true optical 4800 dpi
available. When I say you need 4800 dpi to fully scan Kodachrome 25, I
*mean* 4800 dpi, not processed 2400 dpi.

>> I've been involved on the periphery of some digital cinema developments
>> lately that are using 4x1080p24 systems (i.e. four times the vertical
>> resolution of 1080 line HDTV, for over 4000 lines). We showed these guys
>> a playout display system running at 3840*2400, and they had to scale the
>> images down significantly to display them. You need ultra-wide SCSI 320
>> just to play this stuff back at full frame rate.
>>
>> By those measures, even 10+ megapixel SLRs have a way to go.
>
>Except the measures are not correct.

Of course they are - why would they not be?

> If resolution is your only metric,
>35mm film has been matched in almost every film commonly used.

You are trying to move the goalposts by talking about 'happy snapper'
film, *not* what professionals use when they want high-quality
results.

Basically, you're saying that analogue audio is better because most
people use MP3 at 128kb/sec, and this isn't as good as 1/2 inch 30 ips
analogue tape.

> The slower
>ebd of the spectrum is not yet matched - but it is close.

No, digital photography has to more than double before it gets close
to Kodachrome 25.

>Here are some websites that are worht looking at:
>
>http://www.dlcphotography.net/Digital%20vs%20Film.htm
>http://www.extremetech.com/article2/0,1558,4351,00.asp

Did you miss this paragraph in your second site cite (so to speak)?:

"To set the scale of detail, a digital camera's resolution is measured
according to the total number of lines it can resolve before they
begin to run together. A typical 3 megapixel digital camera has a
resolving power of about 1,000 lines over the entire image sensor. So,
if the CCD is 1/2" in size, that amounts to a total resolution of
2,000 lines per inch. By contrast, the resolution of fine grain 35mm
Kodachrome film is about 2,200 lines--per millimeter! That's more than
50 times better raw resolution than digital. Using this for
comparison, film scientists sometimes peg Kodachrome's digital
equivalent as a 100 megabyte file. Of course, larger film--2 1/4x
31/4, 4x5", etc.--will yield correspondingly more data and overall
resolution than a 35mm frame. "

So much for your claims about digital image resolution. BTW, note that
the SOTA Canon 1DS has a 24x36 11MP sensor, which is only a fraction
over 1,000 lines per inch, so still fifty times less than Kodachrome!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 3, 2004 7:29:19 AM

Archived from groups: rec.audio.high-end (More info?)

In article <1YjOc.178583$a24.9390@attbi_s03>, B&D <bromo@ix.netcom.com>
wrote:

> On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
> Pinkerton" <patent3@dircon.co.uk> wrote:
>
> > There is however one vital difference here: the resolution of
> > fine-grain film still far exceeds that possible with even the best
> > digital cameras.
>
> Actually, that is not true. The current 4+ megapixel cameras equal to
> exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
> range, it approaches or equals medium format film.
>
> The only problem with the digital cameras is color reproduction and
> contrast. And that is right now equal to most common films.

Um, no. A properly exposed 35mm slide has more "pixels" than the best
digital cameras by a big margin, and the digital cams aren't even in the
same ballpark w.r.t. dynamic range.

A competent photographer can make photographic images with considerably
over 1000:1 density on the film; try that digitally.

Isaac
Anonymous
August 4, 2004 2:56:04 AM

Archived from groups: rec.audio.high-end (More info?)

On Tue, 03 Aug 2004 03:29:19 GMT, Isaac Wingfield <isw@witzend.com>
wrote:

>In article <1YjOc.178583$a24.9390@attbi_s03>, B&D <bromo@ix.netcom.com>
>wrote:
>
>> On 7/29/04 8:02 PM, in article cec36g0faq@news2.newsguy.com, "Stewart
>> Pinkerton" <patent3@dircon.co.uk> wrote:
>>
>> > There is however one vital difference here: the resolution of
>> > fine-grain film still far exceeds that possible with even the best
>> > digital cameras.
>>
>> Actually, that is not true. The current 4+ megapixel cameras equal to
>> exceed high quality (low speed) 35mm film. In fact, at the 6+ megapixel
>> range, it approaches or equals medium format film.
>>
>> The only problem with the digital cameras is color reproduction and
>> contrast. And that is right now equal to most common films.
>
>Um, no. A properly exposed 35mm slide has more "pixels" than the best
>digital cameras by a big margin, and the digital cams aren't even in the
>same ballpark w.r.t. dynamic range.
>
>A competent photographer can make photographic images with considerably
>over 1000:1 density on the film; try that digitally.

You need 10 bits for 1000:1, and the vast majority of digital cameras
only use 8 bits. QED.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 4, 2004 3:04:00 AM

Archived from groups: rec.audio.high-end (More info?)

Isaac Wingfield wrote:

>
>
>Um, no. A properly exposed 35mm slide has more "pixels" than the best
>digital cameras by a big margin, and the digital cams aren't even in the
>same ballpark w.r.t. dynamic range.
>
>A competent photographer can make photographic images with considerably
>over 1000:1 density on the film; try that digitally.
>
>
>
==============================================

Reply, off topic of our normal high-end audio discussion subjecst:

This is absolutely true. The common digital camera user doesn't know
this, being happy with small prints rendered with coarse ink blobs on
paper, or low resolution images on a monitor; and their tendency to say
that any photo that shows their grandkids with a nice expression, no
matter how blurry, is a good one.

I will leave it to the rest of you to make an analogy to audio with this!

-Gene Poon
Anonymous
August 4, 2004 3:04:32 AM

Archived from groups: rec.audio.high-end (More info?)

On 8/2/04 6:49 PM, in article cemgej0tn1@news2.newsguy.com, "Stewart
Pinkerton" <patent3@dircon.co.uk> wrote:

>> Here is where you and I will agree 100% about audiophile stuff - Digital
>> sound is much, much, MUCH easier to "clean up" than anything analog - even
>> if the resolution is lost a bit.
>
> No, we don't agree at all. Why will you not understand that the
> resolution of digtal audio - even basic 16/44 - *vastly* exceeds the
> resolution of analogue?

Here is where you are not listening to me. I am agreeing that it is easier
to clean up digital even if the digital signal resolution is lost a little
bit.

Note, I made no reference to resolution or performance of vinyl, nor did I
intend. I was trying to point out that we agreed on something.

If you do not want to agree with me about the fact that it is easier to
manipulate and clean up a digital signal than it is analog - fine by me,
though I do really think we agree about *that*
Anonymous
August 4, 2004 7:24:29 AM

Archived from groups: rec.audio.high-end (More info?)

"Gene Poon" <sheehans@ap.net> wrote in message
news:cep5l002tbf@news4.newsguy.com...
> ==============================================
>
> Reply, off topic of our normal high-end audio discussion subjecst:
>
> This is absolutely true. The common digital camera user doesn't know
> this, being happy with small prints rendered with coarse ink blobs on
> paper, or low resolution images on a monitor; and their tendency to say
> that any photo that shows their grandkids with a nice expression, no
> matter how blurry, is a good one.
>
> I will leave it to the rest of you to make an analogy to audio with this!

Well if they do any b/w darkroom hobbying, perhaps we can learn what audio
equipment they listen to while in their darkrooms. IIRC the safelight I used
with Polycontrast paper was "OC" and the filter passed greenish light so we
are all set to listen to CDs. While in Costco a recent demo showed gorgeous
sharp color saturated ~ 5 X 9 prints on photo paper taken with a 3
megapixel camera, landscapes, etc. (no kids blowing out birthday candles).
Owning a couple of Leicas, other 35 mm stuff as well as 2 medium format
cameras, I'd say the shots were as good as any I'd ever seen (and btw my
eyes are much better than my ears).
Anonymous
August 4, 2004 7:26:01 AM

Archived from groups: rec.audio.high-end (More info?)

Stewart Pinkerton wrote:

> On Sat, 31 Jul 2004 03:31:51 GMT, Tat Chan <le_king_num_7@hotmail.com>
> wrote:
>
>

>>
>>I need to read up on dither. I'm sure they didn't teach this in class!
>
>
> They should have! You'll find it (and lots more!) in Ken Pohlmann's
> seminal text on the matter - Principles of Digital Audio
>

well, they taught too much theory in class and hardly any practical
applications. I think they wanted to push us into maanagement and not a
"lab rat" role ... uhm, no offence meant with that term to all those
reading.

>
>>>The 'stair steps' claim is an intuitive urban myth, with no existence
>>
>>in reality.
>>
>>well, you tend to get those diagrams in signal pro text books. Now,
>>where did I put mine ...
>
>
> Yep, that diagram goes right before the reconstruction filter, which
> removes the HF components which cause the stair steps............
>

I vaguely remember seeing that somewhere, not in the text book I have on
hand (which is fairly rubbish).

So the reconstruction filter is a low pass, anti-aliasing filter right?
I remember that during the signal reconstruction, images of all
frequency components below the sampling frequency will appear at
multiples of the sampling frequency, and hence the need for the
anti-aliasing filter.


> At the output of a properly designed DAC (I put this rider because
> Peter Qvortrup simply didn't include a reconstruction filter in his
> ludicrous Audio Note devices!), the output is a pure analogue signal
> band-limited to 22kHz, with no sign of stairsteps.

if the reconstruction filter is the anti-aliasing filter I mentioned
above, it would be interesting to listen to such a DAC, since the
reproduced signal is not accurate. Hmmmm ... now I'm curious as to how
it would actually sound.
Anonymous
August 5, 2004 3:55:30 AM

Archived from groups: rec.audio.high-end (More info?)

On 4 Aug 2004 03:26:01 GMT, Tat Chan <le_king_num_7@hotmail.com>
wrote:


>So the reconstruction filter is a low pass, anti-aliasing filter right?
>I remember that during the signal reconstruction, images of all
>frequency components below the sampling frequency will appear at
>multiples of the sampling frequency, and hence the need for the
>anti-aliasing filter.

Yes, although strictly speaking it's an anti-imaging filter, the term
aliasing is properly used for the ADC function.

>> At the output of a properly designed DAC (I put this rider because
>> Peter Qvortrup simply didn't include a reconstruction filter in his
>> ludicrous Audio Note devices!), the output is a pure analogue signal
>> band-limited to 22kHz, with no sign of stairsteps.
>
>if the reconstruction filter is the anti-aliasing filter I mentioned
>above, it would be interesting to listen to such a DAC, since the
>reproduced signal is not accurate. Hmmmm ... now I'm curious as to how
>it would actually sound.

Airy but vague is perhaps the nearest I can come to describing the
sound. I can understand why a valve/vinyl guy might like it, but it's
basically broken!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
August 5, 2004 3:56:28 AM

Archived from groups: rec.audio.high-end (More info?)

On 3 Aug 2004 23:04:32 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 8/2/04 6:49 PM, in article cemgej0tn1@news2.newsguy.com, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>>> Here is where you and I will agree 100% about audiophile stuff - Digital
>>> sound is much, much, MUCH easier to "clean up" than anything analog - even
>>> if the resolution is lost a bit.
>>
>> No, we don't agree at all. Why will you not understand that the
>> resolution of digtal audio - even basic 16/44 - *vastly* exceeds the
>> resolution of analogue?
>
>Here is where you are not listening to me. I am agreeing that it is easier
>to clean up digital even if the digital signal resolution is lost a little
>bit.

But with digital, no resolution is *ever* lost, unless you dither down
to a lower bit depth. I'm listening, but you're not making sense. And
what do you mean by 'clean up', in any case?
--

Stewart Pinkerton | Music is Art - Audio is Engineering
!