On Wed, 10 Nov 2004 19:12:05 -0000, "Keith G"
wrote:

A while back I made a reference to a woolly memory of something I had read
in HFW to do with Tim de P's views on bitrates and their vinyl equivalents
and said I would post a reference to it, if it ever appeared. Well it's
popped up out of the blue and is, of course, nothing like I remembered it.

It's on 2 pages of the April 2004 edition:

http://www.apah69.dsl.pipex.com/keit.../article01.jpg

plus the top left paragraph he

http://www.apah69.dsl.pipex.com/keit.../article02.jpg


The 'bitrates' are nothing to do with vinyl it seems - simply Tim De P's
idea of a minimum requirements for digital to come even close.

And we're all well aware that 'crazy Tim'is not fully resident on
planet Earth. Typical valvie, on current evidence............

Now, having said all this, I still have another memory that there are some
pretty impressive figures somewhere that compare vinyl 'information flow'
very favourably with digital bitrates, but I've no idea where from and have
no intention of trying to find out.

Commercial vinyl has a dynamic range of 75dB on the best day of its
life, and a bandwidth of less than 20kHz, regardless of what the
cartridge *could* respond to. This may be fully captured by a 13-bit
digital sampling system running at 40k samples/sec, so in fact the
'information density' is significantly less than that of CD.

- I don't need to, I *know* there's more
detail in vinyl played on decent equipment.

Sure you do, dearie, and you've been told just how this *trick* is
done....................
--

Stewart Pinkerton | Music is Art - Audio is Engineering

In article , Ian Molton wrote:
New Geoff wrote:
"Ian Molton" almost choked on his de-caffinated espresso...

OMFG. 400kHz sampling?
I dont think even a bat could hear the top end of the frequecy range
that allows.

But the point isn't the maximum frequency, it's the content of the audible
waveform . . . .

Remember the idea . . . increased frequency of sampling allows you to
reconstruct a waveform closer to the original analogue form . . . .??

Up to a point, yes. but as someone else here pointed out - over ~8kHz
humans cant distinguish the difference between sine, triangle, sawtooth,
square at all. thats well below 22kHz.

Stepping back, the work of Fletcher & Munson and many others seems to
remain at the core of human hearing research. Indeed about 20 kHz
clearly remains the accepted upper limit for what we can hear.

There have been a few papers on human perception of ultrasound but
compared to the bulk of the literature it is clear that these are
exploring the margins rather than the fundamentals.

A quick search reveals a couple of examples:

- http://home.dmv.com/~tbastian/files/ultrsonc.txt

- http://www.hearultraquiet.com/Pages/...%20Hearing.pdf

It certainly seems that there may be some perception of ultrasound but
just how much that changes how we experience music and other audio is
still not clear. Those who say 44.1 kHz sampling is not enough may
possibly end up with an objective case. Nevertheless, just how much
difference it makes in reality to our experience is a long way from
being established.

We may want to go that way, but it is clear that compared to 44.1 kHz
the effect of upping the sample rate will be distinctly marginal rather
than fundamental.

--
John Phillips

In article , New Geoff m.gjb SPHERICAL
wrote:

"Ian Molton" almost choked on his de-caffinated espresso...

OMFG. 400kHz sampling?

I dont think even a bat could hear the top end of the frequecy range
that allows.

But the point isn't the maximum frequency, it's the content of the
audible waveform . . . .

Remember the idea . . . increased frequency of sampling allows you to
reconstruct a waveform closer to the original analogue form . . . .??

Only if the 'original waveform' contained components at frequencies above
half the original sampling rate chosen.

And that would only matter if we are able to percieve the effect of the
presence/absence of the high frequency components that the lower sampling
rate was not adequate to capture.

Sampling Theorem.

And let's face it, a lot of 'musicality' is based on harmonics and the
interplay of notes, so you really do want a pretty big sampling rate to
make sure you catch it all . . .

Only for the harmonics, etc, at frequencies that have an audible effect.

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

In article , Tat Chan
wrote:


so Tim de P reckons that digital should be 24/400, eh? Most people can't
hear above 20kHz, and 400kHz would allow signals with frequency content
of up to 200kHz be reproduced perfectly. A bit of a waste, isn't it?
Though 200kHz would better capture the harmonics of a square/triangle
wave.

24-bit resolution would imply a dynamic range of 144dB. That's pretty
loud!

I wonder what analog audio system TdeP uses that has a performance that
delivers audio signals to his ears over the bandwidth up to 200kHz with
144dB dynamic range...

Hope he doesn't use it to play SACD's as well... :-)

Still, it must be useful when he is working on valve power amps to be able
to hear it when the amp oscillates at 100 kHz. Perhaps that's why some
magazines like his designs so much... ;-

For some reason, 'bats' come to mind here. 8-]

Although as John has pointed out, there is some work that shows that
'ultrasound' may have audible effects at times.

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

In article , The EggKing
wrote:
So if we should be buying Super-Tweeters if we want to properly listen
to our SACD systems (I don't have one) then what do we need to get the
air moving at 200KHz?

Above about 50kHz the main output from SACD will be the 'hash' from the 1
bit noise shaping scheme. I don't think Philips really want us to hear
that...

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

In article , Nick Gorham
wrote:
Tat Chan wrote:



Vinyl only has (at best) 70 - 78 dB of dynamic range, which equates to
12 - 13 bits resolution, and I am sure vinyl is bandwidth limited as
well (cuts off at 16kHz?).


I could show you a 20kHz sine from a test disk, if that helps.

But at what level, and with how much distortion? And at what point on the
LP?

Isn't the extra "frequency content" associated with vinyl a byproduct
of the mechanical replay system?


No argument that 2nd harmonics will poduce extra extension, but then
unlike CD, it CAN produce harmonics above 20k.

Indeed. But how much of them are due to distortion?

i.e. how well can you record and replay signals at, say, 40kHz via vinyl
LP? (As distinct from finding distortion products.)

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

"Nick Gorham" wrote in message


Ian Molton wrote:

Up to a point, yes. but as someone else here pointed out - over ~8kHz
humans cant distinguish the difference between sine, triangle,
sawtooth, square at all. thats well below 22kHz.

I think the point made was over 8k sine and square was
indistinguisable.

...to review, that's because the first harmonic that is present in the square
wave is at 24 Khz.

The triangle, being symmertrical also has its first present harmonic at 24
KHz.

The sawtooth lacks half-wave symmetry and therefore has substantial content
at 16 KHz.

I would expect someone who's hearing went beyond 16k to tell the rest
apart.

That is not obvious, because masking can prevent people from perceiving the
lack of signal at frequencies that are lower than the limit of hearing for
pure high frequency sine waves. However, the second harmonic of a 8 KHz
sawtooth is probably strong enough to be noticable.

Actually doing this experiment might be non-trivial because it can be hard
to get really good sawtooth waves to work with.

"Tat Chan" wrote in message


Keith G wrote:

Actually it gets better on the second page:

http://www.apah69.dsl.pipex.com/keit.../article03.jpg

... a 'Digital Mastering Consultant' who says "I'd rather listen to
analogue masters than digital" and "analogue still has the
edge".....!!!

(Ya hafta larf....!!! :-)

Agreed.

but that would be his personal preference.

It reflects badly on his work.

He also says "Analogue is superior, theoretically. A digital system
will have analogue front and back ends on the ADC and DAC, and as the
digital section cannot be completely transparent, a purely analogue
system must be better."

He's wrong.

Que?

Can someone please explain

- "analogue is superior, theoretically"

In a sense this is largely correct because acostic instruments effectively
work in the analog domain. There are technical losses in conversion to
digital, even when they are inaudible. However, simply amplifiying and
processing analog also involves technical losses. Analog really falls apart
when you try to distribute it widely.

- "the digital section cannot be completely transparent"

He's wrong. Yet another guy who embarasses himself in public because he is
poorly-informed.

"The EggKing" wrote in message

So if we should be buying Super-Tweeters if we want to properly
listen to our SACD systems (I don't have one) then what do we need to
get the air moving at 200KHz?

super super tweeters.

"Kurt Hamster" wrote in message

On Thu, 11 Nov 2004 07:30:45 +0000 (UTC), Stewart Pinkerton used
to say...

And we're all well aware that 'crazy Tim'is not fully resident on
planet Earth. Typical valvie, on current evidence............

He still has far more credibility in the biz than you have Pinky.

Prove it.