On Wed, 11 Mar 2015 01:19:07 +0000 (GMT), "Dave Plowman (News)"
wrote:

In article ,
Java Jive wrote:
As I said at the top of this sub-thread, but care can not make up for
the low FR of AC, a constraint arising out of its slow tape-speed.

Commercial cassettes were usually duplicated at high speed. So never going
to be at the top end of even that lowly format.

Cassettes made on a good home deck using top quality tape could be
remarkably good, considering.

Not too difficult to achieve considerably higher quality than
commercially recorded music cassettes given a reasonable quality
cassette deck and chrome/SA tape formulations (let alone metal tapes
with Dolby noise reduction with the better decks) when you consider
that the duplication process was often run at 8 times speed and even
as high as 16 speed in some cases.

I'm not sure what the maximum duplication speed was, possibly as high
as 32 speed for audio book recordings were the lower demands for good
quality speech allowed the duplicating equipment to be pushed to its
limits[1] without any obvious degradation becoming evident in the
playback.

As the saying goes, "Time is Money" and there was every temptation to
run the duplication plant as fast as possible, often a choice of 8
speed over quad speed on older kit and probably 16 speed in place of
the 8 speed option on the newer and improved duplicators (where the
16, and possibly 32, speed was intended for audiobook quality alone).

[1] The real limits were down to heating effects and saturation of the
magnetic cores and pole pieces of the special 4 track recording tape
heads used by the slave drives to transfer the master in a single
pass.

Notably the nominal 60KHz bias current being scaled up to 960KHz for
a 16 speed transfer rate. In this case, it was the limits of the heads
that defined the maximum peak limits of the recordings rather than the
limits of the tape formulation itself which would normally be the
limiting factor on a cheap deck's performance at real time speeds.
--
J B Good

In article , Sumatriptan
wrote:
On 10/03/2015 14:38, Don Pearce wrote:

I think a likely explanation is the simple non-linearity of the ear
causing audible intermod products when the ultrasonics were on.


There's a simple test that could confirm that idea.

Two (or more) high level ultrasonic tones a few hundred Hz apart should
result in audible tones. Surely this has been tried.

Yes, it has. See what others have now said, for example.

Also IIRC it is touted by at least one company as a way to 'throw' audio so
that it can be heard in one area but not others. The trick is to send one
set of ultrasonic tones from one speaker and another from a different
speaker. You then are said to only hear the intended intermod in the area
where the beams cross.

No idea how well it works. Only recall hearing reports about it. Must admit
my concern was the exposure to high levels of ultrasound and I wonder if it
would eventually damage hearing or cause some other problem.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

In article , Trevor Wilson trevor@SPA
MBLOCKrageaudio.com.au scribeth thus
On 10/03/2015 2:29 AM, Java Jive wrote:
On Mon, 09 Mar 2015 12:58:54 +1100, Trevor Wilson
wrote:

On 9/03/2015 2:15 AM, Java Jive wrote:
As I said earlier, these days manufacturers take even less care with
vinyl than they do with other media sources.

CD technology could and should have given us a flat FR from near DC up
to 25kHz, easily covering the range of human hearing.

**Bull****. With a sampling rate of 44.1kHz, Nyquist tells us that the
theoretical maximum of CDs is 22.05kz.

As it was actually implemented it was something of a compromise,
sacrificing FR to give greater playing time, but the space available
on the prototype technology "COULD AND SHOULD" (note what I actually
wrote) have been allocated differently to give us a shorter playback
time at a higher sampling rate that would have covered the range of
human hearing.

Even going up
to 22kHz it still covers quite comfortably the range of older
listeners such as myself.

**It actually covers pretty much everyone over the age of 10.

I and several others tested the range of our hearing in the Physics
Lab at college when I was about 17 or 18, and I wasn't the only one in
the group who could hear above 23KHz at that age.

**AT WHAT LEVEL?

The human ear does not have a 'brickwall filter' at 20kHz. No one ever
suggested it did. I recall visiting a warehouse owned by the company I
worked for when I was around 25 years old. I was assaulted by the most
appalling 'feeling' and I had to immediately leave. Curious, I decided
to work out what was going on. Turns out I was reacting to the
ultrasonic burglar alarm system. A microphone, preamp, oscilloscope and
frequency counter showed me that the space in the warehouse was
constantly flooded with a high intensity acoustic signal of
approximately 26kHz. OTOH, using conventional hearing tests, my hearing
extended to around 19kHz at the time.

MORAL: With sufficient intensity, human hearing can extend way past
20kHz. That intensity is NEVER achieved with any commercial recordings.


Didn't Eckersley of the BBC prove the effects of filters operating above
the usual 15 kHz limit did cause audible differences?..






--
Tony Sayer

In article , Don Pearce
wrote:
Actually, 30 and 33kHz are going to generate products at 27 and 36kHZ -
not much help.

Why are you assuming their'd be no f1 - f2 result?

I made a file with 20kHz and 38kHz. The 2kHz product was clearly
audible, and very level dependent. There was actually some kind of
threshold effect. I prevented distortion in the amplifier being a
confounding factor by recording them on separate channels.

2*f1 - f2 ?

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

On 11/03/2015 09:12, Jim Lesurf wrote:


Yes, it has. See what others have now said, for example.

Also IIRC it is touted by at least one company as a way to 'throw' audio so
that it can be heard in one area but not others. The trick is to send one
set of ultrasonic tones from one speaker and another from a different
speaker. You then are said to only hear the intended intermod in the area
where the beams cross.

No idea how well it works. Only recall hearing reports about it. Must admit
my concern was the exposure to high levels of ultrasound and I wonder if it
would eventually damage hearing or cause some other problem.



Interesting result from Don in this thread. I didn't try it myself
because of worries about screaming high ultrasonics affecting my
hearing. So, the verdict seems to be that they can't be heard in
isolation but may impact perception of sound within the audible range.

Just wondering, does material outside the usual hearing range contribute
or detract from perceived 'realism' of recordings? Quite relevant
because we have a whole niche industry selling products offering
performance at at 20 kHz and above that may/may not be relevant to the
sound we can hear.

Of course, subsonics are an example of sounds that may not be heard but
are widely known to be perceived.

I'm interested in Don's report of sudden onset of IM products. I wonder
where in the ear they are being generated...drum, stapes, cochlea,
brain? Plenty of scope for non linearity in that soft squidgy stuff we
are made from.

On Wed, 11 Mar 2015 10:42:43 +0000 (GMT), Jim Lesurf
wrote:

In article , Don Pearce
wrote:
Actually, 30 and 33kHz are going to generate products at 27 and 36kHZ -
not much help.

Why are you assuming their'd be no f1 - f2 result?

I made a file with 20kHz and 38kHz. The 2kHz product was clearly
audible, and very level dependent. There was actually some kind of
threshold effect. I prevented distortion in the amplifier being a
confounding factor by recording them on separate channels.

2*f1 - f2 ?

Jim

I was assuming a third order product as we're dealing with intermods
(2*f1 - f2, as you say). There may be some second order effect, but it
isn't guaranteed.

d

In article , Sumatriptan
wrote:

Just wondering, does material outside the usual hearing range contribute
or detract from perceived 'realism' of recordings?

Its certainly possible. Hard to say more as things stand.

I'm interested in Don's report of sudden onset of IM products. I wonder
where in the ear they are being generated...drum, stapes, cochlea,
brain? Plenty of scope for non linearity in that soft squidgy stuff we
are made from.

This may help

http://www.audiomisc.co.uk/HFN/hearing/index.html

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

On Wed, 11 Mar 2015 10:57:34 +0000, Sumatriptan
wrote:

On 11/03/2015 09:12, Jim Lesurf wrote:


Yes, it has. See what others have now said, for example.

Also IIRC it is touted by at least one company as a way to 'throw' audio so
that it can be heard in one area but not others. The trick is to send one
set of ultrasonic tones from one speaker and another from a different
speaker. You then are said to only hear the intended intermod in the area
where the beams cross.

No idea how well it works. Only recall hearing reports about it. Must admit
my concern was the exposure to high levels of ultrasound and I wonder if it
would eventually damage hearing or cause some other problem.



Interesting result from Don in this thread. I didn't try it myself
because of worries about screaming high ultrasonics affecting my
hearing. So, the verdict seems to be that they can't be heard in
isolation but may impact perception of sound within the audible range.

Just wondering, does material outside the usual hearing range contribute
or detract from perceived 'realism' of recordings? Quite relevant
because we have a whole niche industry selling products offering
performance at at 20 kHz and above that may/may not be relevant to the
sound we can hear.

Of course, subsonics are an example of sounds that may not be heard but
are widely known to be perceived.

Yes, this all rather begs the question of _just_ how far we should go
in pursuit of 'perfect reproduction' of the original 'sonic
experience'. Do we (or even _should_ we) go so far as to risk hearing
damage by reproducing the full range of frequencies and SPLs or do we
make some reasonable compromise in this regard?

One excellent example of using a compromise to successfully emulate a
sonic experience is demonstrated in the "Wolfenstein 3D" 'demo' games
in regard of the pistol shots.

Firing a hand gun without ear defenders causes the ear's own AGC to
kick in in response to the initial high level transient after clipping
the 'crack' of the shot so that the ensuing reverberations and other
incidental noises become very muted, initially swiftly returning back
to more normal sensitivity levels in an exponential return to full
sensitivity.

Obviously, in a computer game, no sane game designer is going to ask
or even expect the serious gamer to invest in a KW rated PC speaker
system or use 10 watts per channel into efficient closed back
headphones to provide the 40 or 50 dB extra dynamic range to permit
the game player the 'full experience' of firing such a weapon under
combat conditions. However, what the clever game designer has done
here, is to impose a simulation of the ear's own dynamic sensitivity
response on the sound track's volume level curve so that what the
listener experiences (in a reasonably quiet environment free of other
sonic distractions that would otherwise 'give the game away') is very
close to the perception of the real experience but without imposing
any damaging stresses on the gameplayer's hearing.

I've no doubt the experience of fire arms enthusiasts were enlisted
to fine tune the simulation of the 'pistol shots' at least to the
point of eliminating unsoliicited criticism from those game players
who've actually fired such weaponery. It certainly met my expectations
of just how one would percieve such sounds, ignoring the physical pain
and damaging effects of the 'real thing'.

I have to say, it was this attention to detail that left a lasting
impression even though, by today's standards, the screen resolution
was ludicrously low (however, it was, at the time, the best resolution
in a "First Person '3D' Point and Shoot 'em up" game, mainly, it has
to be said, on account it was the _only_ "First Person '3D' Point and
Shoot 'em up" game available at that time :-)

The point I was making is that we don't necessarily want to reproduce
the whole of the audio spectrum of an original performance with such
extreme accuracy when very often all we really require is
verisimilitude.

Admittedly such processing to 'simulate' the dynamic effects of the
human ear can only be optimised for a small range of 'ideal' SPLs but
this hasn't stopped us fitting and using volume controls in the past
and to this day. Even extreme audiophiles will accept such compromises
as a necessary evil to accommodate the needs of others and also to
trade off reality against the need to preserve one's sense of hearing
for other novelties. IOW, the volume control is a tacit admission that
we're never going to achieve the Nirvana of sonic perfection (at least
not in this 'Plane of Existence'. :-)


I'm interested in Don's report of sudden onset of IM products. I wonder
where in the ear they are being generated...drum, stapes, cochlea,
brain? Plenty of scope for non linearity in that soft squidgy stuff we
are made from.

A lot of the 'overload' distortion arises in the the modified jaw
bones (stapes, hammer and anvil) when, just like a mistracking stylus,
they part company at their points of contact on sound pressure peaks.

It's actually possible to observe this type of 'in the ear'
distortion' when stood too close to the speakers at a rock music
performance by making the 'overload' distortion go away by the simple
expedient of plugging your ears with your fingers. The disappearance
of such 'clipping distortion' products has nothing to do with the LPF
effect associated with shoving a finger in each ear and more to do
with reducing the SPL at the ear drum.

The other sources of distortion in the cochlea and the central
nervous system become less quantifiable (in that order) but
undoubtedly exist. It's the very last part of the chain that's the
least quantifiable and hardest to understand (the cochlea is merely a
more subtle example of the mechanical problems demonstrated by the
three smallest bones in the human body).

Here we get ourselves into the realm of 'Psycho-acoustics' where a
dichotomy exists between our processing of sounds to allow us to
overcome interference effects in the environment in order to extract
the important 'information' that assisted our survival against other
threats as well as more recently, understand messages from fellow
members of our immediate tribe once our species had evolved into
co-operative hunter gatherer groups, a skill that finally evolved into
the one that lets us speak to each other, even under extremes of
interference. That same skill that allows us to appreciate the
'message' contained in 'musical works of art' ( that is, the ability
to enjoy a modern pop record despite the less than ideal listening
conditions :-).

The thing is, such 'signal processing' powers of the brain, can be
'reprogramed' by the mind contained within that brain. Indeed, such
'reprograming' is an ever continuing process during each individuals'
lifetime. It's a consequence of the 'adaptability through learning'
that gave our otherwise ill armoured, under-weaponised bodies the edge
over other 'better equipped' prey and predator species.

The consequence, of course being that we can choose to ignore
deficiencies in the whole audio system in order to concentrate on the
message in the music (including MP3 encoding/decoding errors) or else,
in looking for 'more detail' in the playing of more complex
orchestrations, find ourselves picking out the deficiencies of the
whole audio system.

Dedicated audio engineers, being engineers, like to define 'The
Problem' as simply as possible (hence the basic definition of working
over a range of frequencies limited to between 20 and 20000Hz with a
dynamic range of 90dB and almost zero distortion within that well
defined audio spectrum).

Such 'limits' are wide enough to satisfy all but a very few dedicated
Hi-Fi enthusiasts so dedicated in their pusuit of sonic Nirvana as to
be willing to take the risks involved in reproducing the more extreme
musical performances (the cannonade in Tchaikovsky's 1812 Overture,
anyone? see the comments for this you youtube example
https://www.youtube.com/watch?v=VbxgYlcNxE8 )

I'm currently listening to it as I type. I started with the volume
set at "Easy Listening Level" and it's at the 4 minute mark. I'll let
you know whether I agree with the comments in another eleven minutes
or so. :-)

The point is, I feel any such such 'Ultra Hi-Fi' system will need to
incorporate a mode switch marked "SOAR"[1] and "Ultra (risk of some
hearing loss)".

I've just finished auditioning that youtube rendition. Sadly, it was
a lttle disappointing since either the recording itself failed to
capture the dynamic range or the the streaming wasn't up to handling
the job. The cannon shots were obviously clipped with no sign of
psycho-acoustic processing as per the Wolfenstein 3D example of pistol
shots which so excellently dealt with the problem.

However, I have to admit I was only using a pair of PC speakers,
admittedly of above average quality and box volume, rather than
through my 50+50W RMS per channel (200W PMPO) mini power amp and large
speakers capable of rattling the otherwise unrattleable windows of the
downstairs lounge.

I suspect I'd have still heard the same shortcoming. There are so
many places in the transmission chain for such clipping effects to
manifest themselves, including the recording itself for which there
was no information as to its origin. Too many unknowns to fathom out
the cause for my sense of disappointment. :-(

[1] SOAR = "Safe Operating Area Restrictions applied"
--
J B Good

On 11/03/2015 9:23 PM, tony sayer wrote:
In article , Trevor Wilson trevor@SPA
MBLOCKrageaudio.com.au scribeth thus
On 10/03/2015 2:29 AM, Java Jive wrote:
On Mon, 09 Mar 2015 12:58:54 +1100, Trevor Wilson
wrote:

On 9/03/2015 2:15 AM, Java Jive wrote:
As I said earlier, these days manufacturers take even less care with
vinyl than they do with other media sources.

CD technology could and should have given us a flat FR from near DC up
to 25kHz, easily covering the range of human hearing.

**Bull****. With a sampling rate of 44.1kHz, Nyquist tells us that the
theoretical maximum of CDs is 22.05kz.

As it was actually implemented it was something of a compromise,
sacrificing FR to give greater playing time, but the space available
on the prototype technology "COULD AND SHOULD" (note what I actually
wrote) have been allocated differently to give us a shorter playback
time at a higher sampling rate that would have covered the range of
human hearing.

Even going up
to 22kHz it still covers quite comfortably the range of older
listeners such as myself.

**It actually covers pretty much everyone over the age of 10.

I and several others tested the range of our hearing in the Physics
Lab at college when I was about 17 or 18, and I wasn't the only one in
the group who could hear above 23KHz at that age.

**AT WHAT LEVEL?

The human ear does not have a 'brickwall filter' at 20kHz. No one ever
suggested it did. I recall visiting a warehouse owned by the company I
worked for when I was around 25 years old. I was assaulted by the most
appalling 'feeling' and I had to immediately leave. Curious, I decided
to work out what was going on. Turns out I was reacting to the
ultrasonic burglar alarm system. A microphone, preamp, oscilloscope and
frequency counter showed me that the space in the warehouse was
constantly flooded with a high intensity acoustic signal of
approximately 26kHz. OTOH, using conventional hearing tests, my hearing
extended to around 19kHz at the time.

MORAL: With sufficient intensity, human hearing can extend way past
20kHz. That intensity is NEVER achieved with any commercial recordings.


Didn't Eckersley of the BBC prove the effects of filters operating above
the usual 15 kHz limit did cause audible differences?..


**Well, it would, wouldn't it? Phase shift and all that.... The human
hearing system is exquisitely sensitive to phase shift.


--
Trevor Wilson
www.rageaudio.com.au

---
This email has been checked for viruses by Avast antivirus software.
http://www.avast.com

On Thu, 12 Mar 2015 06:40:05 +1100, Trevor Wilson
wrote:


**Well, it would, wouldn't it? Phase shift and all that.... The human
hearing system is exquisitely sensitive to phase shift.


The human hearing system is essentially totally deaf to phase shift.
I've seen an experiment in which the phase of the harmonics of a
square wave was cyclically shifted. You could see the waveform
changing shape on a scope. Even with the picture in view, it was
impossible to hear any change in the resulting sound.

Deafness to phase shift is the reason why Orban's Optimod system is
able to operate inaudibly.

d