[Don Pearce[_3_]]

On Mon, 15 Jul 2013 09:34:10 -0700, (hank alrich)
wrote:

Don Pearce wrote:

On Sun, 14 Jul 2013 17:49:54 +0200, (Ralf R.
Radermacher) wrote:

Don Pearce wrote:

I still have no idea what this truth is that you believe you have
revealed to me.

Never mind.

No, come on. All you managed to say to me in this series of posts is
that people have been trying to make speakers better. Was that it,
seriously?

Do you actually believe that it is the quality of speakers that makes
the difference between something that sounds live and something that
sounds recorded? If you are a professional sound person, that makes me
a little sad.

d

I have noted a direcdt relationship between quality (my interpretation
thereof) and the quality of the mimicing of the source, while obviously
I note many other factors that contribute, ambience being a huge one.

I understand you, but the real question is, has a recording ever
fooled you into thinking you were listening live? Obviously it would
have to be happening "round the corner" so you didn't actually know in
advance.

d

[William Sommerwerck]

"hank alrich" wrote in message
...
Gary Eickmeier wrote:

The recording and reproduction process is not intended
to make it sound like you are sitting at the mike position.

Who sez?

Speak for yourself and for the recordings you make. Some of us
do target a bullseye that is the source in the setting, to the best
we can capture that.

Correct. Ideally, we should be able to find a spot in the hall that's
acoustically appropriate for the music and its interpretation, then plop down
the mic and get that sound.

[William Sommerwerck]

"Don Pearce" wrote in message
...
On Mon, 15 Jul 2013 09:34:10 -0700, (hank alrich)
wrote:

I have noted a direcdt relationship between quality (my interpretation
thereof) and the quality of the mimicing of the source, while obviously
I note many other factors that contribute, ambience being a huge one.

I understand you, but the real question is, has a recording ever
fooled you into thinking you were listening live? Obviously it would
have to be happening "round the corner" so you didn't actually know
in advance.

True, from a strictly scientific point of proof.

But I know what concert halls (in general) sound like, and good surround
recordings come within rock-tossing distance of sounding like that.

Then there's the Plasmatronics speaker...

[Frank Stearns]

"Gary Eickmeier" writes:

Frank Stearns wrote:
"Gary Eickmeier" writes:


4. Spatial characteristics. This is the biggie that the Bose research
introduced. If you make a plan view drawing of your speakers and
room, you can draw the reflected sound as virtual images on the
other side of the walls. You can also make an image model drawing of
a live band in the hall. We get the spatial more correct by trying
to make the repro model as much like the live model as possible.

Sorry, but no!!! You've bought into Bose marketing (I remember those
funky zig-zag flared arrow diagrams from 1960s Bose literature).
You're never going to match a domestic living room to a concert hall.
You can only do that by putting your speakers in your own concert
hall (at 1:1 scale to the original venue). But then make sure you
close mic and record anechoically, and put a speaker on your private
stage matching the position of each microphone!

It took me a long time as well to understand the difference between the
spatial and the temporal. You have the classic confusion between the two.
This confusion is shown by your reference to the room sizes not being the
same. Reflection times would be the temporal, not the spatial. Let me
illustrate the difference with a simple example.

- snips -

In terms of acoustics, "temporal" and "spatial" are inexorably
linked. It takes time for sound to travel a distance through space,
and distance traveled implies time.

While I appreciate your example of four surround speakers in a row
failing to work properly, that's a bit like saying all four walls of
the concert hall were lined up in a row, such that we just had a
single long wall, and not a box. Both examples are something of red
herrings.

The Bose idea of some sort of mimicry of concert hall reverb geometry
in your home -- the "spatial" -- falls short because you cannot
detach that aspect from the "temporal". It might work if you could
also scale down the speed of sound to match your miniaturized concert
hall setting, but you cannot.

So, instead of the rich palette of different reverb times ranging
from 1 mS out to several seconds in a large hall, Bose just gives you
the end at 5-20 ms, as defined by your room. Worse, it's _new_ 5-20
mS stuff that's was NOT in the original recording. You superimpose
that on top of original venue spatial cue information at the wrong
time and you have that smeared mess.

Given that an untreated room is actually often worse -- the Bose
smear, being at least somewhat evenly distributed within the 5-20 ms
range -- can actually help. But that "help" is limited and for a
limited range of program material.

When you can properly reproduce original recording spatial cues -- be
it through a LEDE, RFZ, or any number of other proven reproducing
room designs -- you're light years ahead of the Bose approach.

I use LEDE because it's the best option in a small space. Properly
implemented, its ability to make the speakers disappear and put you
in a seemingly much larger and deeper space is remarkable. And you do
not sacrifice razor sharp imaging nor clarity, like you do with Bose.
(But I can also tell you how LEDE fails when *not* set up properly --
my first room went through that evolution, so I got to know and
understand the failure modes.)

In most approaches to room/monitor design, we are indeed adding
_some_ local reverberant field generated in the reproducing room
(listening in an anechoic space is bad at the other extreme so we
do need some sort of local reverb). The question is just exactly
*when* do we add that information, in what distribution, and to what
degree.

My argument with Bose is that the added delay is too high in level,
too soon, and is far too room dependent. And because of that, the
utility of the approach is limited.

Pile on the other issues with the Bose approach -- IM distortion, no
treatment for other bugaboos such as LF problems, etc -- and you're
generally not in for a happy time.

But as noted, if your room is not so good the basic "Bose haze" that
masks various problems is probably an okay way to go. But I'd rather
fix the room (or design it from the ground up) to remove the need for
any "haze," and instead hear what's really there.

But as always, YMMV.

Frank
Mobile Audio
--

[[email protected]]

One of the ways of recognizing live music is due to Dynamic range especially with drums.
Very few recordings capture the full dynamic range of drums.
Mark

[Gary Eickmeier]

William Sommerwerck wrote:
"hank alrich" wrote in message
...
Gary Eickmeier wrote:

The recording and reproduction process is not intended
to make it sound like you are sitting at the mike position.

Who sez?

I sez.


Speak for yourself and for the recordings you make. Some of us
do target a bullseye that is the source in the setting, to the best
we can capture that.

Correct. Ideally, we should be able to find a spot in the hall that's
acoustically appropriate for the music and its interpretation, then
plop down the mic and get that sound.

Ok, then we should want to listen from the perspective of three ears about 6
ft apart hanging about 9 ft above the conductor's head, plus a couple inside
the lid of the piano, another mono ear about a foot in front of the singer's
mouth, and a few spot ears nearer to the guitar and harp than would be
comfortable if you were there.

Gary Eickmeier

[Gary Eickmeier]

Frank Stearns wrote:

In terms of acoustics, "temporal" and "spatial" are inexorably
linked. It takes time for sound to travel a distance through space,
and distance traveled implies time.

While I appreciate your example of four surround speakers in a row
failing to work properly, that's a bit like saying all four walls of
the concert hall were lined up in a row, such that we just had a
single long wall, and not a box. Both examples are something of red
herrings.

The Bose idea of some sort of mimicry of concert hall reverb geometry
in your home -- the "spatial" -- falls short because you cannot
detach that aspect from the "temporal". It might work if you could
also scale down the speed of sound to match your miniaturized concert
hall setting, but you cannot.

So, instead of the rich palette of different reverb times ranging
from 1 mS out to several seconds in a large hall, Bose just gives you
the end at 5-20 ms, as defined by your room. Worse, it's _new_ 5-20
mS stuff that's was NOT in the original recording. You superimpose
that on top of original venue spatial cue information at the wrong
time and you have that smeared mess.

Given that an untreated room is actually often worse -- the Bose
smear, being at least somewhat evenly distributed within the 5-20 ms
range -- can actually help. But that "help" is limited and for a
limited range of program material.

When you can properly reproduce original recording spatial cues -- be
it through a LEDE, RFZ, or any number of other proven reproducing
room designs -- you're light years ahead of the Bose approach.

I use LEDE because it's the best option in a small space. Properly
implemented, its ability to make the speakers disappear and put you
in a seemingly much larger and deeper space is remarkable. And you do
not sacrifice razor sharp imaging nor clarity, like you do with Bose.
(But I can also tell you how LEDE fails when *not* set up properly --
my first room went through that evolution, so I got to know and
understand the failure modes.)

In most approaches to room/monitor design, we are indeed adding
_some_ local reverberant field generated in the reproducing room
(listening in an anechoic space is bad at the other extreme so we
do need some sort of local reverb). The question is just exactly
*when* do we add that information, in what distribution, and to what
degree.

My argument with Bose is that the added delay is too high in level,
too soon, and is far too room dependent. And because of that, the
utility of the approach is limited.

Pile on the other issues with the Bose approach -- IM distortion, no
treatment for other bugaboos such as LF problems, etc -- and you're
generally not in for a happy time.

But as noted, if your room is not so good the basic "Bose haze" that
masks various problems is probably an okay way to go. But I'd rather
fix the room (or design it from the ground up) to remove the need for
any "haze," and instead hear what's really there.

But as always, YMMV.

All I can tell you Frank is the above shows extreme ignorance of
psychoacoustics. You cannot hear 5 to 20 ms of delay as an echo in any way.
All that is happening is a change in the spatial patterns from the front of
the room. Again, the temporal is contained in the recording, not injected by
the reflecting speakers.

So how do you justify the later reflections with the LEDE approach?

Gary Eickmeier

[Gary Eickmeier]

William Sommerwerck wrote:


Correct. But high-accuracy speakers are still desirable, even if only
to assure that the home listener hears what the studio producer
intended.
By the way, my experience has been that even teenage bimbas dragged
in off the street tend to prefer speakers with flat response.

What the devil is a "high accuracy speaker"? What would the radiation
pattern be to achieve this "accuracy"? Accuracy of what to what? Are you
just talking freq response and distortion, and ignoring everything else? Or
do you know about all of the audible properties of speakers and rooms?

Flat response? Which response? The one meter on axis anechoic response? The
room response? The power response? Are you aware of any of these complicated
issues?

Gary Eickmeier

[William Sommerwerck]

"Gary Eickmeier" wrote in message ...

William Sommerwerck wrote:
"hank alrich" wrote in message
...
Gary Eickmeier wrote:

The recording and reproduction process is not intended
to make it sound like you are sitting at the mike position.

Who sez?

I sez.

You're incorrect.


Speak for yourself and for the recordings you make. Some of us
do target a bullseye that is the source in the setting, to the best
we can capture that.

Correct. Ideally, we should be able to find a spot in the hall that's
acoustically appropriate for the music and its interpretation, then
plop down the mic and get that sound.

Ok, then we should want to listen from the perspective of three ears about 6
ft apart hanging about 9 ft above the conductor's head, plus a couple inside
the lid of the piano, another mono ear about a foot in front of the singer's
mouth, and a few spot ears nearer to the guitar and harp than would be
comfortable if you were there.

You completely misunderstand. That is 100% backwards.

Please reread what was said, and think about it.

[jason]

On Mon, 15 Jul 2013 01:49:21 -0400 "Gary Eickmeier"
wrote in article z2MEt.3857$MM1.3075
@fx25.am4

Arny Krueger wrote:
"Scott Dorsey" wrote in message
...

It's also important to point out that Bose marketed a line of
wretchedly bad
speakers (with highly inflated "list" prices) that gave Bose
dealers an opportunity to sell Bose products at a "discount".

The Bose speakers are designed to sound very good with short
listening tests.
They sound brighter and boomier than the competition, because that's
what sells speakers: brief listening comparisons.

Bose looked at how people shop for and buy speakers and then designed
speakers to perform well under those circumstances. That is a sort
of marketing genius.

The most-damning thing one can say about Dr Bose and his company is
that they
did absolutely nothing whatsoever to advance the art of sound
reproduction.
Merchandising is another matter -- Bose mastered The Big Lie long
before Apple.

Actually, Bose has done a lot of research into speaker design and
radiation
patterns, and there are a bunch of papers in the JAES that were
funded by Bose. Mind you, Bose doesn't use this research to make
good sounding speakers, because they aren't in the business of
making good sounding speakers. But you cannot fault Bose for not
helping to advance the state of the science even if not the
technology.

If you get into how Bose does things you can often figure out what
they had in mind when they created the product and generally they do
a good job of creating what they want to create.

Above we see several examples of that.

As you point out, what they want to create is very often something
that people think they want long enough so Bose can capture their
money and the people keep the product.

Methinks Mr. Dorsey doesn't know much about the Bose pro division and all of
the sound reinforcement and DJ equipment and the Modeler program that can
place speakers in a virtual room and play what they would sound like in any
position of the listener in that room. The wonderful L1 speakers and their
successors and now the Panarray line which I don't know that much about yet,
but I am not in that business.

Once upon a time I attended a stage magic show, and rather than pay a live
orchestra they just had a stack of Bose 802s on each side of the stage that
sounded so good I couldn't believe it wasn't live. But of course these
stories are legend all over the world with the Bose pro line.

The automotive stuff, I am not so sure. Sounds a little too heavy on the
bass from what I have heard. Maybe it is because at speed it needs it
(because of road noise).

Gary Eickmeier

Anecdotal-that's-all... On 4 July, I attended an outdoor event at Mohonk
Mountain House in New Paltz. Before the fireworks, there were some
speeches and some a capella performances by a small group and some solo
singing. The PA system comprised 6 Bose speakers, each was a narrow
column of drivers anchored to (I presume) a subwoofer. They did an
excellent job "illuminating" a difficult space, replete with rock-hard
(literally) reflective surfaces and a lake. I had never heard these
before.

[Frank Stearns]

"Gary Eickmeier" writes:

self snips

But as always, YMMV.

All I can tell you Frank is the above shows extreme ignorance of psychoacoustics

Er, my clients would tend to disagree with you...

You cannot hear 5 to 20 ms of delay as an echo in any way.

Not as an discrete echo, no -- 40 ms is less than the Haas boundary.

But again, you can certainly hear the effects of aberrations in the 5-20 ms range --
what we often call good old fashioned comb filtering!!!

I urge you to sit down with your favorite DAW and play with delays in the 2-20 ms
range. Take a "direct" signal, and then route a duplicate through a very short delay
that you can step in 1 ms or less increments. Blend the two and start varying the
delay value. Pay attention to what's happening between 5-10 ms, the delay range
you'd get with a typical 901 placement. Do you get the idea now??? This is what
you're doing to the music with your 901s!

All that is happening is a change in the spatial patterns from the front of
the room. Again, the temporal is contained in the recording, not injected by
the reflecting speakers.

Argh! It's the spatial cue information contained in the source that can be lost,
smeared, comb filtered away by the ham-handed way Bose achieves this with the
typical 901 set up.

So how do you justify the later reflections with the LEDE approach?

While important, actual hard reflection is a very *small* component of the LEDE
sound (perhaps 2% as measured by the treated surface area of the back and rear side
walls). And ideally, it's in the 20-30 ms range. (My previous room had that hard
reflection at 17 ms, but I managed to "steer" it just a bit to mitigate the
shorter-than-optimal time.)

The more important components a

- no early reflections

- no diffraction around the speaker cabinets; no reflections of ANY KIND at any hz
from the wall immediately behind the speakers. This gets tricky 200 hz, but it can
be done.

- carefully placed diffraction and diffusion on the back and rear side walls such
that the RT30 at the sweet spot is 100 mS in the very low end, rising gradually to
around 1.0-1.2 seconds at 10 Khz. The critical range around 3K has an RT30 of around
200-400 ms.

In all those cases the reverb field has a very, very smooth decay. With the Bose
approach, you're simply creating a few wavefront "slaps" off those back and side
walls. And again, those slaps are too loud and happening too soon.

Frank
Mobile Audio
--

[Frank Stearns]

"Gary Eickmeier" writes:

snips

Ok, then we should want to listen from the perspective of three ears about 6
ft apart hanging about 9 ft above the conductor's head, plus a couple inside
the lid of the piano, another mono ear about a foot in front of the singer's
mouth, and a few spot ears nearer to the guitar and harp than would be
comfortable if you were there.

If you were spot placing "ears" as you suggest (ears attached to functioning
brains), it would indeed be atrocious.

Unfortunately, we're dealing with microphones, speakers, and often very poor
listening rooms. Those and many other issues conspire to shoot down that sense of
"being there."

But we capture all that "extra stuff" to give us things we can use to construct an
illusion that hopefully will withstand all the various shortcomings along the way.

We might not use all that we capture, or use it differently from someone else,
depending on experience and what the client might want. But having captured it, it's
available.

Frank
Mobile Audio
--

[Peter Larsen[_3_]]

William Sommerwerck wrote:

"Gary Eickmeier" wrote in message
...

This is a pretty poor commentary Bill. Not worthy of you.

When you consider the people who /have/ made real advancements in
sound recording and playback, I feel this is a pretty accurate
assessment of Dr Bose.

I sincerely ask -- what significant contributions did I miss?

First of all that his point about direct vs. reflected is correct. However
it needs to be fixed in the recording not in the playback.

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Don Pearce wrote:

But do bear in mind that with the recorded source material available,
the chances of even a perfect speaker sounding remotely like live
music are essentially zero. An anechoic recording would be the first
requirement, and musicians generally play very badly in such a
setting.

First of all acoustic instruments lack the feedback they get from a room.

d

Kind regards

Peter Larsen

[Paul[_13_]]

On 7/14/2013 11:50 PM, Trevor wrote:
"geoff" wrote in message
...
Not everybody listens exclusively to live orchestral or chamber music.
Much other music is 'produced', and a 'closest to live' analogy is not
relevant.

Actually "close to live" becomes far easier for electronic instruments and
miked vocals recorded live. Use the same speakers and the only real variable
is the room. Make the listening room dead and add room simulation from the
original and you're pretty close.

Trevor.


I would agree. If the "live" situation is heard mostly through the
PA speakers, and or guitar and keyboard amps, then the sound sources
were originally speakers anyways.

In these cases, the main clue that the music is live usually comes
from the drummer(s), because they are often loud enough to be heard
acoustically, without amplification.

An intimate, un-amplified acoustic performance is different, of
course...

[Frank Stearns]

Les Cargill writes:

-snips-

The Haas limit is 10msec, generally.

Actually, its perception is somewhat variable depending on program material. Might
be as short as 5 ms for transient-type info, such as drum hits (though I'm not sure
I believe that), but up to 40 ms for more complex material, such as music.

The longer time seems to match more closely with experience. It does seem to be
something of a "fuzzy" boundary within human perception.

Frank
Mobile Audio
--

[Gary Eickmeier]

Peter Larsen wrote:
Don Pearce wrote:

But do bear in mind that with the recorded source material available,
the chances of even a perfect speaker sounding remotely like live
music are essentially zero. An anechoic recording would be the first
requirement, and musicians generally play very badly in such a
setting.

First of all acoustic instruments lack the feedback they get from a
room.

As I have mentioned, the final result is a combination of the recorded
acoustic and the playback acoustic, hopefully more of the former in a normal
home environment. Think of it as a continuum from a totally anechoic
recording, which would place the instruments in your listening room, to the
field-type stereophonic recording, in which the microphones are placed close
to the instruments in the recording space in the region of the proscenium,
or soundstage, and finally to the binaural recording with the dummy head
placed in the "best seat in the house" out in the audience so that it can
hear the total concert hall acoustic, listened to on headphones to isolate
the listener from the room he is in.

In a normal stereo recording we need to record the early and some
reverberant sound to get the total sound power output by the instruments in
all directions, plus the spatial and temporal dimensions of the hall that
give the sound the qualities that make for a good performing acoustic. These
qualities are brought back, or decoded, on playback by spreading the sound
wider and deeper with some delay within the fusion time. This can be done
with extra speakers to the front and rear sides on delay, or the most
elegant solution to just model the playback sound patterns after the live
situation by employing reflected sound from a properly designed
multi-directional speaker system. Speaker placement is critical in
establishing these frontal patterns.

Gary Eickmeier

[Gary Eickmeier]

Frank Stearns wrote:
"Gary Eickmeier" writes:

snips

Ok, then we should want to listen from the perspective of three ears
about 6 ft apart hanging about 9 ft above the conductor's head, plus
a couple inside the lid of the piano, another mono ear about a foot
in front of the singer's mouth, and a few spot ears nearer to the
guitar and harp than would be comfortable if you were there.

If you were spot placing "ears" as you suggest (ears attached to
functioning brains), it would indeed be atrocious.

Unfortunately, we're dealing with microphones, speakers, and often
very poor listening rooms. Those and many other issues conspire to
shoot down that sense of "being there."

But we capture all that "extra stuff" to give us things we can use to
construct an illusion that hopefully will withstand all the various
shortcomings along the way.

We might not use all that we capture, or use it differently from
someone else, depending on experience and what the client might want.
But having captured it, it's available.

Frank
Mobile Audio

From a response I just gave a few posts above:

As I have mentioned, the final result is a combination of the recorded
acoustic and the playback acoustic, hopefully more of the former in a normal
home environment. Think of it as a continuum from a totally anechoic
recording, which would place the instruments in your listening room, to the
field-type stereophonic recording, in which the microphones are placed close
to the instruments in the recording space in the region of the proscenium,
or soundstage, and finally to the binaural recording with the dummy head
placed in the "best seat in the house" out in the audience so that it can
hear the total concert hall acoustic, listened to on headphones to isolate
the listener from the room he is in.

In a normal stereo recording we need to record the early and some
reverberant sound to get the total sound power output by the instruments in
all directions, plus the spatial and temporal dimensions of the hall that
give the sound the qualities that make for a good performing acoustic. These
qualities are brought back, or decoded, on playback by spreading the sound
wider and deeper with some delay within the fusion time. This can be done
with extra speakers to the front and rear sides on delay, or the most
elegant solution to just model the playback sound patterns after the live
situation by employing reflected sound from a properly designed
multi-directional speaker system. Speaker placement is critical in
establishing these frontal patterns.

Gary Eickmeier

[William Sommerwerck]

"Gary Eickmeier" wrote in message ...
William Sommerwerck wrote:

Correct. But high-accuracy speakers are still desirable, even if only
to assure that the home listener hears what the studio producer
intended.
By the way, my experience has been that even teenage bimbas dragged
in off the street tend to prefer speakers with flat response.

What the devil is a "high accuracy speaker"?

It's a speaker whose driver motion accurately follows the input signal. Its
radiation pattern is a secondary consideration.

[Scott Dorsey]

Gary Eickmeier wrote:

All I can tell you Frank is the above shows extreme ignorance of
psychoacoustics. You cannot hear 5 to 20 ms of delay as an echo in any way.
All that is happening is a change in the spatial patterns from the front of
the room. Again, the temporal is contained in the recording, not injected by
the reflecting speakers.

I hate to tell you this, Gary, but Frank is a lot closer to the mark than
you are.

And no, you don't hear 5 to 20 mS of delay as an echo... but you see it on
the scope as an echo. It's the same thing, you just hear it differently.

So how do you justify the later reflections with the LEDE approach?

They're what makes the system work, and they are mostly coming from behind
you. That, combined with the fact that they are late, prevents them from
interfering with the image.

Really, honestly, this stuff was very well worked out long ago. Please
stop beating this dead horse.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Gary Eickmeier]

Frank Stearns wrote:
"Gary Eickmeier" writes:

But again, you can certainly hear the effects of aberrations in the
5-20 ms range -- what we often call good old fashioned comb
filtering!!!

Here again I must introduce you to my universe, one in which the radiation
pattern of the speakers is much different from what you are used to with
this "comb filtering" so-called problem. Comb filtering would be the
strongest when the delayed sound and the direct sound are about equal. But
in the 901 design the majority of the sound that you hear is the reflected
and just a small amount direct, eliminating the combing almost completely.

I urge you to sit down with your favorite DAW and play with delays in
the 2-20 ms range. Take a "direct" signal, and then route a duplicate
through a very short delay that you can step in 1 ms or less
increments. Blend the two and start varying the delay value. Pay
attention to what's happening between 5-10 ms, the delay range you'd
get with a typical 901 placement. Do you get the idea now??? This is
what you're doing to the music with your 901s!

The trick to remember in my universe is that the direction from which the
delayed sound comes is all important. It MUST come from incident angles that
are different from the direct sound. In your DAW example, the correct
experiment to try would be to take the normal front 2 channels that you
listen on and feed them to some extra speakers spread way wider and deeper
on time delay of anywhere from 5 to 20 ms I would guess. What it would do is
give a spatial broadening of the recorded sound that makes it sound more
like what you would have heard live than you had with just the two direct
speakers.

Again, the problem with stereo is compressing the immense soundstage of
direct and reflected sounds that were recorded into a set of angles
represented by the two front speakers and causing the recorded early
reflected sound to come from the same limited set of angles as the direct.
Even a brief encounter with some acoustics textbooks will tell you that
can't work.

Part of the problem is stereo theory. You are all thinking that stereo works
by shooting the sound from a couple of front speakers straight to your ears,
and then you will "hear" it. I say that is a first blush ignorant version of
how a field type system works that an audiophile club might think. It is
because the literature on stereo always has this little triangle of two
speakers and a head in an equilateral triangle, and that is how "stereo"
works. Then the engineers think that anything that adds to or interferes
with this happy little idea of the sound straight from the speakers going to
your ears will deteriorate the sound with "smearing" and "comb filtering"
and all manner of problems not contained in the recording.

It is my ongoing battle to try and relate that the recording contains a lot
more information than just the direct sound, and does not work like binaural
by sticking the channels directly into your ears. In addition to the direct
sound contained in the recording there is a set of widely spaced reflected
sounds from the early reflections that were recorded, and those MUST be
brought back from angles that are DIFFERENT from the direct sound - as in
not riding along with the direct sound, which is not the way it happened
live.

Can anyone out there understand all that? It is about the SPATIAL nature of
what was recorded, and the need to decode that on playback, because it is a
field type system and we can hear all spatial characteristics of our
speakers and room and we must not change those characteristics from the huge
live field to the two ****-holes in the snow represented by two speakers at
the front of an LEDE room.

Live sound patterns.

Playback sound patterns.

Bose spider drawings or Eickmeier image model drawings, take your pick.

Do not choose two ****-holes in the snow in a quest for your "accuracy" of
the two raw channels.

Choose realism. Match playback patterns to live patterns.

Amen

Gary Eickmeier

[Gary Eickmeier]

Peter Larsen wrote:

First of all that his point about direct vs. reflected is correct.
However it needs to be fixed in the recording not in the playback.

Kind regards

Peter Larsen

Argh.

Gary Eickmeier

[Scott Dorsey]

Les Cargill wrote:
Gary Eickmeier wrote:
Frank Stearns wrote:

snip

All I can tell you Frank is the above shows extreme ignorance of
psychoacoustics. You cannot hear 5 to 20 ms of delay as an echo in any way.

20ms mixed properly shows up as "echo". 5 ms, not so much. But they
sound "echoey", or "phasey." That's really just about word use.

No, no, this is important. It's more than just word use, it's two different
effects with the same origin. The same thing is happening, but the perception
is totally different. This is a very, very important concept because with
position changes it's possible to change one perceived effect into another.

If you split a mono signal left & right and delay one side 20ms, it
will be perceptible as "ambiance" or something like "stereo." Yeah,
it's more about image than "echo", but it's in that direction.

The comb filtering is _also_ perceived as ambiance. This is kind of cool,
and it's how the Orban stereo simulator works. This is a very handy
technique to have in your bag of tricks.

All that is happening is a change in the spatial patterns from the front of
the room.

I'm not sure of that. There are other things in play. it gets tricky.
In some cases, it's perceives as one; in others, the other. Depends
on the sound.

The Haas limit is 10msec, generally.

Yes, and that "generally" depends on the listener and the frequency. When
you're working with reflection distances around 10 feet or so, you're in
the borderline range between the two perceived effects and things can get
very, very weird.

Again, the temporal is contained in the recording, not injected by
the reflecting speakers.

So how do you justify the later reflections with the LEDE approach?

If there were rear speakers to reproduce the rear image accurately, they
could not be justified. But in a stereo situation, you live without that.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Gary Eickmeier]

Jason wrote:


Anecdotal-that's-all... On 4 July, I attended an outdoor event at
Mohonk Mountain House in New Paltz. Before the fireworks, there were
some speeches and some a capella performances by a small group and
some solo singing. The PA system comprised 6 Bose speakers, each was
a narrow column of drivers anchored to (I presume) a subwoofer. They
did an excellent job "illuminating" a difficult space, replete with
rock-hard (literally) reflective surfaces and a lake. I had never
heard these before.

Thank you Jason. From what I have heard, the Bose systems are famous for
solving sound reinforcement problems that others had failed at for years. I
think Montreaux is another example.

Gary Eickmeier

[Scott Dorsey]

Gary Eickmeier wrote:
Frank Stearns wrote:
"Gary Eickmeier" writes:

But again, you can certainly hear the effects of aberrations in the
5-20 ms range -- what we often call good old fashioned comb
filtering!!!

Here again I must introduce you to my universe, one in which the radiation
pattern of the speakers is much different from what you are used to with
this "comb filtering" so-called problem. Comb filtering would be the
strongest when the delayed sound and the direct sound are about equal. But
in the 901 design the majority of the sound that you hear is the reflected
and just a small amount direct, eliminating the combing almost completely.

No, not at all. You're hearing the reflected sound from a lot of different
paths, with a lot of different delays, all combined together with the live
sound. So what you're hearing is a lot of comb filters laid on top of one
another.

The end result is interesting in that you don't hear it the combing so much;
there are no deep notches because the notch in one path is not the same as
the notch on another path.

But what DOES happen is that if you send a pulse into the system, it comes
out greatly, greatly broadened, because it turns into a bunch of pulses all
delayed different times and superimposed on one another. This is the source
of the "smearing" effect that people dislike so much about those speakers.

I urge you to sit down with your favorite DAW and play with delays in
the 2-20 ms range. Take a "direct" signal, and then route a duplicate
through a very short delay that you can step in 1 ms or less
increments. Blend the two and start varying the delay value. Pay
attention to what's happening between 5-10 ms, the delay range you'd
get with a typical 901 placement. Do you get the idea now??? This is
what you're doing to the music with your 901s!

The trick to remember in my universe is that the direction from which the
delayed sound comes is all important. It MUST come from incident angles that
are different from the direct sound. In your DAW example, the correct
experiment to try would be to take the normal front 2 channels that you
listen on and feed them to some extra speakers spread way wider and deeper
on time delay of anywhere from 5 to 20 ms I would guess. What it would do is
give a spatial broadening of the recorded sound that makes it sound more
like what you would have heard live than you had with just the two direct
speakers.

This is exactly what traditional stereo is trying to prevent. This is
forcing additional ambiance to be created, and that ambiance is _fixed_
in its pattern. It is caused by the room rather than the original source
material, and whether you play a solo guitar recording or an orchestra,
the same ambiance is heard at the listener's position. This is very, very
bad, unless you listen to the same kind of music all the time.

Part of the problem is stereo theory. You are all thinking that stereo works
by shooting the sound from a couple of front speakers straight to your ears,
and then you will "hear" it. I say that is a first blush ignorant version of
how a field type system works that an audiophile club might think. It is
because the literature on stereo always has this little triangle of two
speakers and a head in an equilateral triangle, and that is how "stereo"
works. Then the engineers think that anything that adds to or interferes
with this happy little idea of the sound straight from the speakers going to
your ears will deteriorate the sound with "smearing" and "comb filtering"
and all manner of problems not contained in the recording.

No. Nobody thinks this. Read the New Stereo Soundbook or any introduction
to stereophony.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[William Sommerwerck]

"Gary Eickmeier" wrote in message ...
Peter Larsen wrote:

First of all that his point about direct vs. reflected is correct.
However it needs to be fixed in the recording not in the playback.

Argh.

I'm not quite what your Argh means, but Peter is correct. Broadly speaking,
the recording should contain all that is needed to correctly "present" the
performance. It is not the speaker's job to "interpret" or "enhance" the
recording.

[John Williamson]

William Sommerwerck wrote:
"Gary Eickmeier" wrote in message ...
Peter Larsen wrote:

First of all that his point about direct vs. reflected is correct.
However it needs to be fixed in the recording not in the playback.

Argh.

I'm not quite what your Argh means, but Peter is correct. Broadly
speaking, the recording should contain all that is needed to correctly
"present" the performance. It is not the speaker's job to "interpret" or
"enhance" the recording.

Although, IMHO, the recording should take into account the limitations
of likely playback mechanisms. Hence I make a slightly different mix of
the same tracks for internet streaming and CD playback.

--
Tciao for Now!

John.

[William Sommerwerck]

"John Williamson" wrote in message ...
William Sommerwerck wrote:
"Gary Eickmeier" wrote in message ...

I'm not quite what your Argh means, but Peter is correct. Broadly speaking,
the recording should contain all that is needed to correctly "present" the
performance. It is not the speaker's job to "interpret" or "enhance" the
recording.

Although, IMHO, the recording should take into account the limitations of
likely playback mechanisms. Hence I make a slightly different mix of the
same tracks for internet streaming and CD playback.

The problem is that carrying that approach to its logical extreme results
in -- at the very least -- musically unnatural recordings (eg, Dynagroove).

[Les Cargill[_4_]]

Scott Dorsey wrote:
Les Cargill wrote:
Gary Eickmeier wrote:
Frank Stearns wrote:

snip

All I can tell you Frank is the above shows extreme ignorance of
psychoacoustics. You cannot hear 5 to 20 ms of delay as an echo in any way.

20ms mixed properly shows up as "echo". 5 ms, not so much. But they
sound "echoey", or "phasey." That's really just about word use.

No, no, this is important. It's more than just word use, it's two different
effects with the same origin. The same thing is happening, but the perception
is totally different.

This will sound weird, but I think of the percpetion and word use as a
lot the same thing. they exist mainly in the brain and aren't ... real.

Maybe it's a bad habit, but I tend to lump things caused by the same
"system" together.

No doubt "word use" above is poor ... word use, but .... both are
cognitive artifacts imposed rudely on a nice signal that has no idea
what's being done to it.

"Cognitive effects" sound pretentious and probably is. So...

This is a very, very important concept because with
position changes it's possible to change one perceived effect into another.



Delaying something percussive by 20msec will have a different
effect than with a low, long cello note.

If you split a mono signal left & right and delay one side 20ms, it
will be perceptible as "ambiance" or something like "stereo." Yeah,
it's more about image than "echo", but it's in that direction.

The comb filtering is _also_ perceived as ambiance. This is kind of cool,
and it's how the Orban stereo simulator works. This is a very handy
technique to have in your bag of tricks.



Oh, absolutely. It's one of my favorite things. But it's subject to
overuse - you have to be careful with it.

All that is happening is a change in the spatial patterns from the front of
the room.

I'm not sure of that. There are other things in play. it gets tricky.
In some cases, it's perceives as one; in others, the other. Depends
on the sound.

The Haas limit is 10msec, generally.

Yes, and that "generally" depends on the listener and the frequency.

Yep.

When
you're working with reflection distances around 10 feet or so, you're in
the borderline range between the two perceived effects and things can get
very, very weird.


Absolutely.

Again, the temporal is contained in the recording, not injected by
the reflecting speakers.

So how do you justify the later reflections with the LEDE approach?

If there were rear speakers to reproduce the rear image accurately, they
could not be justified. But in a stereo situation, you live without that.
--scott



--
Les Cargill

[John Williamson]

William Sommerwerck wrote:
"John Williamson" wrote in message
...
William Sommerwerck wrote:
"Gary Eickmeier" wrote in message ...

I'm not quite what your Argh means, but Peter is correct. Broadly
speaking, the recording should contain all that is needed to
correctly "present" the performance. It is not the speaker's job to
"interpret" or "enhance" the recording.

Although, IMHO, the recording should take into account the limitations
of likely playback mechanisms. Hence I make a slightly different mix
of the same tracks for internet streaming and CD playback.

The problem is that carrying that approach to its logical extreme
results in -- at the very least -- musically unnatural recordings (eg,
Dynagroove).

Moderation is the key, as in all things, although I have been known to
remix from a totally blank canvas, rather than just nudge the faders a bit.

--
Tciao for Now!

John.

[Scott Dorsey]

In article ,
William Sommerwerck wrote:
"John Williamson" wrote in message ...
William Sommerwerck wrote:
"Gary Eickmeier" wrote in message ...

I'm not quite what your Argh means, but Peter is correct. Broadly speaking,
the recording should contain all that is needed to correctly "present" the
performance. It is not the speaker's job to "interpret" or "enhance" the
recording.

Although, IMHO, the recording should take into account the limitations of
likely playback mechanisms. Hence I make a slightly different mix of the
same tracks for internet streaming and CD playback.

The problem is that carrying that approach to its logical extreme results
in -- at the very least -- musically unnatural recordings (eg, Dynagroove).

It depends on the distribution method too. It's not unusual to make special
radio-friendly mixes that are a little brighter and denser than typical, and
it's not unusual to make a different mix for an MP3 single that will be
probably listened to on headphones than a CD release.

But it sure would have been nice if someone had marked some records
WARNING: DYNAGROOVE, PLAY ONLY ON GE PHONOGRAPHS WITH SPHERICAL STYLI.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Peter Larsen[_3_]]

Gary Eickmeier wrote:

Peter Larsen wrote:

Don Pearce wrote:

But do bear in mind that with the recorded source material
available, the chances of even a perfect speaker sounding remotely
like live music are essentially zero. An anechoic recording would
be the first requirement, and musicians generally play very badly
in such a setting.

First of all acoustic instruments lack the feedback they get from a
room.

As I have mentioned, the final result is a combination of the recorded
acoustic and the playback acoustic, hopefully more of the former in a
normal home environment.

I think you perhaps didn't quite get my point. Instruments hear the room
response and are influenced by it AND instruments hear each as well. A cello
played near a concert grand with a closed lid sounds quite different from
one played near the same concert grand with the lid open.

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

William Sommerwerck wrote:

"Gary Eickmeier" wrote in message
... William Sommerwerck wrote:

Correct. But high-accuracy speakers are still desirable, even if
only to assure that the home listener hears what the studio producer
intended.
By the way, my experience has been that even teenage bimbas dragged
in off the street tend to prefer speakers with flat response.

What the devil is a "high accuracy speaker"?

It's a speaker whose driver motion accurately follows the input
signal. Its radiation pattern is a secondary consideration.

No, as Harwood showed the directivity index has to be as close to constant
as possible and steps in it are a known cause of coloration.

Kind regards

Peter Larsen

[John Williamson]

Peter Larsen wrote:
Gary Eickmeier wrote:

Peter Larsen wrote:

Don Pearce wrote:

But do bear in mind that with the recorded source material
available, the chances of even a perfect speaker sounding remotely
like live music are essentially zero. An anechoic recording would
be the first requirement, and musicians generally play very badly
in such a setting.

First of all acoustic instruments lack the feedback they get from a
room.

As I have mentioned, the final result is a combination of the recorded
acoustic and the playback acoustic, hopefully more of the former in a
normal home environment.

I think you perhaps didn't quite get my point. Instruments hear the room
response and are influenced by it AND instruments hear each as well. A cello
played near a concert grand with a closed lid sounds quite different from
one played near the same concert grand with the lid open.

A`lot of that is the piano strings resonating to the cello's notes.
Close mic the cello, and you don't hear it anywhere near as much.

Players will also react to the room acoustic, making the instrument
sound different, as you'll no doubt have noticed.
--
Tciao for Now!

John.

[William Sommerwerck]

"Peter Larsen" wrote in message
...
William Sommerwerck wrote:

It's a speaker whose driver motion accurately follows the input
signal. Its radiation pattern is a secondary consideration.

No, as Harwood showed the directivity index has to be as close
to constant as possible and steps in it are a known cause of coloration.

If the radiating surface doesn't accurately follow the input, it doesn't
matter what the directivity index is.

[Don Pearce[_3_]]

On Tue, 16 Jul 2013 11:08:06 -0700, "William Sommerwerck"
wrote:

"Peter Larsen" wrote in message
k...
William Sommerwerck wrote:

It's a speaker whose driver motion accurately follows the input
signal. Its radiation pattern is a secondary consideration.

No, as Harwood showed the directivity index has to be as close
to constant as possible and steps in it are a known cause of coloration.

If the radiating surface doesn't accurately follow the input, it doesn't
matter what the directivity index is.

And conversely if the directivity index is poor, it doesn't matter how
accurately the radiating surface follows the input. The radiation
pattern is a long way from secondary in speaker design. It almost
defines what the speaker sounds like. "Following" errors are usually
caused by phenomena such as cone break-up which should not be a
problem with modern materials and computer-aided design.

d

[William Sommerwerck]

"Don Pearce" wrote in message
...
On Tue, 16 Jul 2013 11:08:06 -0700, "William Sommerwerck"
wrote:
"Peter Larsen" wrote in message
k...
William Sommerwerck wrote:

It's a speaker whose driver motion accurately follows the input
signal. Its radiation pattern is a secondary consideration.

No, as Harwood showed the directivity index has to be as close
to constant as possible and steps in it are a known cause of coloration.

If the radiating surface doesn't accurately follow the input, it doesn't
matter what the directivity index is.

And conversely if the directivity index is poor, it doesn't matter how
accurately the radiating surface follows the input. The radiation
pattern is a long way from secondary in speaker design. It almost
defines what the speaker sounds like. "Following" errors are usually
caused by phenomena such as cone break-up which should not be a
problem with modern materials and computer-aided design.

This is absolutely untrue. The belief that radiation pattern (and total
radiated power) are the principal components of a speaker's subjective quality
is demonstrably untrue.

Get a pair of STAX earspeakers and a pair of really good dynamic headphones
(such as Sennheiser). Play some good recordings. Which reproduces the
recording more faithfully?

Case closed.

[Don Pearce[_3_]]

On Tue, 16 Jul 2013 13:05:35 -0700, "William Sommerwerck"
wrote:

"Don Pearce" wrote in message
...
On Tue, 16 Jul 2013 11:08:06 -0700, "William Sommerwerck"
wrote:
"Peter Larsen" wrote in message
. dk...
William Sommerwerck wrote:

It's a speaker whose driver motion accurately follows the input
signal. Its radiation pattern is a secondary consideration.

No, as Harwood showed the directivity index has to be as close
to constant as possible and steps in it are a known cause of coloration.

If the radiating surface doesn't accurately follow the input, it doesn't
matter what the directivity index is.

And conversely if the directivity index is poor, it doesn't matter how
accurately the radiating surface follows the input. The radiation
pattern is a long way from secondary in speaker design. It almost
defines what the speaker sounds like. "Following" errors are usually
caused by phenomena such as cone break-up which should not be a
problem with modern materials and computer-aided design.

This is absolutely untrue. The belief that radiation pattern (and total
radiated power) are the principal components of a speaker's subjective quality
is demonstrably untrue.

Get a pair of STAX earspeakers and a pair of really good dynamic headphones
(such as Sennheiser). Play some good recordings. Which reproduces the
recording more faithfully?

Case closed.

I have Stax earspeakers and some Sennheiser phones. What I don't have
is a clue what you are talking about. The analogy escapes me entirely.

d

[William Sommerwerck]

"Don Pearce" wrote in message
...
On Tue, 16 Jul 2013 13:05:35 -0700, "William Sommerwerck"
wrote:

This is absolutely untrue. The belief that radiation pattern (and total
radiated power) are the principal components of a speaker's subjective
quality is demonstrably untrue.

Get a pair of STAX earspeakers and a pair of really good dynamic
headphones (such as Sennheiser). Play some good recordings.
Which reproduces the recording more faithfully?

I have Stax earspeakers and some Sennheiser phones. What I don't have
is a clue what you are talking about. The analogy escapes me entirely.

What you are hearing is more or less the direct output of the drivers, without
the influence of room acoustics or radiation pattern.

A good listening room is on the dead side of "neutral". Most of the sound we
hear is the direct sound from the driver(s). Are we to believe that a dynamic
system with broad dispersion will sound better than a system with superior
electrostatic drivers -- even if they don't have as broad dispersion?

[geoff]

"Gary Eickmeier" wrote in message
...

As I have mentioned, the final result is a combination of the recorded
acoustic and the playback acoustic, hopefully more of the former in a
normal
home environment.

It just the bit where it somehow sounds good thru a bunch of extremely
****ty little drivers powered via a sledgehammer of an equaliser, that gets
me....

geoff