ADD or AAD vs. DDD
JTT
actually prefer recordings (particularly of piano) that were recorded in the
60 or 70's via analog equipment, then translated to the digital format. I
know this doesn't make a lot of sense because it ends up digital anyway, and
if I were hearing something in digital that bothered me I would think that 1
transfer would do it. However, when I hear digitally recorded piano there is
almost an emptiness or lack of warmth. Maybe a lack of real presence? I
should break out my turntable, but short of that, I was wondering if anyone
else has noticed this?
Stephen McElroy
<jamest...@hotmail.com> wrote:
Yep, but it varies so much from recording to recording and transfer to
transfer that I would hesitate to make a sweeping statement about inherent
sound quality.
I like my records and turntable, but some people are so sensitive to
minute pitch variations that cd (or digital) is their best bet.
Do you have any favorite analog recordings soundwise? For me, on this
subject, it's interesting to compare Ashkenazy analog to Ashkenazy digital
from Decca.
Stephen
David Wake
> I just wanted to ask if others have had this sensation. I am starting to
> actually prefer recordings (particularly of piano) that were recorded in the
> 60 or 70's via analog equipment, then translated to the digital format. I
> know this doesn't make a lot of sense because it ends up digital anyway, and
> if I were hearing something in digital that bothered me I would think that 1
> transfer would do it.
>
It may make sense: with recording being as much art as science, it may
well be that some engineers have still not completely mastered the art
of recording pianos digitally (microphone placement etc.). If the
original analog sounds superior, it should still do so after a digital
remastering.
David (who feels somewhat the same way)
vaneyes
Early 1980's was the worst period for harsh-sounding DDD CDs. You shouldn't be
experiencing any widespread problems on majors or quality indies any longer.
Perhaps the vinyl/analogue propaganda has you gun-shy?
Of more recent CD vintage, IMO Naxos has been the worst culprit, however in the
last couple of years they've vastly improved...witness last year's Szymanowski
Piano Works Volume 3, and Shostakovich Preludes & Fugues.
Regards
David Wake
>
> Do you have any favorite analog recordings soundwise? For me, on this
> subject, it's interesting to compare Ashkenazy analog to Ashkenazy digital
> from Decca.
>
David
David M. Cook
Analog recordings do seem to have more presence and warmth. I particularly
like EMI recordings from the late 50s and early 60s.
However, with the Arcam Alpha 9 CD player that I've had for the last year,
digital recordings sound a lot better. I used to dislike Chandos, for
instance, but with this new player these are some of my favorite recordings
qua recordings, even some of the early ones like Järvi's Rimsky overture
set, which sounds lovely on the new player.
Dave Cook
Simon Roberts
"JTT" <jamest...@hotmail.com> wrote in message
news:QC0L6.56166$Ce4.4...@news1.rdc1.ct.home.com...
There are good and bad digital recordings, and good and bad
analogue recordings. It so happens that many superb-sounding
recordings were made in the 60s and 70s (especially the 60s), and
if you happen to catch good CD transfers of those, and only
inferior digital recordings, you might well reach that
conclusion. But you might revise your opinion a bit if you
compared the best digital recordings with some of the bad
analogue recordings made in the 60s and 70s....
Simon
jt
If you prefer to be modern, perhaps SACD is the way to go.
For me, 16-bit digital is unacceptable, but you should listen and make up
your own mind. It's best to do this in a home setting over a long period.
-jt
JTT
I have many. The Alkan by Ronald Smith recorded in 77 comes to mind. I have
recently really enjoyed the Rubinstein Chopin in GP Ar III. No doubt, there
were variable recordings from those years and I have heard some bad ones.
it's just when they are good, they sound really good to me.
James
JTT
"David M. Cook" <rh2...@mindspring.com> wrote in message
news:slrn9fp8b7...@localhost.localdomain...
So you really noticed a difference with the player? I imagine better
speakers must play a role as well and maybe my trusty old boxes were just
made for analog.
James
Paul Kintzele
"David M. Cook" wrote:
>
> I used to dislike Chandos, for
> instance, but with this new player these are some of my favorite recordings
> qua recordings, even some of the early ones like Järvi's Rimsky overture
> set, which sounds lovely on the new player.
I've recently had this exact same change of mind about Chandos. I
didn't understand the fuss about Jansons' Tchaikovsky 5, for example,
until I upgraded my amp and speakers. Now that my system can handle
recordings with a wider dynamic range and larger soundstage, this
Tchaikovsky 5 sounds great. I almost actually like the symphony now....
Paul
JTT
"Paul Kintzele" <kint...@english.upenn.edu> wrote in message
news:3AFCBF4D...@english.upenn.edu...
Akira Allen
not digitally remastered. They can sound a bit hard to my ears.
--
Alex
alex....@NOSPAMbradford.gov.uk
JTT <jamest...@hotmail.com> wrote in message
news:QC0L6.56166$Ce4.4...@news1.rdc1.ct.home.com...
Simon Roberts
I
> didn't understand the fuss about Jansons' Tchaikovsky 5, for
example,
> until I upgraded my amp and speakers. Now that my system can
handle
> recordings with a wider dynamic range and larger soundstage,
this
> Tchaikovsky 5 sounds great. I almost actually like the
symphony now....
You might just plain like it if you had a better performance than
Jansons'....
Simon
Frank Galvin
In article <QC0L6.56166$Ce4.4...@news1.rdc1.ct.home.com>,
"JTT" <jamest...@hotmail.com> wrote:
> I just wanted to ask if others have had this sensation. I am starting
> to actually prefer recordings (particularly of piano) that were
>recorded in 60 or 70's via analog equipment, then translated to the
>digital format [...] I know this doesn't make a lot of sense because
> it ends up digital anyway[.] I was wondering if anyone else has
>noticed this?
JTT:
Perhaps you missed this exchange from a few days ago, but then,
so did Matty...
In article <9d23pl$35g$1...@news.ox.ac.uk>, "Matthew Silverstein"
<matthew.s...@corpus-christi.oxford.ac.uk> wrote:
>
> The Muti/Philadelphia Beethoven cycle on EMI sounds wretched. Whenever
> I listen to it, I feel the need to take the wet cotton out of my ears.
> There is no definition in the bass at all, just a general (and
>generally inaudible) rumbling mush, and the horns and timpani often
>sound as though they are off stage. Occasionally they break through the
>fog and sound out gloriously, but only occasionally. It's quite a shame.
> While I doubt this would become a favorite of mine, I think that there
>is quite a bit I would really love were I able to hear it . . .
> Matty
[from M's description, I assumed that the Muti was a DDD and
replied]
Matty:
It is (and has been for about 15 years) my contention that *all*
DDDs are out of phase [*not* ADD or AAD or any permutation that
has A as the first letter].
Listen up, now: By *turning off your amp* then reversing the
leads to ONE of your speakers you can correct this anomaly. Then
turn your amp back on. And don't forget: Reverse this procedure
when you go back to analog [in-phase] recordings (do you?). NEVER
do any lead-switching with your amp *on*.
My report? I find that: 1) the 'hole in the middle' stereo effect
is corrected, 2) instrumental localization is restored,
3) frequency response is improved and 4) most of the glare is
removed.
I've actually rigged a DPDT switch at my speaker and one at the
amplifier taps for my Stax earspeakers so I can
easily change polarity [*pace* signal path purists, the gains, to
me, far outnumber any loss]. And I once saw a preamp that had a
built-in polarity switch. (Why would that be?)
Any time I talk about this I feel like some mumbling obsessive
crying out in the wilderness, and because of the possibility of
blowing fuses, etc. am loath to recommend it to anyone. But
RCMR-ers, breed apart as they are, might have no fear[...].
[and Frank, mumbling on, still feeling like an obssessive, and
itching to use the word "whilst" adds, on 12 May]:
Essentially, you want both speakers zigging together, not, as
with an out-of-phase signal, one zigging whilst the other zags.
In deep analog sympathy,
Frank
JTT
"Frank Galvin" <fga...@mediaone.net> wrote in message
news:pN8L6.30076$V%6.82...@typhoon.ne.mediaone.net...
Actually, this would explain why I do NOT have the problem with ADD or AAD.
And the description:
>My report? I find that: 1) the 'hole in the middle' stereo effect
is corrected, 2) instrumental localization is restored,
3) frequency response is improved and 4) most of the glare is
>removed.
really describes what I hear. There is s hole in the middle and I do not
get a sense that the instrument actually exists in a certain location. I
turn up, I turn down when it blasts, all the things that I never do with A**
recordings.
But WHY would they engineer digital recordings to be out of phase? Has
everyone grown so deaf it doesn't make a difference? Anyway, I did miss
this and thanks very much for the info.... interesting stuff.
James
Matthew B. Tepper
<9dirss$fvo$1...@newsreaderm1.core.theplanet.net>:
>If you like AAD piano sound, quite a few of Pollini's DG CDs are AAD,
>i.e. not digitally remastered. They can sound a bit hard to my ears.
Nonsense. If they are on CD, they have been digitally remastered, by
definition. What you mean is that they were not digitally mixed or edited.
--
Matthew B. Tepper: WWW, science fiction, classical music, ducks!
My personal home page -- http://home.earthlink.net/~oy/index.html
My main music page --- http://home.earthlink.net/~oy/berlioz.html
To write to me, do for my address what Androcles did for the lion
Top 3 worst UK exports: Mad-cow; Hoof-and-mouth; Charlotte Church
JTT
"Matthew B. Tepper" <oy兀earthlink.net> wrote in message
news:9djjk...@news2.newsguy.com...
> alex....@bradford.gov.uk (Akira Allen) wrote in
> <9dirss$fvo$1...@newsreaderm1.core.theplanet.net>:
>
> >If you like AAD piano sound, quite a few of Pollini's DG CDs are AAD,
> >i.e. not digitally remastered. They can sound a bit hard to my ears.
>
> Nonsense. If they are on CD, they have been digitally remastered, by
> definition. What you mean is that they were not digitally mixed or
edited.
>
> --
I understand that the middle letter represents the remastering process, i.e.
tonally readjusting and adjusting volume and eq etc. on the recording tracks
before transfer to the final digital format. If this is a direct analog to
digital copy, or analog equipment was used to master the tracks, it is AAD.
I am not expert on this but that was my understanding, any experts on this
out there??
James
Frank Galvin
----------
In article <9djjk...@news2.newsguy.com>, oy兀earthlink.net
(Matthew B. Tepper) wrote:
> alex....@bradford.gov.uk (Akira Allen) wrote in
> <9dirss$fvo$1...@newsreaderm1.core.theplanet.net>:
>
>>If you like AAD piano sound, quite a few of Pollini's DG CDs are AAD,
>>i.e. not digitally remastered. They can sound a bit hard to my ears.
>
> Nonsense [MBT replies with unusual diplomacy]. If they are on CD,
> they have been digitally remastered, by definition. What you mean is
> that they were not digitally mixed or edited.
And not digitally *recorded*, which is where I think the big
problem is. See my post above. Would love to get some discussion
going on this; mumble, mumble, mumble...
Cheers (in spite of DDD),
Frank
Chel van Gennip
I understand that the middle letter represents the remastering process, i.e.
> tonally readjusting and adjusting volume and eq etc. on the recording tracks
> before transfer to the final digital format. If this is a direct analog to
> digital copy, or analog equipment was used to master the tracks, it is AAD.
> I am not expert on this but that was my understanding, any experts on this
> out there??
That is correct. The whole problem with this discussion is that some of
the factors are totally undefined. The terms AAD, ADD and DDD were defined
about 25 years ago. Since then the meaning (or implementation) of the first
2 D's has changed a lot. Current digital techniques are almost perfect,
early digital techniques were not. Furthermore with digital sound a
number of new sound processing techniques were introduced, it is
questionable if one should "improve" a good recording of a good
performance.
A related factor could be that experienced sound technicians sticked
to the analogue equipment they knew for a long time, while young
inexperienced sound technicians switched to digital more rapidly.
Chel van Gennip
Listen to http://www.mp3.com/SergvanGennip
vaneyes
Another thought. Many times people's speaker cable and connections have
gradually deteriorated to such a degree, that they think they need new
components. The less expensive solution should be considered first.
Regards
vaneyes
JTT
"Chel van Gennip" <ch...@vangennip.nl> wrote in message
news:3AFD6D8C...@vangennip.nl...
If current digital techniques are perfect, how does that explain that I can
listen to recordings from 99/00/01 and still come away with that same
feeling I originally described? The lack of focus and warmth. I once heard
it digital music was looking at a beautiful landscape through a screen. For
some reason the piano brings that aspect out for me as I am starting to get
a sense of what that means. The sound is clean, hiss free, unwavering, yet
free of that essence that pulls you in, a central solid body of sound. It
is heard to explain and it could very well be helped by my equipment
upgrade. The out of phase argument perfectly described what I hear and I was
amazed by that revelation, if it is true!
Paul Kintzele
Well, I do have Markevitch/Philips and Svetlanov/Melodiya, but as I
recall, neither made a overwhelming impression on me, although I haven't
done any comparative listening since improving my gear.
Paul
David Wake
I've only heard the old DG full-price CD which had a romantic flowery
picture of a young woman. Haven't heard the DG Originals. Anyone
know whether there's a difference?
David
Matthew Silverstein
> Well, I do have Markevitch/Philips and Svetlanov/Melodiya, but as I
> recall, neither made a overwhelming impression on me, although I haven't
> done any comparative listening since improving my gear.
Try Gergiev/VPO on Philips. Stunning.
Matty
David M. Cook
>So you really noticed a difference with the player?
I upgraded my whole system piecemeal, but the CD player upgrade seemed to
make the biggest difference, the amp upgrade (to a Bryston B-60) a little
more. The evaluation was made by the usual "switch it out and listen"
method, which is probably not very reliable. I was also going by the
glowing reviews that the Arcam got everywhere.
I do feel that I jumped too quickly to the Arcam after auditioning a Rotel.
I wish I'd auditioned more equipment at an intermediate price point. At
least I didn't spend $20k on a Linn.
I do believe that D/A converters have improved greatly in recent years. That
Rotel sounded a lot better than my old Sony from the early 90s. It's
probably not necessary to spend very much to get the benefit of the newer
chips.
> I imagine better
>speakers must play a role as well and maybe my trusty old boxes were just
>made for analog.
I upgraded from B&W 603s to Soliloquy 5.3s a few months after the CD player
upgrade. The new speakers do handle seem to handle a wider dynamic range
without getting boxy sounding. That may have a lot to do with my greater
enjoyment of DDD recordings.
Dave Cook
David M. Cook
>Listen up, now: By *turning off your amp* then reversing the
>leads to ONE of your speakers you can correct this anomaly. Then
>turn your amp back on. And don't forget: Reverse this procedure
>when you go back to analog [in-phase] recordings (do you?). NEVER
>do any lead-switching with your amp *on*.
>
>My report? I find that: 1) the 'hole in the middle' stereo effect
>is corrected,
Er, yeah, because your speakers are now wired out of phase, so there is no
stereo effect.
2) instrumental localization is restored,
Rather dubious if your speakers are wired out of phase.
I don't doubt that this can sound more pleasant, though.
Dave Cook
Frank Galvin
In article <jWgL6.56992$Ce4.5...@news1.rdc1.ct.home.com>,
"JTT" <jamest...@hotmail.com> wrote:
> If current digital techniques are perfect, how does that explain that I
> can listen to recordings from 99/00/01 and still come away with
> that same feeling I originally described? The lack of focus and warmth.
> [...] The sound is clean, hiss free, unwavering, yet free of that
> essence that pulls you in, a central solid body of sound.
The heart and essence of the matter. A vague feeling of
dissatisfaction; nothing you can quite put your finger on; oh,
all very 'crisp' and so on, sure. But lacking...what?
> I once heard digital music was looking at a beautiful landscape through
> a screen.
Funny, I remember in the first years of digital, when the
recordings were released on LP, _Gramophone_ reviewers
inevitably would say, 'Sounds like a veil has been removed
between the listener & the music.' *Moi* used to inevitably say,
'What the ---- are they talking about?'
>The out of phase argument perfectly described what I hear and I
> was amazed by that revelation, if it is true!
Amazed because you and I both realize where this line of thinking
would eventually lead....
[JTT's post was in response to Chel van Gennip, who also wrote]:
>> A related factor could be that experienced sound technicians sticked
>> to the analogue equipment they knew for a long time, while young
>> inexperienced sound technicians switched to digital more rapidly.
True, but I think that might be changing. There are an awful lot
of very well engineered DDDs out there [Albany Records Troy 416,
with the Long Beach SO conducted by JoAnn Falletta, is a recent
and spectacular example]. Problem is, it's out of phase....
Best,
Frank
Tang Huyen
Frank Galvin wrote: <<It is (and has been for about 15 years) my contention
that *all* DDDs are out of phase [*not* ADD or AAD or any permutation that
has A as the first letter].
Listen up, now: By *turning off your amp* then reversing the leads to ONE of
your speakers you can correct this anomaly. Then turn your amp back on. And
don't forget: Reverse this procedure when you go back to analog [in-phase]
recordings (do you?). NEVER do any lead-switching with your amp *on*.
My report? I find that: 1) the 'hole in the middle' stereo effect is
corrected, 2) instrumental localization is restored, 3) frequency response is
improved and 4) most of the glare is removed.
I've actually rigged a DPDT switch at my speaker and one at the amplifier
taps for my Stax earspeakers so I can easily change polarity [*pace* signal
path purists, the gains, to me, far outnumber any loss]. And I once saw a
preamp that had a built-in polarity switch. (Why would that be?) [snip]
Essentially, you want both speakers zigging together, not, as with an
out-of-phase signal, one zigging whilst the other zags.
In deep analog sympathy,
Frank>>
What you are talking about here is relative phase, and incorrect relative
phase, one speaker zigs while the other zags and they are out of phase with
each other, and you are attributing it to all DDDs.
This relative phase is different from absolute phase, and in incorrect
absolute phase both speakers zig when they should zag and zag when they
should zig, but otherwise are in relative phase with each other, while in
correct absolute phase they both zig when they should zig and zag when they
should zag, and of course they are in relative phase with each other. My
friend Clark Johnsen (Listening Studio, Boston) wrote a smal book on absolute
phase, the Wood Effect.
If the two sides of DDDs are out of phase with each other, as you say, and
you reverse the leads to ONE of your speakers to correct the relative phase,
then either the result is correct absolute phase (in which case you don't
have to do anything further) or incorrect absolute phase, and in the latter
case you still have a problem of hollow and muffled sound, the hole in the
middle of the stereo image, etc., in which case you have to reverse the leads
on BOTH speakers from what they presently are (after your initial reversal of
one side only). In effect, you switch the initially reversed leads back to
where they were and then reverse the leads on the OTHER speaker.
By the way, you ought to post your post to rec.audio.high-end also.
Tang Huyen
Frank Galvin
> On Sat, 12 May 2001 11:04:53 GMT, Frank Galvin <fga...@mediaone.net>
> wrote:
>>[by switching speaker phase on digital recordings] I find that: 1) the
'hole in the middle' stereo effect
>>is corrected [...]
In article <slrn9frt7f...@localhost.localdomain>,
rh2...@mindspring.com (David M. Cook) replied:
> Er, yeah, because your speakers are now wired out of phase, so there is
> no
> stereo effect.
s'not true...
> 2) instrumental localization is restored,
>
> Rather dubious if your speakers are wired out of phase.
>
> I don't doubt that this can sound more pleasant, though.
>
> Dave Cook
Dave:
[...mumble, mumble] uh, I do this only with DDD recordings,
which, if I'm right, are out of phase to begin with. Again, the
trick is to get the speaker cones zigging together [in phase]
rather than [say] left zigging; right zagging [out of phase]. I
think the original phase problem is brought about by something
inherent in the *digital* recording process - what that
'something' is, I leave to the engineers. [Any engineers out
there??]
Your last point is well taken as it goes to subjectivity, a very
large part of the equation. What really interests me here is that
two people (JTT and the gentleman who wrote about dissatisfaction
with a Muti Beethoven set, now in Newsgroup Valhalla, I guess)
sounded to me as if they were addressing the very things I've
been trying to correct. Interesting?
Like to hear more from you on this. You stopped at # 2 in the
list of improvements - out of exasperation, or do you think 3 & 4
are valid?
Best,
Frank
>
>
>
Norman M. Schwartz
>From: Tang Huyen tang_...@yahoo.com
>Date: 5/13/01 7:43 AM Eastern Daylight Time
>Message-id: <3AFE736D...@yahoo.com>
If *any* recording (LP, CD, tape, etc.) is recorded out of ABSOLUTE PHASE, and
played back through a pre-amp having a mono setting, which is engaged, the
signal heard through the loudspeakers is reduced to ZERO or very close to it.
Neither ADD, AAD or DDD CDs are universally recorded out of absolute phase!
(Additionally, stereo signals which are recorded out of absolute phase are very
easily observed as being so on a 'scope.)
Tony Movshon
> It is (and has been for about 15 years) my contention that *all*
> DDDs are out of phase [*not* ADD or AAD or any permutation that
> has A as the first letter].
>
> Listen up, now: By *turning off your amp* then reversing the
> leads to ONE of your speakers you can correct this anomaly. Then
> turn your amp back on. And don't forget: Reverse this procedure
> when you go back to analog [in-phase] recordings (do you?). NEVER
> do any lead-switching with your amp *on*.
>
> My report? I find that: 1) the 'hole in the middle' stereo effect
> is corrected, 2) instrumental localization is restored,
> 3) frequency response is improved and 4) most of the glare is
> removed.
It is easy to verify that all commercial stereo recordings, analog or
digital, new or old, have correct relative phase between the channels
as recorded. The idea that DDD recordings are out of phase is nonsense.
If you find listening to recordings with the phase reversed more
pleasant, that suggests to me that there is something fundamentally
wrong with your system or your listening room. It would be interesting
to know whether you have the same preference when listening through
good headphones. The subjective impression most people get from out
of phase reproduction is of an odd, echoey, poorly localized sound
source without spatial definition. It's conceivable, of course, that
you might prefer that. It's inexplicable that your preference would
depend on whether the recording was analog or digital.
The question of relative phase is entirely different from the issue
of *absolute* phase. The absolute phase of reproduction determines
whether compression or rarefaction signals received at the microphone
are reproduced the "right" way round. This is a nonissue for human
listeners, since it has been repeatedly established that the human
ear is entirely insensitive to absolute phase. Claims to the contrary
which pop up from time to time have been rather securely shown to
depend on errors of reproduction or errors in test procedure.
--
Tony Movshon
mov...@nyu.edu
David M. Cook
>[...mumble, mumble] uh, I do this only with DDD recordings,
>which, if I'm right, are out of phase to begin with.
I think you're confusing absolute polarity with electrical phase. To change
the absolute polarity, you'd have to reverse the leads on *both* speakers.
Unfortunately, you can't determine ahead of time whether a recording is in
absolute polarity or not. Some pre-amps have a polarity switch.
The idea that all DDD recordings have their channels 180 degrees out of
phase is riduculous.
Dave Cook
Tony Movshon
> By the way, you ought to post your post to rec.audio.high-end also.
There's no need to be nasty.
--
Tony Movshon
mov...@nyu.edu
Kalman Rubinson
>If *any* recording (LP, CD, tape, etc.) is recorded out of ABSOLUTE PHASE, and
>played back through a pre-amp having a mono setting, which is engaged, the
>signal heard through the loudspeakers is reduced to ZERO or very close to it.
You have it wrong, I am afraid. What you are describing is when there
is an error in relative phase, that is, one channel relative to the
other. Absolute phase is, as Tony Movshon has explained, inaudible.
>Neither ADD, AAD or DDD CDs are universally recorded out of absolute phase!
Right.
>(Additionally, stereo signals which are recorded out of absolute phase are very
>easily observed as being so on a 'scope.)
Again, you are referring to relative phase.
Kal
Stephen W. Worth
(Norman M. Schwartz) wrote:
> (Additionally, stereo signals which are recorded out of absolute
> phase are very easily observed as being so on a 'scope.)
It's also easy to spot by ripping the CD and looking at the
waveform in an audio editing program. I've ripped lots of
classical CDs and they aren't out of phase.
There is an interesting rear channel theory called the
"Haffler-Doss Sum and Difference Matrix System". You take
the positive leads from the left and right speakers and
attach them to the positive leads of two rear channel
speakers. Then you connect the negatives of both speakers
to each other. This throws any sound not common to both
speakers to the rear. It also seems to throw out of
phase sound to the rear channel. I have it like this on
my TV stereo. It sounds pretty good... a poor man's surround
sound.
See ya
Steve
--
Visit Spumco's Wonderful World of Cartoons:
http://www.spumco.com alt.animation.spumco
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Learn about animation art (without going BROKE!)
Vintage Ink & Paint http://www.vintageip.com
David M. Cook
>My report? I find that: 1) the 'hole in the middle' stereo effect
>is corrected, 2) instrumental localization is restored,
>3) frequency response is improved and 4) most of the glare is
>removed.
The problem might be that the DDD recordings have a much wider stereo
separation, leading to the "hole in the middle." You might try one or all of
the following:
* toe the speakers in
* move your listening position further away
* move the speakers closer together
Dave Cook
Frank Galvin
----------
In article <3AFEB347...@nyu.edu>, Tony Movshon
<mov...@nyu.edu> wrote:
> Frank Galvin wrote:
>> It is (and has been for about 15 years) my contention that *all*
>> DDDs are out of phase [*not* ADD or AAD or any permutation that
>> has A as the first letter].
> It is easy to verify that all commercial stereo recordings, analog or
> digital, new or old, have correct relative phase between the channels
> as recorded. The idea that DDD recordings are out of phase is nonsense.
Tony:
Verify how [serious question]?
> [...]It would be interesting
> to know whether you have the same preference when listening through
> good headphones.
Stax electrostatics that *can* be phase inverted [I'd be out of
luck with regular 'phones, wouldn't I?] and the difference is
even *more* evident, as are so many other things with 'phones.
> The question of relative phase is entirely different from the issue
> of *absolute* phase. The absolute phase of reproduction [...]
> is a nonissue for human
> listeners, since it has been repeatedly established that the human
> ear is entirely insensitive to absolute phase. Claims to the contrary
> [...] have been rather securely shown to
> depend on errors of reproduction or errors in test procedure.
Have always thought of myself, although definitely human, as
something of a slight error of reproduction, so maybe there's the
rub!
/;-->)
Best,
Frank
P.S. And I do like that '*rather* securely' [emphasis mine]...
Frank Galvin
[David M. Cook wrote]:
> On Sun, 13 May 2001 12:28:06 GMT, Frank Galvin <fga...@mediaone.net>
> wrote:
[in a pit of fique, obviously]
>>[...mumble, mumble] uh, I do this only with DDD recordings,
>>which, if I'm right, are out of phase to begin with.
and Dave Cook replied:
> I think you're confusing absolute polarity with electrical phase. [...]
> Some pre-amps have a polarity switch.
Why?
> The idea that all DDD recordings have their channels 180 degrees out of
> phase is riduculous.
I'm willing to be convinced...
[Later, Dave also wrote]:
>The problem might be that the DDD recordings have a much wider stereo
> separation, leading to the "hole in the middle." You might try one or
> all of the following:
> * toe the speakers in
> * move your listening position further away
> * move the speakers closer together
None of which I can do with my Stax electrostatic phones, where I
claim I can *hear* the improvements best - obviously.
My ADS L-810s [from 1978; no laughing allowed, now] are toed in,
about 5 feet apart, and sound terrific - no holes; instrument
location, depth of soundstage imaging, bass response etc. fine -
with analog recordings. Hole-y, diffused imaging, thin bass, and
glare is what I get with DDDs, *until* I do you-know-what. All
other things are equal.
Cheers & thanks,
Frank
P.S. [fyi] I use the old Shure 'An Audio Obstacle Course' record
to check my speaker phasing. Primitive and amateur, perhaps, but
it do work.
F.
Tony Movshon
> > It is easy to verify that all commercial stereo recordings, analog or
> > digital, new or old, have correct relative phase between the channels
> > as recorded. The idea that DDD recordings are out of phase is nonsense.
>
> Tony:
>
> Verify how [serious question]?
Any of a number of ways. The traditional control room method would be to
put the L signal into the vertical deflection of a CRT and the R into the
horizontal. In-phase signals show up as lines with a positive slope.
Simpler in this digital era is to do what another poster suggested: rip
the track of your choice into a computer, and then use any available WAV
editor to view the waveforms. Expand the time step until you can see each
waveform cycle, and look to see whether the waveforms go up and down
together in each channel (in phase), or oppose each other (out of phase).
Another approach is to use the computer to calculate the L+R and L-R
signals from the given L and R inputs -- for in-phase L and R, the L+R
signal is always larger than the L-R signal (typically by 6-12 dB).
Interestingly, the relative strength of the L-R signal is a good
indication of the spatial separation you will hear in a recording. If
you have a fast computer, a wave editor, and a burner, you can easily
modify the effective channel separation by computing L+R and L-R,
attenuating L-R, and then recombining the signals into L and R. This
can also be easily done in the analog domain with a couple of op-amps.
Which, by the way, leads me to wonder why in the early days of stereo
the engineers didn't choose to encode stereo as L+R/L-R. This would
make it easy to adjust stereo imaging with an L-R gain control, and
would have guaranteed a clean mono signal from the L+R channel. The
L-R could have been encoded on totally mono-compatible LPs by using
horizontal movements for L+R and vertical ones for L-R, and there
would have been no need to press stereo and mono versions of the same
LP.
> > The question of relative phase is entirely different from the issue
> > of *absolute* phase. The absolute phase of reproduction [...]
> > is a nonissue for human
> > listeners, since it has been repeatedly established that the human
> > ear is entirely insensitive to absolute phase. Claims to the contrary
> > [...] have been rather securely shown to
> > depend on errors of reproduction or errors in test procedure.
> ...
> P.S. And I do like that '*rather* securely' [emphasis mine]...
A scientist's caution. It's impossible to prove the null hypothesis,
even when you try to do it yourself.
I've made up some demonstration material to show the inaudibility of
absolute phase, and to show that "absolute phase" becomes audible in
the presence of rather small amounts of odd-order distortion. I can
put the WAV files on the Web if anyone's interested.
--
Tony Movshon
mov...@nyu.edu
Frank Galvin
In article <bigshot-1405...@pm03-29.ktb.net>,
> In article <20010513083940...@ng-ch1.aol.com>, nm...@aol.com
> (Norman M. Schwartz) wrote:
>
>> (Additionally, stereo signals which are recorded out of absolute
>> phase are very easily observed as being so on a 'scope.)
>
> It's also easy to spot by ripping the CD and looking at the
> waveform in an audio editing program. I've ripped lots of
> classical CDs and they aren't out of phase.
(Bad boy, the RIAA'll be coming for you if you don't watch out!!)
/;-->)
> There is an interesting rear channel theory called the
> "Haffler-Doss Sum and Difference Matrix System". You take
> the positive leads from the left and right speakers and
> attach them to the positive leads of two rear channel
> speakers. Then you connect the negatives of both speakers
> to each other. This throws any sound not common to both
> speakers to the rear. It also seems to throw out of
> phase sound to the rear channel. I have it like this on
> my TV stereo. It sounds pretty good... a poor man's surround
> sound.
> See ya
> Steve
Yes, and my new Sony DVD player does an awfully good job of
*deriving* a 5.1-like effect from my miserable little Pro-Logic
set-up (no doubt by a Haffler-Doss circuit?). Trouble is, all the
movies su--. But that's another story and another newsgroup, god
help them if I ever find time to lurk 'n' leap.
Apologies to all for my misuse of terms. So now I gotta re-state:
IM[ns]HO all DDDs ('scopes and ripping info notwithstanding) are
released having incorrect relative phase, which causes this
person's speaker cones to make excursions independent of one
another, not together, thus causing him much aural fatigue when
he plays digitally *recorded* music.
If anyone can think of a better way to say this, please feel free
to re-ph[r]ase. (Or am I just asking for it???)
Best to all,
Frank
P.S. Oh yes, sorry, forgot - uh, reversing both sets of leads
gets us back to square one, doesn't it?
Kalman Rubinson
wrote:
>Which, by the way, leads me to wonder why in the early days of stereo
>the engineers didn't choose to encode stereo as L+R/L-R. This would
>make it easy to adjust stereo imaging with an L-R gain control, and
>would have guaranteed a clean mono signal from the L+R channel.
In fact, there were stereo discs cut this way when stereo was
experimental. Some of these were demonstrated at shows.
Kal
Kalman Rubinson
<fga...@mediaone.net> wrote:
>IM[ns]HO all DDDs ('scopes and ripping info notwithstanding) are
>released having incorrect relative phase, which causes this
>person's speaker cones to make excursions independent of one
>another, not together, thus causing him much aural fatigue when
>he plays digitally *recorded* music.
>
>If anyone can think of a better way to say this, please feel free
>to re-ph[r]ase. (Or am I just asking for it???)
Perfectly clear and perfectly untrue, of course.
Kal
Simon Roberts
"Tony Movshon" <mov...@nyu.edu> wrote in message
news:3AFFDB1D...@nyu.edu...
[snip]
> I've made up some demonstration material
[snip]
Now there's an engagingly ambiguous phrase....
Simon
Tony Movshon
Do not confuse my ambiguity with your own ambivalence.
--
Tony Movshon
mov...@nyu.edu
Stephen W. Worth
Galvin" <fga...@mediaone.net> wrote:
> IM[ns]HO all DDDs ('scopes and ripping info notwithstanding) are
> released having incorrect relative phase, which causes this
> person's speaker cones to make excursions independent of one
> another, not together, thus causing him much aural fatigue when
> he plays digitally *recorded* music.
It isn't the CDs. Perhaps your speakers' crossovers are
wired backwards. Or perhaps it's time for a new set of
speakers.
samir golescu
> It isn't the CDs. Perhaps your speakers' crossovers are
> wired backwards. Or perhaps it's time for a new set of
> speakers.
This is the nicest euphemism yet for the Korenian "it's time for a new
pair of ears"....
Simon Roberts
: > It isn't the CDs. Perhaps your speakers' crossovers are
It's about time you bought a second speaker so you can hear what these
gentlemen are talking about.
Simon (who doesn't understand a word of it but instinctively believes
Tony M.)
Alain Dagher
Simon Roberts wrote:
> Simon (who doesn't understand a word of it but instinctively believes
> Tony M.)
I also don't understand the whole phase reversal issue. And I got an A+ in
my acoustics course in electrical engineering too! But this never made sense
to me.
There was even a speaker in the 80s that was cone shaped and I think the
adverts suggested that this equalized the phase between low and high
frequencies. Perhaps Tony remembers this. I also didn't understand this.
Alain
Kalman Rubinson
<al...@bic.mni.mcgill.ca> wrote:
>There was even a speaker in the 80s that was cone shaped and I think the
>adverts suggested that this equalized the phase between low and high
>frequencies. Perhaps Tony remembers this. I also didn't understand this.
Aren't most speakers 'cone shaped?' Perhaps you are referring to the
Ohm speakers which used an elongated cone driver on top of the
enclosure, facing down into it.
Kal
Alain Dagher
Kalman Rubinson wrote:
Yes - that's the one.
Alain
Kalman Rubinson
<al...@bic.mni.mcgill.ca> wrote:
>
>
>Kalman Rubinson wrote:
>
>> On Mon, 14 May 2001 21:15:46 GMT, Alain Dagher
>> <al...@bic.mni.mcgill.ca> wrote:
>>
>> >There was even a speaker in the 80s that was cone shaped and I think the
>> >adverts suggested that this equalized the phase between low and high
>> >frequencies. Perhaps Tony remembers this. I also didn't understand this.
>>
>> Aren't most speakers 'cone shaped?' Perhaps you are referring to the
>> Ohm speakers which used an elongated cone driver on top of the
>> enclosure, facing down into it.
>
>Yes - that's the one.
Ugh. Regardless of what they claimed about phase coherance, they
sounded strange.
Kal
Tony Movshon
> Simon (who doesn't understand a word of it but instinctively believes
> Tony M.)
We scientists love it when lay people say that.
--
Tony Movshon
mov...@nyu.edu
Raymond Hall
news:utTL6.32310$V%6.93...@typhoon.ne.mediaone.net...
>
> Apologies to all for my misuse of terms. So now I gotta re-state:
>
> IM[ns]HO all DDDs ('scopes and ripping info notwithstanding) are
> released having incorrect relative phase, which causes this
> person's speaker cones to make excursions independent of one
> another, not together, thus causing him much aural fatigue when
> he plays digitally *recorded* music.
>
> If anyone can think of a better way to say this, please feel free
> to re-ph[r]ase. (Or am I just asking for it???)
The only explanation I can offer, is that your equipment is already "not in
relative phase" (it only requires one incorrect reversed connect somewhere),
and by reversing the leads to one speaker you are correcting the problem of
relative phase.
I've noticed no problem at all with any DDD CD I have purchased.
Regards,
# Classical Music WebSite Links :
http://www.users.bigpond.com/hallraylily/tassiedevil2.htm
# Main Page, To Conductors, Jazz Songstresses :
http://www.users.bigpond.com/hallraylily/index.html
# Ormandy page and New Furtwängler link
Ray, Sydney
Tony Movshon
> I also don't understand the whole phase reversal issue. And I got an A+ in
> my acoustics course in electrical engineering too! But this never made sense
> to me.
>
> There was even a speaker in the 80s that was cone shaped and I think the
> adverts suggested that this equalized the phase between low and high
> frequencies. Perhaps Tony remembers this. I also didn't understand this.
It was one of a collection of speakers that claimed "phase coherence".
The reasoning is that different drivers within a multi-way speaker
respond at slightly different times -- a woofer is typically more
massive and slower to respond than a tweeter. So the idea behind some
of these designs was to place the tweeters some way behind the woofer
so that the low frequencies would get a head start and everything would
arrive at your ear in phase. The coneheaded speaker was an extreme case
whose precise physical basis I now forget.
While this all sounds like a nice idea, the phase characteristics of
dynamic loudspeakers are such a mess that it was probably a doomed
effort to begin with, and no one seemed to like it much (vide Kal's
post).
Incidentally, this is really a third aspect of phase. So far we've
drifted through relative phase (i.e. the polarity of the speakers with
respect to each other), absolute phase (the polarity of the speakers
with respect to the original sound), and now phase coherence (the
preservaton of relative phase across frequencies within a channel).
The only one of these that is [clearly] audible is the first.
--
Tony Movshon
mov...@nyu.edu
Simon Roberts
> Simon Roberts wrote:
> > Simon (who doesn't understand a word of it but instinctively
believes
> > Tony M.)
>
> We scientists love it when lay people say that.
That's why we're here.
Simon
Alain Dagher
Tony Movshon wrote:
> So far we've
> drifted through relative phase (i.e. the polarity of the speakers with
> respect to each other), absolute phase (the polarity of the speakers
> with respect to the original sound), and now phase coherence (the
> preservaton of relative phase across frequencies within a channel).
> The only one of these that is [clearly] audible is the first.
I guess that's the part I am not sure I get. What is it that one hears, and how is
it detectable?
Alain
Kalman Rubinson
wrote:
>It was one of a collection of speakers that claimed "phase coherence".
>The reasoning is that different drivers within a multi-way speaker
>respond at slightly different times -- a woofer is typically more
>massive and slower to respond than a tweeter. So the idea behind some
>of these designs was to place the tweeters some way behind the woofer
>so that the low frequencies would get a head start and everything would
>arrive at your ear in phase.
Sorry, Tony, but the reason for the displacement has nothing to do
with the speed of response since any moving diaphragm reproducing
a particular frequency must move at the same speed, else it would
not the the same frequency. The reason for the displacement is to
compensate for the phase shift due to the fact that larger, deeper
cones have an effective source location further back from the front
panel than a smaller, shallower cone or dome. Displacing the latter
backward, places the source in the same radiating plane.
>Incidentally, this is really a third aspect of phase. So far we've
>drifted through relative phase (i.e. the polarity of the speakers with
>respect to each other), absolute phase (the polarity of the speakers
>with respect to the original sound), and now phase coherence (the
>preservaton of relative phase across frequencies within a channel).
>The only one of these that is [clearly] audible is the first.
Thanks so much for the [clearly].
Kal
Tony Movshon
> >of these designs was to place the tweeters some way behind the woofer
> >so that the low frequencies would get a head start and everything would
> >arrive at your ear in phase.
>
> Sorry, Tony, but the reason for the displacement has nothing to do
> with the speed of response since any moving diaphragm reproducing
> a particular frequency must move at the same speed, else it would
> not the the same frequency.
It's not the speed of the diaphragm that's the issue, it's the
phase delay or latency of response. Two drivers responding to the
same input frequency don't necessarily respond at the same time,
even if they move at the same velocity. That's phase lag.
> The reason for the displacement is to
> compensate for the phase shift due to the fact that larger, deeper
> cones have an effective source location further back from the front
> panel than a smaller, shallower cone or dome. Displacing the latter
> backward, places the source in the same radiating plane.
That's equivalent to what I was saying. The phase lag is what's
responsible for the "effective source location".
> >Incidentally, this is really a third aspect of phase. So far we've
> >drifted through relative phase (i.e. the polarity of the speakers with
> >respect to each other), absolute phase (the polarity of the speakers
> >with respect to the original sound), and now phase coherence (the
> >preservaton of relative phase across frequencies within a channel).
> >The only one of these that is [clearly] audible is the first.
>
> Thanks so much for the [clearly].
The evidence on absolute phase sensitivity is very weak. On deviations
from perfect phase coherence, the case is stronger but I think the
jury is still out.
--
Tony Movshon
mov...@nyu.edu
Tony Movshon
The approximate truth is that the ear is exquisitely sensitive to the
relative amounts of different frequency components in a sound, but very
insensitive to the phase of those components except when they are very
close together in frequency, within what's called a "critical band".
Our deafness to phase information cannot, however, be absolute, and
there's reason to believe that inaccuracies in the reproduction of
phase ("phase coherence") can make an audible difference to reproduced
sound under some conditions (hence the speaker design issues that you
and Kal discussed). The evidence on this is not decisive, and there
are other possible explanations for some of the phenomena that people
report, but there's also a general belief that phase-coherent sound
reproduction is a Good Thing, and it's hard to argue with that. In
practice, outside the domain of high-end electrostatic headphones, it's
almost impossible to achieve good phase coherence from real transducers.
One special case of phase perception is so-called "absolute phase", i.e.
whether humans can detect inversion of a waveform. This is the same
as shifting the phase of all components exactly 180 deg, which leaves
phase coherence unaffected. There is no credible evidence that absolute
phase is detectable.
--
Tony Movshon
mov...@nyu.edu
Kalman Rubinson
wrote:
>It's not the speed of the diaphragm that's the issue, it's the
>phase delay or latency of response. Two drivers responding to the
>same input frequency don't necessarily respond at the same time,
>even if they move at the same velocity. That's phase lag.
That would be phase lag but, in practice, it is due, at crossover
frequencies, to the relative physical position of the sources. To
a much smaller degree, it is also due to the relative inductances of
the drivers. However, I think you are suggesting something that sounds
like different response latencies. What causes that? Mass?
>> The reason for the displacement is to
>> compensate for the phase shift due to the fact that larger, deeper
>> cones have an effective source location further back from the front
>> panel than a smaller, shallower cone or dome. Displacing the latter
>> backward, places the source in the same radiating plane.
>
>That's equivalent to what I was saying. The phase lag is what's
>responsible for the "effective source location".
The other way. The phase lag is due to the differences in effective
source locations due to their construction and size. One simply moves
the drivers to compensate.
>The evidence on absolute phase sensitivity is very weak. On deviations
>from perfect phase coherence, the case is stronger but I think the
>jury is still out.
Yes. Most of the better-sounding designs have smooth phase shift
across the audible range rather than uniform phase response.
Kal
Tony Movshon
> On Mon, 14 May 2001 21:34:28 -0400, Tony Movshon <mov...@nyu.edu>
> wrote:
>
> >It's not the speed of the diaphragm that's the issue, it's the
> >phase delay or latency of response. Two drivers responding to the
> >same input frequency don't necessarily respond at the same time,
> >even if they move at the same velocity. That's phase lag.
>
> That would be phase lag but, in practice, it is due, at crossover
> frequencies, to the relative physical position of the sources. To
> a much smaller degree, it is also due to the relative inductances of
> the drivers. However, I think you are suggesting something that sounds
> like different response latencies. What causes that? Mass?
Among other things, yes. Capacitance can also be an issue with some
designs.
> >> The reason for the displacement is to
> >> compensate for the phase shift due to the fact that larger, deeper
> >> cones have an effective source location further back from the front
> >> panel than a smaller, shallower cone or dome. Displacing the latter
> >> backward, places the source in the same radiating plane.
> >
> >That's equivalent to what I was saying. The phase lag is what's
> >responsible for the "effective source location".
>
> The other way. The phase lag is due to the differences in effective
> source locations due to their construction and size. One simply moves
> the drivers to compensate.
"Effective source location" and "phase lag" are the same thing. In
other words, if the drivers were perfectly in phase, they would have
the same effective source location.
This seems to have drifted way off topic for rmcr. Apologies to those
who care about such things.
--
Tony Movshon
mov...@nyu.edu
Alain Dagher
> The approximate truth is that the ear is exquisitely sensitive to the
> relative amounts of different frequency components in a sound, but very
> insensitive to the phase of those components except when they are very
> close together in frequency, within what's called a "critical band".
>
> Our deafness to phase information cannot, however, be absolute, and
> there's reason to believe that inaccuracies in the reproduction of
> phase ("phase coherence") can make an audible difference to reproduced
> sound under some conditions (hence the speaker design issues that you
> and Kal discussed). The evidence on this is not decisive, and there
> are other possible explanations for some of the phenomena that people
> report, but there's also a general belief that phase-coherent sound
> reproduction is a Good Thing, and it's hard to argue with that. In
> practice, outside the domain of high-end electrostatic headphones, it's
> almost impossible to achieve good phase coherence from real transducers.
Before I can understand this I need to know: is the problem physical (i.e. the
sound waves cancel each other out when they are out of phase) or
psychophysical (i.e. an illusion). In the former case, the phenomenon should
completely disappear when listening to headphones, in the latter it should be
far worse with headphones. You seem to be saying that it is the latter - i.e.
that we perceive phase differences in some circumstances.
While we are at it, does anyone understand the "moon illusion" and why mirrors
reverse left-right but not up-down? I have no certificates to offer I'm
afraid.
cheers,
Alain
Frank Galvin
----------
In article <3B00A04D...@nyu.edu>, Tony Movshon
<mov...@nyu.edu> wrote:
> [...]
> This seems to have drifted way off topic for rmcr. Apologies to those
> who care about such things.
> --
> Tony Movshon
> mov...@nyu.edu
Yup, my thought was for all of us to take up residence, bag and
baggage, at some (any) audio newsgroup - just barge in
unannounced and keep this up until someone asks, 'Who *are* you
people?' 'Oh, we're just refugees from RMCR; don't mind us.' But
then the whole subject of audio reproduction fascinates me and I
do believe that RMCR people who aren't interested just stay away
(in droves!). Me, I'm archiving the discussion. (Duh, on my
little notepad until I find a better way! Direct-to-Disc?)
When Tang Huyen moved this to a more reasonable subject heading I
think (from their responses) some people may have missed this
exchange:
>>[Later, Dave (Cook) also wrote]:
>The problem might be that the DDD recordings have a much wider stereo
> separation, leading to the "hole in the middle." You might try one or
> all of the following:
> * toe the speakers in
> * move your listening position further away
> * move the speakers closer together
and I says,
>>None of which I can do with my Stax electrostatic phones, where I
>>claim I can *hear* the improvements best - obviously.
>>My ADS L-810s [from 1978; no laughing allowed, now] are toed in,
>>about 5 feet apart, and sound terrific - no holes; instrument
>>location, depth of soundstage imaging, bass response etc. fine -
>>with analog recordings. Hole-y, diffused imaging, thin bass, and
>>glare is what I get with DDDs, *until* I do you-know-what. All
>>other things are equal.
>>P.S. [fyi] I use the old Shure 'An Audio Obstacle Course' record
>>to check my speaker phasing. Primitive and amateur, perhaps, but
>>it woiks.
...and, with all due respect to Samir, my ears are just fine,
thank you very much...
Best to all,
Frank
(...who will eventually summarize and restate - count on it...)
Tony Movshon
> Tony Movshon wrote:
> Before I can understand this I need to know: is the problem physical (i.e. the
> sound waves cancel each other out when they are out of phase) or
> psychophysical (i.e. an illusion).
It's psychophysical but not an illusion. It's a specific deafness. You don't
call it an "illusion" that we can't see ultraviolet light?
> While we are at it, does anyone understand the "moon illusion" and why mirrors
> reverse left-right but not up-down?
Sort of; and they don't.
--
Tony Movshon
mov...@nyu.edu
Alain Dagher
Tony Movshon wrote:
> Alain Dagher wrote:
> > Tony Movshon wrote:
> > > The approximate truth is ...
>
> > Before I can understand this I need to know: is the problem physical (i.e. the
> > sound waves cancel each other out when they are out of phase) or
> > psychophysical (i.e. an illusion).
>
> It's psychophysical but not an illusion. It's a specific deafness. You don't
> call it an "illusion" that we can't see ultraviolet light?
I don't see the analogy. Suppose I have two speakers out of phase. If I hear the
sound from one speaker alone, then turn it off and hear the sound from the other
speaker, they sound identical. But if I have them both on at the same time, it
sounds different? That is not the same as not seeing UV.
>
> > While we are at it, does anyone understand the "moon illusion" and why mirrors
> > reverse left-right but not up-down?
>
> Sort of; and they don't.
>
The latter is without a doubt the least satisfying answer to the mirror reversal
illusion. (Not that that makes it wrong.) There is a good discussion about this in
the Oxford Companion to Philosphy.
Did you see that paper in PNAS on the moon illusion?
Alain
Tony Movshon
> Tony Movshon wrote:
> > Alain Dagher wrote:
> > > Tony Movshon wrote:
> > > > The approximate truth is ...
> >
> > > Before I can understand this I need to know: is the problem physical (i.e. the
> > > sound waves cancel each other out when they are out of phase) or
> > > psychophysical (i.e. an illusion).
> >
> > It's psychophysical but not an illusion. It's a specific deafness. You don't
> > call it an "illusion" that we can't see ultraviolet light?
>
> I don't see the analogy. Suppose I have two speakers out of phase. If I hear the
> sound from one speaker alone, then turn it off and hear the sound from the other
> speaker, they sound identical. But if I have them both on at the same time, it
> sounds different? That is not the same as not seeing UV.
Perhaps I misunderstood your question. The auditory system uses the sound
pressure waveforms it encodes to try to reconstruct the disposition of sound
sources in the environment. Neglecting room and speaker effects, the sound
from a pair of in-phase loudspeakers corresponds *within the limits of
auditory encoding* to the sound pattern produced by localized sources in a
"typical" acoustic environment. The auditory system thus happily reconstructs
a percept of those sources. However, the sound from a pair of out-of-phase
loudspeakers does not correspond to the sound pattern produced by localized
sources any typical acoustic environment but to some strange and atypical
pattern of diffuse sources. The auditory system tries to reconstruct the
sound sources anyway (that's its job), with the result that you perceive
diffuse and nonlocal sources in an "odd-sounding" acoustic environment.
These are both "illusions", if you like, since the sources do not exist in
the room as we hear them. But it's an odd use of the word "illusion", since
within the limits of hearing, the information at the ear corresponds to the
information that real sound sources would produce. I prefer to think of it
as a selective deafness, because if we could accurately encode all the
information that reaches the ear (e.g. relative phase) then we might be
able to tell that the sound came from two point sources, and we would not
perceive the sound from two speakers as defining a "complete" sound field.
When we look at a painting of a flower, we perceive the color of the flower
as matching the color of real flowers, despite the fact that the wavelength
distribution of light from the painting and the real flowers are quite
different. The percept is due to the fact that we are blind to many
variations in wavelength distribution and therefore cannot distinguish
the real flowers' distribution from the painting's. Is that an "illusion"?
I suppose you could call these percepts "illusions". But it's more usual in
psychophysics to use the word "illusion" to describe situations in which
you *mis*perceive something despite encoding correct physical information;
textbook examples of illusions are typically of this kind (e.g. the Muller-
Lyer illusion).
The situations here are ones in which the information encoded by the senses
is not sufficient to discriminate the "real" from the "illusory", and our
brains therefore make the only reasonable assumption: that the "illusory"
sources are like "real" ones. Perhaps we should call them "deceptions", not
"illusions".
> Did you see that paper in PNAS on the moon illusion?
Yes, it's by an old friend and colleague of mine. I don't think I'd call
it an "explanation" in the sense that I would use the word.
--
Tony Movshon
mov...@nyu.edu
Tony Movshon
> In article <3B00A04D...@nyu.edu>, Tony Movshon wrote:
> > [...]
> > This seems to have drifted way off topic for rmcr. Apologies to those
> > who care about such things.
>
> baggage, at some (any) audio newsgroup - just barge in
> unannounced and keep this up until someone asks, 'Who *are* you
> people?' 'Oh, we're just refugees from RMCR; don't mind us.'
I assume you've not visited the rec.audio newsgroups much. Some of the
folks over there make the flat-earth crowd and the Holocaust revisionists
seem like paragons of reasonableness.
--
Tony Movshon
mov...@nyu.edu
Rodger Whitlock
<mov...@nyu.edu> wrote:
> Which, by the way, leads me to wonder why in the early days of stereo
> the engineers didn't choose to encode stereo as L+R/L-R.
But on LP that's precisely what we have: L+R in the horizontal
plane, L-R in the vertical.
Attained by mechanical means, admittedly.
--
Rodger Whitlock
Victoria, British Columbia, Canada
Tony Movshon
>
> On Mon, 14 May 2001 09:18:21 -0400, Tony Movshon
> <mov...@nyu.edu> wrote:
>
> > Which, by the way, leads me to wonder why in the early days of stereo
> > the engineers didn't choose to encode stereo as L+R/L-R.
>
> But on LP that's precisely what we have: L+R in the horizontal
> plane, L-R in the vertical.
>
> Attained by mechanical means, admittedly.
True. I should have said "decode" instead of "encode".
--
Tony Movshon
mov...@nyu.edu
Alain Dagher
Tony Movshon wrote:
> Alain Dagher wrote:
>
> > I don't see the analogy. Suppose I have two speakers out of phase. If I hear the
> > sound from one speaker alone, then turn it off and hear the sound from the other
> > speaker, they sound identical. But if I have them both on at the same time, it
> > sounds different? That is not the same as not seeing UV.
>
> Perhaps I misunderstood your question. The auditory system uses the sound
> pressure waveforms it encodes to try to reconstruct the disposition of sound
> sources in the environment. Neglecting room and speaker effects, the sound
> from a pair of in-phase loudspeakers corresponds *within the limits of
> auditory encoding* to the sound pattern produced by localized sources in a
> "typical" acoustic environment. The auditory system thus happily reconstructs
> a percept of those sources. However, the sound from a pair of out-of-phase
> loudspeakers does not correspond to the sound pattern produced by localized
> sources any typical acoustic environment but to some strange and atypical
> pattern of diffuse sources. The auditory system tries to reconstruct the
> sound sources anyway (that's its job), with the result that you perceive
> diffuse and nonlocal sources in an "odd-sounding" acoustic environment.
Thanks for the explanation.
However, I still have a problem. The wavelength of musical sounds is from about 4 m to
less than 5 cm. That means when I listen to my in-phase speakers, reflected sounds above
a certain frequency will be out of phase compared to the direct sounds. Even worse, in a
concert hall, each individual violin in the violin section of the orchestra is separated
from the other violins by a distance of the same order of magnitude as the wavelength.
So the sounds coming from the violin section consists of numerous out of phase signals.
What I mean is that hearing out of phase sounds is the norm. So why does it feel "weird"
when it is coming from your stereo?
Alain
Stephen McElroy
> Frank Galvin wrote:
> > Yup, my thought was for all of us to take up residence, bag and
> > baggage, at some (any) audio newsgroup - just barge in
> > unannounced and keep this up until someone asks, 'Who *are* you
> > people?' 'Oh, we're just refugees from RMCR; don't mind us.'
> I assume you've not visited the rec.audio newsgroups much. Some of the
> folks over there make the flat-earth crowd and the Holocaust revisionists
> seem like paragons of reasonableness.
The moderated group would be the best bet for a civil answer. Some genuine
authoritues in the field do participate, if grumpily.
Stephen
Kalman Rubinson
<al...@bic.mni.mcgill.ca> wrote:
>However, I still have a problem. The wavelength of musical sounds is
> from about 4 m to less than 5 cm. That means when I listen to my in-phase
> speakers, reflected sounds above a certain frequency will be out of phase
> compared to the direct sounds.
That's true and that's why speaker placement, room dimensions and room
treatment have huge effects on the quality of reproduced sound.
> Even worse, in a concert hall, each individual violin in the violin section of
> the orchestra is separated from the other violins by a distance of the same
> order of magnitude as the wavelength.
That's part of what makes them sound like an ensemble instead of one
BIG violin. This information, along with the reflected/delayed sound
of the performance site, need to be included in order to (re)create
the illusion of the event. A recording made under anechoic conditions
sounds unnatural (because it is).
>So the sounds coming from the violin section consists of numerous out of
> phase signals. What I mean is that hearing out of phase sounds is the norm.
> So why does it feel "weird" when it is coming from your stereo?
You are hearing the phase/amplitude effects of the listening room
superimposed on those of the performance site. There are devices,
such as the SigTech, Tact RCS and others, which can correct for the
listening room effects, at a price.
Kal
Tony Movshon
> Tony Movshon wrote:
> > pressure waveforms it encodes to try to reconstruct the disposition of sound
>
> Thanks for the explanation.
>
> However, I still have a problem. The wavelength of musical sounds is from about 4 m to
> less than 5 cm. That means when I listen to my in-phase speakers, reflected sounds above
> a certain frequency will be out of phase compared to the direct sounds. Even worse, in a
> concert hall, each individual violin in the violin section of the orchestra is separated
> from the other violins by a distance of the same order of magnitude as the wavelength.
> So the sounds coming from the violin section consists of numerous out of phase signals.
> What I mean is that hearing out of phase sounds is the norm. So why does it feel "weird"
> when it is coming from your stereo?
The simplest answer is that not all "out-of-phase" patterns are the same.
The patterns that arise from multiple violins in a hall are not perfectly
in-phase, but they are not perfectly out of phase either; the phases are
roughly random, and different sources therefore tend both to cancel and
reinforce. Out of phase speakers give you perfect inverted phase, which
you would never encounter in nature.
The more general answer is that in ways that are poorly understood, the
auditory system "knows" the characteristics of sound sources and acoustic
environments, e.g. that sound sources tend to be local, that sounds tend
to reach the ear by multiple direct and reflected paths, etc. The result
in any particular acoustic situation is a mind-numbingly complex pattern
of delays, but in normal environments only a small subset of the possible
patterns is ever encountered. That is, there are strong statistical
regularities in both natural sound sources and natural acoustic
environments. The auditory system is extremely good at exploiting these
regularities to calculate the likely nature of both the sound source
and the environment, and what you perceive is the result of that
calculation. To reproduce music, you have to master the art of fooling
the calculator.
For a rather lengthy discussion of this material, you might want to browse
through a book called "Auditory Scene Analysis" by Albert Bregman, who is
just down the road from you. You can get more information about this and
hear some demos at
http://www.psych.mcgill.ca/labs/auditory/laboratory.html
--
Tony Movshon
mov...@nyu.edu
Tony Movshon
> The moderated group would be the best bet for a civil answer. Some genuine
> authoritues in the field do participate, if grumpily.
Which group is the moderated one?
--
Tony Movshon
mov...@nyu.edu
Kalman Rubinson
wrote:
>Stephen McElroy wrote:
>> The moderated group would be the best bet for a civil answer. Some genuine
>> authoritues in the field do participate, if grumpily.
>
>Which group is the moderated one?
rec.audio.high-end but there's still a lot of muck on the floor.
Kal
Frank Galvin
----------
Tony Movshon <mov...@nyu.edu> wrote:
>> > [...]
>> > This seems to have drifted way off topic for rmcr. Apologies to
> those
>> > who care about such things.
[I replied]:
>> Yup, my thought was for all of us to take up residence, bag and
>> baggage, at some (any) audio newsgroup - just barge in
>> unannounced and keep this up until someone asks, 'Who *are* you
>> people?' 'Oh, we're just refugees from RMCR; don't mind us.'
[and Tony had this to say]
:
> I assume you've not visited the rec.audio newsgroups much. Some of the
> folks over there make the flat-earth crowd and the Holocaust
> revisionists
> seem like paragons of reasonableness.
> --
> Tony Movshon
> mov...@nyu.edu
A friend of mine (PhD in E.E., MIT - be impressed, or at least
don't laugh) who would never, never doubt my ear, suggested this
morning that my left and right brains are wired absolutely out of
phase, thus contributing mightily to the problem! Question is,
would I seem more reasonable at rec.audio? Probably would, huh?
Do hope you'll address the remainder of my post, the reasonable
bit, before I am forced to sum up and jump ship...
Best & thanks,
Frank
Simon Roberts
: A friend of mine (PhD in E.E., MIT - be impressed, or at least
: don't laugh) who would never, never doubt my ear, suggested this
: morning that my left and right brains are wired absolutely out of
: phase, thus contributing mightily to the problem! Question is,
: would I seem more reasonable at rec.audio? Probably would, huh?
: Do hope you'll address the remainder of my post, the reasonable
: bit, before I am forced to sum up and jump ship...
You should try listening upside down first.
Simon
Kalman Rubinson
<fga...@mediaone.net> wrote:
>Do hope you'll address the remainder of my post, the reasonable
>bit, before I am forced to sum up and jump ship...
Don't sum up whatever you do. Your two out-of-phase halves
will cancel and your mind will disappear.
Kal (Assoc. Prof. of Physiol. & Neurosci., so you can believe me.)
mt
Kal, we believed you *before*...
MrT
Alain Dagher
Frank Galvin wrote:
> A friend of mine (PhD in E.E., MIT - be impressed, or at least
> don't laugh) who would never, never doubt my ear, suggested this
> morning that my left and right brains are wired absolutely out of
> phase, thus contributing mightily to the problem!
But actually the left and right hemispheres ARE wired differently for sound.
The connections from the auditory cortex have more myelin on the left than
the right, which means that they conduct faster (myelin is the insulator).
I wonder whether stereo systems are "tuned" for left hemisphere dominance -
which could be done by introducing a slight delay in the right speaker
channel.
Alain
Kalman Rubinson
><<Kal (Assoc. Prof. of Physiol. & Neurosci., so you can believe me.)>>
>
>Kal, we believed you *before*...
Exactly what I intended.
Kal
Frank Galvin
----------
In article <ajv2gtcjrkobqdv9i...@4ax.com>, Kalman
Rubinson <k...@nyu.edu> wrote:
I do believe you, and what's more, I think it's already starting.
Frank
Tony Movshon
Since nerve signals from both ears go to both hemispheres, that would
be a bad idea.
--
Tony Movshon
mov...@nyu.edu