I'm just too dense to figure this one out. Now that I've asked twice and the same answer has come out, I'll have to ask more directly and reveal my utter lack of understanding:
How does one take an audio signal and create 0.5x at the speaker? Is it merely an 8-ohm resistor in series with the 8-ohm driver? How would the 0.5x be achieved at the line-level signal?
This thread is riveting stuff and I'm enjoying the content, but please forgive my technical ignorance.
How does one take an audio signal and create 0.5x at the speaker? Is it merely an 8-ohm resistor in series with the 8-ohm driver? How would the 0.5x be achieved at the line-level signal?
This thread is riveting stuff and I'm enjoying the content, but please forgive my technical ignorance.
The answer is too simple, and can be easily ignored !
In the schema in post 97, it is a self matrixing configuration ! Just feed the 'hot' leads of your amplifier output Left and Right stereo signals to L and R taps, and connect the negative terminals of your amplifier output to the common tap. That's all !
The x = 0.5 if formed by the fact that according to principle of superposition the both L and R signals as connected see a voltage division ratio of 0.5 between center and side element signals.
- Elias
I'm not sure if single point MS recording microphone is optimal at all !
There were posted samples of Schoeps microphone demonstrations, including XY and spaced techniques, it was in the Stereolith thread. Better to find them out !
I concluded that time, and just recently I listened them again with SSSx5, that pure XY was not the best sounding because it did not render the ambiance. The best was a spaced mic pair, spacing about 17 cm (?), presumable because it produces time cues as well.
- Elias
Bingo. I too have been "ruined" for conventional stereo. In my case it happened by using pink noise to balance L-R. When using steady mono pink noise to generate a sharp phantom center image, one discovers just how fragile and fake a phantom center is. The smallest head movements--even with dipole and omni speakers--reveal the gross colorations from HRTF, and the phantom collapses. It's impossible to ignore. After that experience it has become much harder to relax into a believable 2-channel soundstage. [BTW, anyone who does not experience this effect with pink noise may have a room that is too live or some other artifact going on; try listening for the effect in nearfied]
The widespread acceptance of 2-ch stereo is simply the result of conditioning. If one always listens in this format, the brain is prediposed to ignore the problems, and adaptation takes place pretty quickly. The changeup introduced by SSS or other format gives the brain a whole new reference, and adaptation to 2 ch is no longer quick and unnoticeable for me.
'Pure' M-S recordings may not be the answer to everything. But I guess they save your bacon while exploring other configurations than triangle stereo, and certainly insulates you from close-miked 'metre-wide-mouthed' singers, 'head inside the piano' and such like atrocities, and that tyrannous gadget, the panpot.
I am totally with Tubamark as regards the insanity of 'setting up' with pink noise. Conditioning??!!!
Here is a link that I re-read recently <http://www.holophony.net/index.htm>, and here are those (almost) forgotten LEDR tests <http://www.audiocheck.net/audiotests_ledr.php> that often (with most speakers, in most rooms) are very very revealing. Purely a personal and subjective observation--no need for any ruffled feathers.
Once again--does anyone know of a good source of M-S component downloads for testing purposes?
I am totally with Tubamark as regards the insanity of 'setting up' with pink noise. Conditioning??!!!
Here is a link that I re-read recently <http://www.holophony.net/index.htm>, and here are those (almost) forgotten LEDR tests <http://www.audiocheck.net/audiotests_ledr.php> that often (with most speakers, in most rooms) are very very revealing. Purely a personal and subjective observation--no need for any ruffled feathers.
Once again--does anyone know of a good source of M-S component downloads for testing purposes?
Possible proof of a reinvent ?
AES project report for professional audio — Specifications for audio on high-capacity media, AES-R1-1997
pdf download:
http://citeseerx.ist.psu.edu/viewdoc/download;jsessionid=2DEC9C47720A818816675E09D9A8B325?doi=10.1.1.196.2438&rep=rep1&type=pdf
"An alternative to use of an array of monopole loudspeakers facing the listening area and attempting to produce a diffuse
field from multiple uncorrelated sources is to use purpose-built loudspeakers which emphasize the diffuse-field
component of their output to reproduce the reverberant field as reflected sound in the reproducing space. This technique
dates back to Lauridsen21 who used a mid-side (MS) loudspeaker for stereo."
and the reference 21 is:
"21Lauridsen, H., Schlegel, F., "Stereofonie und richtungsdiffuse Klangwiedergabe," Gravesaner Blatter, 5 28-
50 (August, 1956)."
Ok, so in 1956 someone used MS loudspeaker for a stereo reproduction, and utilised the listening room wall reflections as a transmission medium.
- Elias
AES project report for professional audio — Specifications for audio on high-capacity media, AES-R1-1997
pdf download:
http://citeseerx.ist.psu.edu/viewdoc/download;jsessionid=2DEC9C47720A818816675E09D9A8B325?doi=10.1.1.196.2438&rep=rep1&type=pdf
"An alternative to use of an array of monopole loudspeakers facing the listening area and attempting to produce a diffuse
field from multiple uncorrelated sources is to use purpose-built loudspeakers which emphasize the diffuse-field
component of their output to reproduce the reverberant field as reflected sound in the reproducing space. This technique
dates back to Lauridsen21 who used a mid-side (MS) loudspeaker for stereo."
and the reference 21 is:
"21Lauridsen, H., Schlegel, F., "Stereofonie und richtungsdiffuse Klangwiedergabe," Gravesaner Blatter, 5 28-
50 (August, 1956)."
Ok, so in 1956 someone used MS loudspeaker for a stereo reproduction, and utilised the listening room wall reflections as a transmission medium.
- Elias
Ok, so in 1956 someone used MS loudspeaker for a stereo reproduction, and utilised the listening room wall reflections as a transmission medium.
- Elias
..and this was before stereo (music) was really available to the masses (which started to occur around '58).
Honestly, I wouldn't be at all surprised if Blumlein himself didn't experiment with this at some point (late '30s).
"An alternative to use of an array of monopole loudspeakers facing the listening area and attempting to produce a diffuse
field from multiple uncorrelated sources is to use purpose-built loudspeakers which emphasize the diffuse-field
component of their output to reproduce the reverberant field as reflected sound in the reproducing space. This technique
dates back to Lauridsen21 who used a mid-side (MS) loudspeaker for stereo."
_______________________________________________
And vice versa (with slight modifications):
An alternative to use purpose-built loudspeakers which emphasize the diffuse-field component of their output to reproduce
the reverberant field as reflected sound in the reproducing space is to use an array of (monopole) loudspeakers facing the
listening area to produce a diffuse field from multiple uncorrelated sources.
One design option are 2 or more bending wave loudspeakers, each mimicking an array of multiple uncorrelated sources, thereby
increasing the number of sources per channel with the order of the modes excited (which is an increase of sources with
frequency). The problem of interference in conventional stereo loudspeakers, which becomes detrimental to imaging in the
Khz range, can be circumvented. Of course that approach is also possible with phase steered arrays of multiple very small
transducers while having the decorrelation by signal processing.
Stereo format can be transmitted by either a single panel using multi channel exciters or separated panels for each channel.
The advantages over the (nearfield coherent radiating) single source mid-side loudspeaker are:
- undelayed reproduction of principal left and principal right side images in relation to the center image,
as it was recorded: There is no additional delay for side images arising from reflection paths, which is added
to any delay present on the recording.
- operates far more independently from room and setup: The speaker takes his primary "builtin decorralation algorithm"
into every room in the same way. Instead the "decorrelation algorithm" of the Mid-Side system family is room dependent
especially due to
- asymmetry in frequency dependent absorption/diffusion of the side walls
- absolute length of reflection paths
- left/right relative length of reflection paths
The single source mid side speaker IMO mixes up the solutions to two different kinds of problem without solving
just one of them concisely:
1st task IMO is the imaging of the direct sound of left/center/right phantom sources as intended on the recording.
To give up on that task is leaving the "reproduction" paradigm and turning over to a "production" kind of paradigm.
I do not state that this is forbidden, but one should be aware of the fact in doing so.
(I am a bit conservative due to that issue, nevertheless able to accept compromise...)
2nd (but not the "lesser") task IMO is to have a soundfield which comes close to a "natural" diffuse sound field
according to parameters like e.g. IACC, especially in the reverberant part.
A principal left positioned phantom source e.g. is not naturally "way more" diffuse than a centered one.
Reverberance is not (over) proportional to the (virtual) source being "off center".
But this discussion is useful in making us aware (again) of the fact, that stereo does neither solve the
mentioned tasks by the recording format itself, nor by conventional speaker setups.
So the "single source mid side family" of approaches circumvents especially the interference problems in
centered phantom images and their associated spatial instability (which i interpret as a kind of "spatial aliasing"
due to spatial "undersampling") and associated lack of tonal quality at higher frequencies, especially when
sitting outside the median plane while listening.
Since centered sources are very important i do not doubt that it may offer an alternative possibly having
better fidelity than stereo with conventional (fullrange coherent) speakers, if implemented in a canny way.
field from multiple uncorrelated sources is to use purpose-built loudspeakers which emphasize the diffuse-field
component of their output to reproduce the reverberant field as reflected sound in the reproducing space. This technique
dates back to Lauridsen21 who used a mid-side (MS) loudspeaker for stereo."
_______________________________________________
And vice versa (with slight modifications):
An alternative to use purpose-built loudspeakers which emphasize the diffuse-field component of their output to reproduce
the reverberant field as reflected sound in the reproducing space is to use an array of (monopole) loudspeakers facing the
listening area to produce a diffuse field from multiple uncorrelated sources.
One design option are 2 or more bending wave loudspeakers, each mimicking an array of multiple uncorrelated sources, thereby
increasing the number of sources per channel with the order of the modes excited (which is an increase of sources with
frequency). The problem of interference in conventional stereo loudspeakers, which becomes detrimental to imaging in the
Khz range, can be circumvented. Of course that approach is also possible with phase steered arrays of multiple very small
transducers while having the decorrelation by signal processing.
Stereo format can be transmitted by either a single panel using multi channel exciters or separated panels for each channel.
The advantages over the (nearfield coherent radiating) single source mid-side loudspeaker are:
- undelayed reproduction of principal left and principal right side images in relation to the center image,
as it was recorded: There is no additional delay for side images arising from reflection paths, which is added
to any delay present on the recording.
- operates far more independently from room and setup: The speaker takes his primary "builtin decorralation algorithm"
into every room in the same way. Instead the "decorrelation algorithm" of the Mid-Side system family is room dependent
especially due to
- asymmetry in frequency dependent absorption/diffusion of the side walls
- absolute length of reflection paths
- left/right relative length of reflection paths
The single source mid side speaker IMO mixes up the solutions to two different kinds of problem without solving
just one of them concisely:
1st task IMO is the imaging of the direct sound of left/center/right phantom sources as intended on the recording.
To give up on that task is leaving the "reproduction" paradigm and turning over to a "production" kind of paradigm.
I do not state that this is forbidden, but one should be aware of the fact in doing so.
(I am a bit conservative due to that issue, nevertheless able to accept compromise...)
2nd (but not the "lesser") task IMO is to have a soundfield which comes close to a "natural" diffuse sound field
according to parameters like e.g. IACC, especially in the reverberant part.
A principal left positioned phantom source e.g. is not naturally "way more" diffuse than a centered one.
Reverberance is not (over) proportional to the (virtual) source being "off center".
But this discussion is useful in making us aware (again) of the fact, that stereo does neither solve the
mentioned tasks by the recording format itself, nor by conventional speaker setups.
So the "single source mid side family" of approaches circumvents especially the interference problems in
centered phantom images and their associated spatial instability (which i interpret as a kind of "spatial aliasing"
due to spatial "undersampling") and associated lack of tonal quality at higher frequencies, especially when
sitting outside the median plane while listening.
Since centered sources are very important i do not doubt that it may offer an alternative possibly having
better fidelity than stereo with conventional (fullrange coherent) speakers, if implemented in a canny way.
Last edited:
If we take as a conclusion from the Bennett et al. paper
http://decoy.iki.fi/dsound/ambisoni...ssment of stereophonic_Bennett et al_1985.pdf
that stereo is flawed especially at high frequencies, where sound wavelengths
are small compared to speaker separation and common listener movement, are
we really surprised ?
Is it wise to neglect the good share of the "news", which can be paraphrased
like "stereo (and conventional speaker setup) works quite well for low to mid
frequencies, given that some restrictions (listening position) are met" ?
The natural way of improving a system is IMO maintaining its virtues and
approaching its odds.
The odds are rather specific: Interference at higher frequencies has significant impact
on reproduction quality above say 1Khz.
That frequency range of interest (strangely !) quite coincides with the range where
our hearing gives up on evaluating interaural phase difference (because it gets ambiguous)
for evaluation of (horizontal plane) sound source direction and starts relying on alternative
cues like IDT and HRTF.
The good news is: We are allowed to "mess up" phase difference between left and right
signal now at the listening area !
Are we allowed to mess up group delay in the range of several ms regarding the "direct
sound" component ? Surely not.
Btw. if Toole finds that subjects prefer wide dispersion speakers, what is it specifically
that makes most or many listeners prefer those ?
Maybe i am a bit "battlesome" today, so please take my words with a grain of salt ...
A wide dispersion speaker is the only one in todays common technologogical environment
of (mainly) pistonic acting (and mainly coherent radiating) speakers, which is able to
partially heal interference artefacts and lacking similarity of speaker generated soundfield parameters
in comparison to natural soundfields, as experienced at "acoustical" music venues.
That is mainly achieved by lateral reflections, given the listening room provides a supporting
environment.
http://decoy.iki.fi/dsound/ambisoni...ssment of stereophonic_Bennett et al_1985.pdf
that stereo is flawed especially at high frequencies, where sound wavelengths
are small compared to speaker separation and common listener movement, are
we really surprised ?
Is it wise to neglect the good share of the "news", which can be paraphrased
like "stereo (and conventional speaker setup) works quite well for low to mid
frequencies, given that some restrictions (listening position) are met" ?
The natural way of improving a system is IMO maintaining its virtues and
approaching its odds.
The odds are rather specific: Interference at higher frequencies has significant impact
on reproduction quality above say 1Khz.
That frequency range of interest (strangely !) quite coincides with the range where
our hearing gives up on evaluating interaural phase difference (because it gets ambiguous)
for evaluation of (horizontal plane) sound source direction and starts relying on alternative
cues like IDT and HRTF.
The good news is: We are allowed to "mess up" phase difference between left and right
signal now at the listening area !
Are we allowed to mess up group delay in the range of several ms regarding the "direct
sound" component ? Surely not.
Btw. if Toole finds that subjects prefer wide dispersion speakers, what is it specifically
that makes most or many listeners prefer those ?
Maybe i am a bit "battlesome" today, so please take my words with a grain of salt ...
A wide dispersion speaker is the only one in todays common technologogical environment
of (mainly) pistonic acting (and mainly coherent radiating) speakers, which is able to
partially heal interference artefacts and lacking similarity of speaker generated soundfield parameters
in comparison to natural soundfields, as experienced at "acoustical" music venues.
That is mainly achieved by lateral reflections, given the listening room provides a supporting
environment.
Last edited:
To circumvent interference patterns alternatively the number and spatial density of real sound
sources may be increased with frequency, due to the frequency range of interest, and an
appropriate decorrelation between these sources may be established.
If both are matched to the grade of deficiency (rising with frequency) of the stereo setup and
the "mode" in spatial hearing - which both are a gradually shifting with frequency -
we have improved reproduction of the stereo format (and also other multichannel formats you
can think of).
sources may be increased with frequency, due to the frequency range of interest, and an
appropriate decorrelation between these sources may be established.
If both are matched to the grade of deficiency (rising with frequency) of the stereo setup and
the "mode" in spatial hearing - which both are a gradually shifting with frequency -
we have improved reproduction of the stereo format (and also other multichannel formats you
can think of).
Last edited:
The spaciousness of soundstage, and naturalness of center images is quite addictive with SSS, but I agree - I do miss accurate image placement on some source material.
A workable improvement to some of these issues in the SSS would be to 1) make the drivers (especially sides) highly directional (that is, CD, and not just beaming effects), by use of waveguides, dipoles (tricky at close spacing) or multi-driver arrays, and then simply 2) apply a delay to the Mid speaker to equal path difference to listener (or push the center speaker back several feet, but this just creates additional problems), and apply sohisticated spectrum + delay correction to the sides, similar to Audyssey. Very do-able with digital technology . . . Of course all off this ruins the beautiful simplicity of the passive matrixed SSS . . . but the matrixing itself could still be passively performed at line-level before all the correction hoo-ha.
Granted that lower midrange and below could still present a direct signal to the listener, but it would be relatively benign (or could also be corrected digitally). The signals of all three need to be in-sync from upper bass and below. All do-able with digital processing of each channel.
It still won't be 100% perfect, but would provide really impressive performance in a wide range of domestic size rooms. I'm not currently using any digital gear or PC in my signal chain (aside from a CD player), but would love to try it sometime.
--- Mark
Last edited:
Oliver, I see you've been pondering this a lot, but have you build a prototype yet and listened by yourself ? But in a forum like this the endless theorising is part of the fun too
You've been advocating this a lot in the past too, but why no one is using them ?
Yes, this is the problem with stereo triangle ! However I have a feeling some people may not be able to see it as a problem, afterall some say they perceive solid phantoms even at high frequencies. How is it possible ?
This delay thing is a hoax, and I'm not convinced it has much of a meaning when we are targeting small room acoustic reflective spaces like normal living rooms.
This may be true for a direct sound component, but then again it is only a small portion of what we perceive in a small room.
Yes, so it sort of nullifies the comparison to start with, considering the tasks stated, but it may also indicate that the tasks are not definitive requirement goals for a stereo system ?
Yes I think it may be the most important, and the most easily and immediately noticeable property of SSS.
After that its really hard to go back to conventional triangle. Since I build a prototype of SSS I have been able to do a A/B comparison between stereo triangle and SSS and I have not yet encountered a situation or a recording where I would have prefered stereo triangle. And why is that ?
- Elias
But in a forum like this the endless theorising is part of the fun too You've been advocating this a lot in the past too, but why no one is using them ?
Yes, this is the problem with stereo triangle ! However I have a feeling some people may not be able to see it as a problem, afterall some say they perceive solid phantoms even at high frequencies. How is it possible ?
This delay thing is a hoax, and I'm not convinced it has much of a meaning when we are targeting small room acoustic reflective spaces like normal living rooms.
This may be true for a direct sound component, but then again it is only a small portion of what we perceive in a small room.
Yes, so it sort of nullifies the comparison to start with, considering the tasks stated, but it may also indicate that the tasks are not definitive requirement goals for a stereo system ?
Yes I think it may be the most important, and the most easily and immediately noticeable property of SSS.
After that its really hard to go back to conventional triangle. Since I build a prototype of SSS I have been able to do a A/B comparison between stereo triangle and SSS and I have not yet encountered a situation or a recording where I would have prefered stereo triangle. And why is that ?
- Elias
Everyone should know it by now, even the most conservative stereo advocates, that stereo does not work as intended by any conventional means !
Surely ? The pudding has been put on the table without a proof ! Where are the references explicitely stating that group delay in a small room acoustic reflective spaces has much relevance ?
I accept group delay can be perceived in anechoic conditions and / or with headphones and using synthetic clicks etc nonrelevancies, but come on, where are the studies on group delay in music reproduction in small rooms ?
I wish Toole had tested and reported any unconventional stereo speaker
In the past I was advocating a speaker with high directivity low and mid freqs and wide directivity at high freqs, totally opposite to general speaker, and I had partially satisfying experience with a dipole line array and a ceiling firing tweeter. It does exaclty what you say at high freqs. I also liked the midrange detail brought by high direct to reflection ratio. But it did not have the 'magic' of SSS, which I still have no full explanation why it works so good.
- Elias
Last edited:
An later attempt of a reinvent ?
I'm happy to realise that the general family of center placed side firing single loudspeakers are approaching the stereo localisation principles presented here:
project 184
from where quoted:
"Project 184: The Inaudible Loudspeaker -
Unmasking audio sound information by applying monaural stereophonics"
"We can not recognise the size of the sound source in the reproduced sound
We are all familiar with Von Bekésy’s problem (1960): the ‘in the box sound’ effect seems to increase with the decrease of the loudspeaker’s dimensions. In an experimental research on the relation between acoustic power, spectral balance and perceived spatial dimensions and loudness, Von Bekésy’s test subjects were unable to correctly indicate the relative dimensional shape of a reproduced sound source as soon as the source’s dimensions exceeded the actual shape of the reproducing loudspeaker box. One may conclude that the loudspeaker’s spatio-spectral properties introduce a message-media conflict when transmitting sound information.
We listen to the properties of the loudspeaker
Why does the ear lock so easily to the loudspeaker characteristics? Based on the hypotheses of Brian Moore (2002) et al., we may conclude that this is because the ear, by nature, produces two dimensional nerve signals to the brain that reflect the three-dimensional wave interference due to direct interaction of both the ear’s and the physical sound source’s spatio-spectral properties. Therefore any three-dimensional spatio-spectral ‘message’ information embedded in the signal to be transmitted is masked by the physically present ‘media’ information related to the loudspeaker’s properties. One could say that in prior art solutions the input signal of the loudspeaker system merely functions as a carrier signal that modulates the loudspeaker’s characteristics.
A phantom sound source appears to be more realistic
We all experience a ‘phantom’ sound source reproduced by a stereo-system to appear more realistic than the same sound reproduced from one box. However, spatial sound reproduction by means of stereophonic ‘virtual sources’ is a partial solution because it fails for stereophonic information that does not meet the L=R requirement for perfect phantom imaging. In fact stereophonics is no more than improved mono with the listening area restricted to a hot spot.
A new approach
For a new approach of this problem we considered that one better could overrule the monaural spatio-spectral coding than trying to manipulate it (as in crosstalk suppression or Bose-like diffusion approaches), thus preventing inter-aural cross correlation which is the root mechanism for the listener’s ability to localize any non-virtual sound sources i.e. loudspeakers.
A phantom loudspeaker
The resulting patent application describes a sound system that produces coherent vertical phantom images that cannot monaurally nor binaurally be decomposed to their root sources i.e. the actual loudspeaker drivers. The loudspeaker becomes audibly non-localizable and visually transparent and any multi-channel stereophonic horizontal configuration of these virtual speakers is possible without adding masking ‘media’ properties. As a consequence the listening area is not anymore restricted to a hot spot. Loudness perception is dramatically improved because the huge ‘out of the box’ sound screen will now fit to even the largest sound source shapes."
Some pics can be seen here:
terugblik 184
And the concept is being disscussed here:
http://www.diyaudio.com/forums/multi-way/187537-monaural-stereophonics.html
The similarity is simply the fact that side firing single loudspeaker produces virtual loudspeakers which cannot be localised, and these virtual speakers generate the 'second order' phantom images which form the perceived soud stage from the recording.
- Elias
I'm happy to realise that the general family of center placed side firing single loudspeakers are approaching the stereo localisation principles presented here:
project 184
from where quoted:
"Project 184: The Inaudible Loudspeaker -
Unmasking audio sound information by applying monaural stereophonics"
"We can not recognise the size of the sound source in the reproduced sound
We are all familiar with Von Bekésy’s problem (1960): the ‘in the box sound’ effect seems to increase with the decrease of the loudspeaker’s dimensions. In an experimental research on the relation between acoustic power, spectral balance and perceived spatial dimensions and loudness, Von Bekésy’s test subjects were unable to correctly indicate the relative dimensional shape of a reproduced sound source as soon as the source’s dimensions exceeded the actual shape of the reproducing loudspeaker box. One may conclude that the loudspeaker’s spatio-spectral properties introduce a message-media conflict when transmitting sound information.
We listen to the properties of the loudspeaker
Why does the ear lock so easily to the loudspeaker characteristics? Based on the hypotheses of Brian Moore (2002) et al., we may conclude that this is because the ear, by nature, produces two dimensional nerve signals to the brain that reflect the three-dimensional wave interference due to direct interaction of both the ear’s and the physical sound source’s spatio-spectral properties. Therefore any three-dimensional spatio-spectral ‘message’ information embedded in the signal to be transmitted is masked by the physically present ‘media’ information related to the loudspeaker’s properties. One could say that in prior art solutions the input signal of the loudspeaker system merely functions as a carrier signal that modulates the loudspeaker’s characteristics.
A phantom sound source appears to be more realistic
We all experience a ‘phantom’ sound source reproduced by a stereo-system to appear more realistic than the same sound reproduced from one box. However, spatial sound reproduction by means of stereophonic ‘virtual sources’ is a partial solution because it fails for stereophonic information that does not meet the L=R requirement for perfect phantom imaging. In fact stereophonics is no more than improved mono with the listening area restricted to a hot spot.
A new approach
For a new approach of this problem we considered that one better could overrule the monaural spatio-spectral coding than trying to manipulate it (as in crosstalk suppression or Bose-like diffusion approaches), thus preventing inter-aural cross correlation which is the root mechanism for the listener’s ability to localize any non-virtual sound sources i.e. loudspeakers.
A phantom loudspeaker
The resulting patent application describes a sound system that produces coherent vertical phantom images that cannot monaurally nor binaurally be decomposed to their root sources i.e. the actual loudspeaker drivers. The loudspeaker becomes audibly non-localizable and visually transparent and any multi-channel stereophonic horizontal configuration of these virtual speakers is possible without adding masking ‘media’ properties. As a consequence the listening area is not anymore restricted to a hot spot. Loudness perception is dramatically improved because the huge ‘out of the box’ sound screen will now fit to even the largest sound source shapes."
Some pics can be seen here:
terugblik 184
And the concept is being disscussed here:
http://www.diyaudio.com/forums/multi-way/187537-monaural-stereophonics.html
The similarity is simply the fact that side firing single loudspeaker produces virtual loudspeakers which cannot be localised, and these virtual speakers generate the 'second order' phantom images which form the perceived soud stage from the recording.
- Elias
Oliver, I see you've been pondering this a lot, but have you build a prototype yet and listened by yourself ?
I experimentally used 90 degrees "toed out" dipoles with narrow baffles
standing close ... which is not too far from proposed M-S behaviour IMO.
It was nice on some recordings but e.g. on jazz big band the lateral
phantoms sounded "distant" or "faint" or "diffusely localized" to my ears.
So i recognized this not being my cup of tea. Nevertheless with some effort
spent in optimization and proper placement in an appropriate room, i can
imagine it being an alternative approach.
Concerning bending wave loudspeakers:
Hard to believe that Ohm, Manger, Goebel and some other manufacturers
keep up their business without having customers ... some of them since
decades.
When you ask instead "Why are those BWL not more commonly used?" then
i may try a simple answer:
Because most BWL suffer from specific drawbacks and because they are
mostly expensive. "Use a BWL" is too simple an answer to deficiencies in
stereophony. A BWL has to be constructed in a specific way to achieve
this. I mentioned the intended behaviour before ...
Concerning group delay:
My experience says otherwise. Only for bass to lower midrange i may accept a lack of relevance.
Even in a smaller room, say >20 sqm, i often found audible anomalies to be more easily mapped
on measurable quantities when using group delay than when using FR (anechoic).
But to get there you ought to be an "RFZ believer", place your speakers accordingly and have
speakers which already have quite flat group delay at mid to high frequencies.
When living in a forest (of early reflections) even some close trees might be invisible at first
sight (as an allegory, not personally).
Last edited:
The Dipol 8 ?
Well, actually I'm not proposing M-S (which can be indeed reproduced by crossed dipoles) but the 'optimal' matrix with x = 0.5.
I have tested configuration with x = 0 and x = 0.5 and the difference is a night-day. I'll expect similarly great difference between x = 0.5 and x = 1 (MS).
So if you tested MS, you really havent tested what I propose. On the other hand I did not yet try pure M-S, so I also cannot give indication other than theoretical revolving.
I was more thinking of DIY designs ! Please do not suggest we should be bying ready made speakers
I want to see DIY BWL, with a proven measurements achieving the claimed 'decorrelation' !
Peer reviewed references please
- Elias
Indeed ! It has opened eyes to see what stereo can provide !
I do not miss at all the localisation of the tweeters of a conventional stereo triangle
Even the localisation of the tweeters was pretty accourate I believe the purpose is not to generate laser like side wall reflections, but to generate virtual loudspeakers which cannot be localised as such. If there would be only one reflection acting as a 'secondary speaker', it would be perceivable as a sound source. See one of the earlier post about virtual speakers.
On the other hand all the three speaker stereo matrix reproduction methods allways provide signal from the center speaker during the side panned signal !
- Elias
Everyone should take this as words of caution: Do not do pink noise tests with a conventional stereo triangle !
You can be disappointed, for a life time !
- Elias