A new theoretical model for stereo imaging

[Curvature]

I'm under the impression that most of our sound image localization happens well north of the transition region, while in the transition region and below we're mostly getting envelopment cues rather that image localization cues. So it seems to me that an Image Model Theory set-up would work well in the frequency region that matters most for sound image localization (soundstage size).

This is a really good point. IMT can be adjusted to have slightly different explanatory principles below the transition region. That's one objection down.

On my other two objections, my final comments about the lack of a perceptual model concern something I've been thinking about for a long time:

• Recordings can sound really good, even though they are compromised in a spatial sense.
• Speakers can sound really good, even when suboptimally positioned or designed.

IMT doesn't have enough explanatory power to address either point, I think.

Somewhat separately, I'd like to refer to architectural acoustics, where this modelling method used.

There are applications (for reverberation modelling) where a 3D model is used, and what's seen is a multiplication of images that accounts for later and later reflections. The initial model ends up exploding in complexity when you start drawing mirror images of virtual sources as well.

I think IMT acknowledges this complexity in the articles implicitly, but certainly doesn't dive into it. There is probably something here worth thinking about. Not sure exactly what at the moment.

 

[Curvature]

I think that would work very well. If possible, "shield" the little Genelecs, as seen from the listening area, such that none of their side radiation has a line-of-sight path to your ears and therefore arrives too early.

I was going to hide them behind the TV to ensure that only late reflections hit the listening position.

 

[Bwaslo]

I was going to hide them behind the TV to ensure that only late reflections hit the listening position.

Yeah, but you can probably save some $$ by using some waveguides with dome drivers rather than Genelecs. The requirements aren't that precise, and at least on mine only doing upper frequencies seems to work well enough. Oddly enough, the sound seems to get more full at lower midrange (where the guide has no output!) when the splash levels are set too high, psychoacoustic illusion I guess. But having a digital delay on the signal seems kind of important to keep the ambience uncorrelated with the mains.

 

[pma]

Speaking about room acoustics, to me the most valued publication is this one:

Acoustics of Small Rooms

Much time is spent working out how to optimize the acoustics of large rooms, such as auditoria, but the acoustics of small rooms and environments can be just as vital. The expensive sound equipment of a recording studio or the stereo in a car or living room is likewise rendered useless if the...

www.routledge.com

 

[krabapple]

ok so not exactly
a speaker he made according to his theories beat out (A) a linkwitz orion with no sub or EQ (lol) and (B) a terrible Beringer speaker

Except, it's not a terrible speaker. It's a champ in its category (budget 'monitor') , as is the passive version. Maybe you're prejudiced against the name?

in a test with 13 listeners. I'm no expert, but it seems to me they did a commendable job trying to produce a controlled test, so that's good.

The test unequivocally can't support this claim, however. And that's what I was asking about.

What speaker blind tests produce 'unequivocal' results? There's always someone who can equivocate.

 

[krabapple]

The BOSE 901 at those old times was not bad but also not perfect as far I remember. The backfired speaker sound depends much on the back surfaces whether damped or reflective. So the sound was not perfect compared with todays high quality speakers. For me to broaden the acoustical scene the distance between the stereo speakers can be increased without the need of early reflections. But anyway this all is subject for personal taste.

there's a limit to how far you can separate the front L/R before there's a 'hole' in the middle.

and AIUI 'broadening' is usually due to reflections from the side walls. Front wall reflection add 'depth'. But I could be wrong.

Thank you! Again giving credit where credit's due, AJ was using highly adjustable, fairly wide-band rear-firing drivers before me. (But I didn't copy him.)

as in, Florida AJ? I used to see him all over audio forums, fighting the good fight. Whatever happened to him?

 

[kyuu]

What speaker blind tests produce 'unequivocal' results? There's always someone who can equivocate.

He didn't say anything about the test producing unequivocal results (or not). He said that the test unequivocally cannot support the claim being made.

 

[Flaesh]

to me the most valued publication is this

I've been recommending this book to everyone for a long time. But just the other day I realized that the authors' last names mean "small" and "silent.)
But I don’t remember the book promoting +6 dB reflections and the like.

 

[MarkS]

you really like it when the soundstage width extends beyond the speakers. ... suppose you read Floyd Toole's book and learn that this happens because strong early same-side-wall reflections expand the Apparent Source Width in the direction of those reflections.

I'm not sure what Floyd Toole actually writes about this in his book, but this is definitely not the whole story.

Listen to out-of-phase pink noise in stereo on narrow directivity speakers far from side walls, and you will hear the sound coming from well beyond the left edge of the left speaker and the right edge of the right speaker.

This effect has nothing to do with wall reflections. Wall reflections actually reduce the effect (by messing up the left-right cancellations that are tricking your brain about where the sound is coming from), and so is less prominent with wide directivity speakers.

 

[Duke]

I'm not sure what Floyd Toole actually writes about this in his book, but this is definitely not the whole story.

Listen to out-of-phase pink noise in stereo on narrow directivity speakers far from side walls, and you will hear the sound coming from well beyond the left edge of the left speaker and the right edge of the right speaker.

This effect has nothing to do with wall reflections. Wall reflections actually reduce the effect (by messing up the left-right cancellations that are tricking your brain about where the sound is coming from), and so is less prominent with wide directivity speakers.

Image broadening because of out-of-phase signals is not the same thing as image broadening because of a strong sidewall reflection. Neither precludes the other.

 

[Thomas_A]

Phase effects should be a different thing. The acoustic mirror of a speaker in the direction of the reflection is another effect.

 

[Thomas_A]

Not sure about the back-wall firing stuff. It has been used for various reasons, one was using tweeter for diffuse sound to compensate for the usual loss occurring in the highest octave. It gives effects for sure, but I can't really see the depth unless signal is delayed (or speaker far out in the room). I still think it's best having frontal arriving sound as "pure" as possible, meaning the front wall should be acoustically invisible.

 

[MarkS]

Image broadening because of out-of-phase signals is not the same thing as image broadening because of a strong sidewall reflection. Neither precludes the other.

Yes, exactly. Your post seemed to imply that sidewall relection was the only source, and I was pointing out that there is another source. I'm glad that you agree.

 

[RexrothPigeon]

What speaker blind tests produce 'unequivocal' results? There's always someone who can equivocate.

You misunderstand. Go back and reread the claim I was responding to and then review the test. The test is not testing for that.

 

[Thomas_A]

If a model with ”they are here” there is a room extending behind the listening room and the front wall should be removed or acoustically invisible. There are still the reflections present in the recording and the real reflections of the sidewalls, roof and floor. But there are no reflections from the front wall. So with that model there is no use of backfiring speakers. Fixing cues that cause the event being a few meters behind the wall is the tricky one. One thing is to remove cues that disclose the speaker positions.

 

[gnarly]

In this model, speakers in a room can be thought of as a lamp in a room where every wall is covered in a full length mirror. Whilst a hall of mirrors will generate an almost infinite number of reflections, it is different with speakers. Only the first and second order reflections are important for imaging, because the other reflections die off and become more diffuse in time. It is argued that the position of these phantom speakers (my term, not his) is responsible for apparent soundstage width (ASW) - again my term and not his. All imaging takes place within the area bound by the two actual speakers and the six phantom speakers as shown.

It can be seen that the radiating pattern of the speaker, and the position of the speakers relative to each other and to the room, has a major influence on the strength of the phantom speakers.

I have a hard time putting much legitimacy into the idea of 'full-range' virtual speakers from boundary mirror images. Once frequencies polar radiations move away from onmi, it seems to me the lateral reflections, and especially the rearward reflections, probably are weak to the point of meaningless.

I've been taught in prosound sub-deployment classes, to visualize all the virtual subs that get formed with perpendicular boundaries.....due to the more-omni-than-not sub radiation. Which makes sense.
What doesn't make sense to me, is how can there be a virtual speaker in play, without near full strength radiation going directly towards the perpendicular boundaries (forming the mirror images.)

I dunno.....theory makes sense for low frequencies, but by the time we are in the top half of the spectrum, seems pretty bogus to me. Just my 2c thoughts

 

[Keith_W]

I have a hard time putting much legitimacy into the idea of 'full-range' virtual speakers from boundary mirror images. Once frequencies polar radiations move away from onmi, it seems to me the lateral reflections, and especially the rearward reflections, probably are weak to the point of meaningless.

Bingo! Excellent point! This model does not work for traditional non-omnidirectional loudspeakers. Monopoles have very little rearward radiation, so if anything the phantom speaker would be an extremely spectrally distorted copy of the speaker. It's like pointing a torchlight towards yourself and looking at the reflection of the torch in the mirror. You see the arse end of the torch, you don't see another source of light.

I was excited to read the paper when I first got my hands on it, but by the second day my sentiment had cooled considerably. Now I don't think that the model holds up unless speakers are specifically designed to exploit this effect.

 

[olieb]

I dunno.....theory makes sense for low frequencies, but by the time we are in the top half of the spectrum, seems pretty bogus to me. Just my 2c thoughts

Mirror idea makes sense, but of course the radiation pattern has to be taken into account. Like wenn you direct a torch to the listener, there will be not much light reflected in a side or back mirror. (You are seeing the back of the speaker in the mirror.)
And all this is only valid to the degree that the walls are flat and hard reflecting like a tiled brick wall.
In this case more than 6 images must be taken into account.
When absorption or diffusion takes place the "mirror" transforms into one build from frosted, tinted or matt glass.
And each generation of reflection will get weaker and weaker (fuzzier and fuzzier).

But all these reflections do not invalidate the law of first wave front. The phantom images do not sum like linear combinations, but on a "first come - creates the image" basis.
The reflections create interference (comb filtering), image modification (apparent source width, image shift/distortion, echos) and slow decay.
So I do not buy the theory but I am still amazed by the outcome of the "auditory scene shoot out".
What is going on?
How can this

create a more convincing scene than the competitors?

 

[Duke]

If a model with ”they are here” there is a room extending behind the listening room and the front wall should be removed or acoustically invisible. There are still the reflections present in the recording and the real reflections of the sidewalls, roof and floor. But there are no reflections from the front wall. So with that model there is no use of backfiring speakers. Fixing cues that cause the event being a few meters behind the wall is the tricky one. One thing is to remove cues that disclose the speaker positions.

Most loudspeaker paradigms probably do work best when the front wall is absorptive, but there are those which make deliberate use of the front wall. Dipoles would be the most common example. Ime in order for the rear-firing energy to be beneficial, it should have approximately the same spectral balance as the direct sound, and it should not arrive too soon.

Gary Eickmeier's approach of creating a deliberate (and arguably optimized) pattern of phantom sources is imo a fascinating innovation. It's not the paradigm I use, but that doesn't lessen my respect for what he has created.

Bingo!

... I don't think that the model holds up unless speakers are specifically designed to exploit this effect.

Bingo!

Yes, if I understand correctly, it's both the specific loudspeaker design and the specific placement in a room with suitable walls. It might be feasible to simulate Gary's model with two pairs of speakers, back-to-back, and with the rear-firing pair 6 dB louder than the front-firing pair. Might need to EQ out some of the rear-firing speaker's low end because it will wrap around and potentially over-reinforce the front-firing speaker's low end.

Mirror idea makes sense, but of course the radiation pattern has to be taken into account.

Agreed.

But all these reflections do not invalidate the law of first wave front.

Yes!

How can this
View attachment 483536
create a more convincing scene than the competitors?

I would say that, evidently, Gary's "outside the box" thinking produces the results he intended. Interesting orientation of the IMP speaker as well, very strong toe-in.

Can anybody here tell whether that's the original Orion, or the Orion+, or the Orion++?

 

[olieb]

I would say that, evidently, Gary's "outside the box" thinking produces the results he intended. Interesting orientation of the IMP speaker as well, very strong toe-in.

So it seems, though I would not call it out of the box thinking, more like extreme inside the box thinking. ;-)
As they say: You have to either swim against the tide or with it, but much faster than the others.
I redid the pic with his "design" where the back drivers are 6dB louder than the one facing the listener (indicated by darker color of the circles representing those). So the "images" would have a similar amplitude than the direct radiation in spite of having longer pathway.
Hard to fathom that this actually works as intended.