What exactly is "imaging" ?

[Tony L]

I am not factoring in room reflections at all. If the mono centre signal is coming from two speakers, it starts off equal amplitude from each speaker and in phase. If you have an ideal triangle setup like in @tuga Fig (a) a few posts ago, the sound from the R speaker will reach your R ear exactly at the same time and in phase with the sound arriving at your L ear from the L speaker. It is the sound crossing over round your head that takes a longer path, R speaker to L ear and vice versa. The path is several cm longer, which works out to be a half wavelength at about 2 kHz. This partly cancels the direct sound from the other speaker

Ok, I can now see the argument, but I rate it as utterly inconsequential and more a line from someone trying to sell me a centre speaker that I really do not want!

I am perfectly happy with two-channel stereo and personally view the distances between drivers and corresponding phase irregularities on typical multi-driver moving coil speakers as vastly more of an issue. I’d be arguing for a full-range phase accurate point-source long, long before bothering about this sort of stuff!

 

[davidsrsb]

I argue that this is why you don't want centre speakers, if the original mix was done in stereo. Trying to correct this issue is undoing what the studio team fixed in the first place

 

[adamdea]

This is another reason Linkwitz's approach of working with room reflections, instead of suppressing them, is a beneficial one, with stereo. Stereo is a trick that hacks the listener's natural aural processes, and delayed reflections in the playback space of output from the speakers are part of the hack. The cross-talk is simple to reason about in an anechoic chamber situation, but in a room, a lot more goes on!

The quote at https://pinkfishmedia.net/forum/threads/what-exactly-is-imaging.257921/page-8#post-4433852 is dense with meaning, it's worth (re-)reading carefully.

The first sentence of the quote is basically marketing for his approach and I dare say that many speaker designers would object very strongly. The rest is not really objectionable apart from the implication towards the end that everyone get stereo equally including newbs. I seem to remember that there is some evidence that it is a somewhat learned "skill". Aside from that, the fact that some people think that suppressing reflections is the key, while he wants to work with them demonstrates the point that once we get beyond left right panning, fairly general distance and sense of space, the results are likely to be very personal and room dependent and distantly (if at all) connected with features of the recording.

 

[darrenyeats]

Linkwitz was retired by then - linkwitzlab.com was non-profit. I know you didn't write it was commercial marketing, but I am making it clear it wasn't.

The main reason I'm banging on about Linkwitz is that tuga referenced him, and I know for a fact Linkwitz's views were not compatible with tuga's position. From email conversations I had with Siegfried, it became apparent to me, despite his famous and less famous achievements, he was a humble man and I think his motivation was altruistic.

 

[adamdea]

I argue that this is why you don't want centre speakers, if the original mix was done in stereo. Trying to correct this issue is undoing what the studio team fixed in the first place

I guess you had better hope they didn't carry out the whole process using headphones.

 

[Tony L]

The quote at https://pinkfishmedia.net/forum/threads/what-exactly-is-imaging.257921/page-8#post-4433852 is dense with meaning, it's worth (re-)reading carefully.

Interesting stuff, and well worth contrasting with say the writings of Paul Klipsch who I guess can be seen as the polar opposite. His view (which is well documented in various white papers ‘dope from hope’ letters etc, but I don’t have specific links to hand) is narrow/controlled directivity and corner location is the way to control room reflection as you remove any prospect of delayed sound reflection from behind the speaker, the room itself becomes the horn to a large degree, and the narrow directivity of the mid and treble horns gets the sound straight to the listener long before any room reflection.

I’m not saying it is right/wrong, but it goes to show there are many legitimate approaches. There is certainly something remarkably focused about a good horn system that sounds more like headphones in many ways, I suspect it is the high directivity and correspondingly lower/later room-splash. The exact opposite of omnidirectional speakers.

I guess we have at least four very distinct ‘schools’; panels/dipoles, corner horns, omnidirectional speakers, and conventional boxes. Some overlap and further variety within type, but there is certainly no accepted ‘right’ to my mind. It is still all up for grabs. I’m certainly not buying into any school or ideology on this one as there are so many solutions that work in their own way.

 

[tuga]

The main reason I'm banging on about Linkwitz is that tuga referenced him, and I know for a fact Linkwitz's views were not compatible with tuga's position.

Which position are you referering to, phantom image height and/or location in the vertical plane?

In Linkwitz's piece I mentioned earlier he makes reference to a (vague) "sense of height", but doesn't provide evidence or describe a mechanism:

2.3.4. Phantom source distance & space (page 14)


The loudspeakers are hidden when the real signals from left and right loudspeakers have merged with the phantom sources in a spatial continuum that is not hard-bounded by the loudspeakers. Spectrally similar room reflections allow the auditory scene to be dominated by the direct loudspeaker signals, which carry information about the recorded acoustic event in its spatial context. The auditory scene becomes 3-dimensional, located behind the loudspeakers, with depth, width and even some height. In many cases it can only be a miniaturization of reality, as heard from a distance. Indeed the volume control acts as the auditory scene‟s size and distance control and its program specific setting is critical to obtain a believable auditory scene, one that creates full enjoyment.

(...)

2.4.1. Perspective and distance of the reproduced Auditory Scene (pages 15 & 16)

When loudspeakers are designed properly and set up in a room in such a way that both, loudspeakers and room, are no longer recognized in the auditory scene, then it becomes possible to make meaningful comparisons between the auditory scene experienced at a live concert and its reproduction in the living room. Such comparisons are important to me when I evaluate loudspeakers, their setup and room influences. Seen from the “sweet spot” „A‟ in my room the two loudspeakers subtend a 600 angle, Fig. 14. The phantom source will essentially occupy this angle in front of me and it will not be closer than the distance to the real sources of sound, the two loudspeakers. In the concert hall the musicians, the sources of sound, are seen at a 600 angle from seat „X‟. Let us make a recording from this location with small omni-directional microphones placed on each side of the head, Fig. 11. Later, left and right microphone signals are fed to left and right loudspeakers without any further processing and listened to from location A. The signal streams from the two loudspeakers contain ILD and ITD cues for angular positions within the auditory scene, as well as some distance, elevation and HRTF cues.

The rear hemisphere and much of the side sounds that the microphones picked up will now be heard coming from the front, in addition to the orchestra. Thus there will be more hall sound in the reproduced auditory scene then was probably heard live. The direct-to-reverberant sound ratio has been reduced. The ear-brain perceptual processor performs a running correlation between left and right eardrum signal streams to locate sound sources in phantom space and to recognize its spatial attributes of width, depth, height and continuity. This is an immediate learning process as different sound sources illuminate parts of that space.

Amplitude and timing differences between the signals streams at the ears, their amplitude envelopes and spectrum are decoded in the brain to place phantom sources in locations between the loudspeakers. Stream content that cannot be mapped between the loudspeakers is lumped into left and right loudspeakers as monaural sources. It is not clear how the auditory scene is affected by this.

Distance and size of the auditory scene are playback volume dependent. The auditory scene moves closer with increasing volume, it becomes larger and more detailed. There is a limit to the maximum volume setting, when the perceived distance to the auditory scene becomes incongruous with its loudness and size. Listening from location „B‟ will not increase the maximum acceptable volume setting significantly. It will merge the auditory scene into the room with a loss of imaging precision and detail. Listening from location „C‟ is more immersive into the auditory scene and actually my preferred seat for this type of head-related recording.

In the concert hall location „Y‟ and the last four rows of chairs are underneath the balcony. From here the orchestra and hall are seen as through a very wide and open window. I am aware of being in a smaller space with a low ceiling listening out into a larger space. This is also audible when listening to a recording from this area of the hall. In general, a sense of height is captured with this simple recording technique, which adds to the realism of the auditory scene reproduced at „A‟. Location „Z‟ and closer to the orchestra produce a very unbalanced auditory scene because musicians are visually hidden and distances to them vary greatly. Commercial recordings in this hall typically use a large number of microphones hanging above different sections and instruments of the orchestra. A widely spaced pair above row „D‟ picks up hall sound. Recordings here have won prestigious awards.

 

[tuga]

Interesting stuff, and well worth contrasting with say the writings of Paul Klipsch who I guess can be seen as the polar opposite. His view (which is well documented in various white papers ‘dope from hope’ letters etc, but I don’t have specific links to hand) is narrow/controlled directivity and corner location is the way to control room reflection as you remove any prospect of delayed sound reflection from behind the speaker, the room itself becomes the horn to a large degree, and the narrow directivity of the mid and treble horns gets the sound straight to the listener long before any room reflection.

I’m not saying it is right/wrong, but it goes to show there are many legitimate approaches. There is certainly something remarkably focused about a good horn system that sounds more like headphones in many ways, I suspect it is the high directivity and correspondingly lower/later room-splash. The exact opposite of omnidirectional speakers.

I guess we have at least four very distinct ‘schools’; panels/dipoles, corner horns, omnidirectional speakers, and conventional boxes. Some overlap and further variety within type, but there is certainly no accepted ‘right’ to my mind. It is still all up for grabs. I’m certainly not buying into any school or ideology on this one as there are so many solutions that work in their own way.

Ultimately the balance between direct sound (recording) and reflections (listening-room sound) it's a matter of preference.
And then one has wide directivity box speakers which produce strong early reflections from side walls and dipoles that produce only late reflections from front and back walls, they produce a completely different effect and listening experience.

 

[darrenyeats]

Linkwitz talked about a 3D auditory scene that could extend somewhat beyond the speakers in left/right dimension and in height, with of course depth too. Your position isn't compatible with that tuga. Further he talked about the listening room's surfaces disappearing from the auditory scene, not sure where you stand with that.

 

[adamdea]

Linkwitz was retired by then - linkwitzlab.com was non-profit. I know you didn't write it was commercial marketing, but I am making it clear it wasn't.

The main reason I'm banging on about Linkwitz is that tuga referenced him, and I know for a fact Linkwitz's views were not compatible with tuga's position. From email conversations I had with Siegfried, it became apparent to me, despite his famous and less famous achievements, he was a humble man and I think his motivation was altruistic.

I'm sure you are right.

 

[tuga]

Interesting stuff, and well worth contrasting with say the writings of Paul Klipsch who I guess can be seen as the polar opposite. His view (which is well documented in various white papers ‘dope from hope’ letters etc, but I don’t have specific links to hand) is narrow/controlled directivity and corner location is the way to control room reflection as you remove any prospect of delayed sound reflection from behind the speaker, the room itself becomes the horn to a large degree, and the narrow directivity of the mid and treble horns gets the sound straight to the listener long before any room reflection.

I've uploaded the one piece by Klipsch that I have in my library, I wonder if this is the one you're referring to:

Symposium On Auditory Perspective
https://pdfhost.io/v/Fd7oql5CX_Symposium_on_Auditory_Perspective.pdf

 

[adamdea]

I guess we have at least four very distinct ‘schools’; panels/dipoles, corner horns, omnidirectional speakers, and conventional boxes. Some overlap and further variety within type, but there is certainly no accepted ‘right’ to my mind. It is still all up for grabs. I’m certainly not buying into any school or ideology on this one as there are so many solutions that work in their own way.

I guess you might add to that soffit mounted, which makes a lot of sense
and maybe the cardiod pattern/ B&O controlled direction thing too.

 

[darrenyeats]

Dipole is dipole technically speaking, but I think it's a mistake to judge dipoles as a whole by listening to panels. That's going by the Orions anyway - I'd like to hear also the LX521!

 

[camverton]

A single source straight in front of you will reach both ears at the same time for all frequencies.
Sources L and R will each reach one ear first and the other a bit later. The delay to the further ear reaches a half wavelength at about 2 kHz (depending on your head and details of the speaker angles)
This means that the far ear sound is out of phase with the direct sound from the other speaker and partly cancels it.
Toole is right on this effect
This means that a purist mono system will sound very different to a stereo system playing a mono recording

Very interesting and well explained, thank you. In practice I wonder how much it matters and to what extent different types of speaker are affected by this.

For a while I used meridian G68 processor with a 5.1 speaker setup, and found the trifield processing very effective in the way it stabilised the centre image by utilising a centre speaker. It even made the “box” speakers I was using at the time acceptable for me. After that I went back to Martin Logan’s and found they gave a much more realistic presentation than a pair of monitor speakers, albeit with the necessity to sit equidistant from each speaker, so I abandoned Trifield processing, five Martin Logan’s in a room being excessive even by my standards.

I then went back to Quads and found they were a good compromise but that soloists, whilst sounding realistic, had a slightly two dimensional feel about them, which for me hindered the illusion of having musicians in front of me.

Recently I’ve gone for German Physiks omnis and, for my requirements, they make for a very satisfactory presentation (I’m not interested in studying the recording so much as hearing music as if it were being played in front of me). The extraordinary thing with the German Physiks is that you don’t need to sit equidistant from each speaker to get a good centre image, and that, subject to the recording, the centre image of say a soloist isn’t any wider than with a pair of monitors. In fact I did a comparison with my LS50s and found that the omnis gave a better defined centre image of a soloist. Of course, the fact that they make use of the room may offend purists who demand absolute fidelity to the recording, but for those who want fidelity to the music performance they are worth a try.

Perhaps I just have funny shaped ears - I certainly seem to end up with funny shaped speakers!

 

[Tony L]

I've uploaded the one piece by Klipsch that I have in my library, I wonder if this is the one you're referring to:

Symposium On Auditory Perspective
https://pdfhost.io/v/Fd7oql5CX_Symposium_on_Auditory_Perspective.pdf

I can’t see a file there, but googling that title suggests it was just something he was sharing, not his own work. I’m certainly referring to either one (or more) of his white papers or his regular ‘Dope From Hope’ letters. This stuff used to be available from the Klipsch forum, but I can’t find it now. It was certainly his own thinking and struck me as very interesting.

 

[davidsrsb]

The effect of room reflections will be to reduce the extreme peaks and troughs from the simple two paths to two ears model. This is a heap of compromises and why there is room for horns, planars, boxes and omnis with no "best speaker in the world"
What I do know is that when I work in a RF test chamber, that also happens to be anechoic acoustically, that I get a headache from the acoustics being so "wrong" just for people talking

 

[darrenyeats]

The effect of room reflections will be to reduce the extreme peaks and troughs from the simple two paths to two ears model.

The reflections are delayed in time, so it's a bit more complicated. Answering how this will be perceived overall isn't simple, and could be affected by other context (e.g. off-axis FR). But yeah, I expect the perception will be different in room compared to anechoic.

 

[camverton]

The reflections are delayed in time, so it's a bit more complicated. Answering how this will be perceived overall isn't simple, and could be affected by other context (e.g. off-axis FR). But yeah, I expect the perception will be different in room compared to anechoic.

Completely different. I've worked in a few anechoic chambers of various types; horrible places and they feel very unnatural and not somewhere one can stand being in for long, and certainly not somewhere to listen to music. When it comes to listening to music in a room it is a matter of taste how much one wants the room/room reflections to influence the sound, but trying to go too far in removing the effects of the room leads to unrealistic reproduction for me.

 

[tuga]

The reflections are delayed in time, so it's a bit more complicated. Answering how this will be perceived overall isn't simple, and could be affected by other context (e.g. off-axis FR). But yeah, I expect the perception will be different in room compared to anechoic.

I listened to an amplified orchestra and also an amplified staging of Tosca outdoors. It was an awfully awkward experience...

 

[darrenyeats]

Amplified orchestra? Ugh!