Pink noise volume pan for evaluating stereo imaging

[Ghoostknight]

No, we can’t think of it like that. You might think of it like that but here in the sound science forum we think of it according to the facts/science, that’s why we’re here in the first place! This is particularly well researched area going back and has been since before the 1930’s. The number of “freq peaks you squeeze into the -1 to -10dB range” is NOT how we humans perceive overall volume, it’s where in the frequency spectrum those peaks are. See the Fletcher/Munson “Loudness Contours” from 1933!

at "reference levels" the fletcher/munson curve doesnt as much tho...

you are right its a variable i forget to mention in this but its not the only variable...

Dynamic range can play a significant role but only if you DON’T squeeze it into the -1 to -10dB range!

you didnt got what i was trying to say.... you get "loadness war optimized tracks" if you squeeze as much as possible into the upper db range close to 0db

Yes, the loudness war definitely can “make songs more pleasant to listen to”, that’s why there has been a loudness war in the first place!

loudness war messes with our perception of what is "good" .... "louder = better"

now if you listen to a loudness war track at 60db you might even prefer it over the same track non-loudnesswar-optimized at 80db


.... if you compare both versions at 80db reference level the case is pretty clear what "sounds best" tho if you include the playback volume it isnt clear.... imo thats why loudness tracks are so pushed.... many consumers dont listen at reference levels but actual lower levels... where loudness tracks "shine" so to speak...

 

[MrHaelscheir]

Genelec’s tend to have very good imaging, although that depends on exactly how they’re setup of course (EG. The room they’re in and their position within that room). I’m not quite sure what you mean by “out-there-ness” but whatever that is, along with the “proper lateral positioning” is not only defined by the speakers and speaker positioning, it’s defined in the first instance by how the recording has been mixed. Obviously the amount of width and depth and the instruments’ positioning within that stereo field are defined by the choices/preferences of the producer and is therefore likely to vary to some degree from recording to recording.

Again, I’m not quite sure what you mean here. Obviously no one has a listening room “with the scale or frontal distance of a concert hall”, nor do the recording studio control rooms where the recordings are mixed. Commercial orchestral recordings are designed to sound convincing in a typical consumer listening environment, not in an actual concert hall. However, that cannot be perfectly achieved of course, a sitting room (and a studio control room) is obviously not a concert hall. Therefore, we’re talking about creating an illusion defined by the subjective choices of the engineers/producer, which is of course dependent/limited by the mic’ing layout chosen and based on the control room system/acoustics the engineers/producer are using to monitor their mix. Only after all that does it come down to the fidelity of the playback system/environment in reproducing that illusion. But even given a hypothetically perfect consumer reproduction, the illusion can never be perfect, only subjectively convincing enough for some (hopefully most/nearly all) consumers.

G

Of course, the stereo width and the depth cues greatly depend on the recording and mixing methods employed, whereby I was focusing on the subjective assessment that some speaker systems amid their interactions with the room seemed to lack the channel-to-channel image width I would expect of certain recordings played through my Genelecs or binaural head-tracking, but at least had the sound roughly on a line between said channels, primarily occupying the phantom center, whereas a few notable speaker systems, not too fancy, successfully created the illusion of the stereo line being positioned in space a foot or so above the tweeters and midrange drivers and a few feet behind the speakers. "Out-there-ness" is the sense of the chance of some sound sources indeed imaging from beyond the front wall of the room or there being a large virtual space intersecting into rather than being confined by the physical listening space; e.g. if I were able to create a perfect personalized binaural recording of a concert hall from the middle of the fifth row and convincingly hear that imaging even if standing with my face against a wall.

I at the moment cannot set up my Genelecs to produce the illusion of imaging further than the line between the channels, though ceiling or wall reflections in my crappy setup could push some things like vocals above or leftward on the stereo line which I mainly find to be an annoyance. With my binaural head-tracking, for the grand majority of recordings that don't do weird time panning or phase tricks, I end up with a largely 1D stereo image between the two virtual channels, whereby even depth cues seem to image along that fixed line between the channels. From my DSP simulation of reflections using upmixing and delays, if the delay is small enough, the sound sources end up getting panned/stretched in the direction of the simulated reflection, but I hadn't been able to simulate increased depth/distance so long as my measured HRTF ITDs are tied to the 1.5 m distance; maybe I need to play with front wall reflections. i.e. If I play my DSP while facing out into the living room, I still feel like the stereo line is expectedly no further than 1.5 m in front of me due to the distance at which I measured my HRTF and ITDs. As such, I don't know whether I would rather attempt to play with speaker choice, placement, and room treatment to position the stage exactly where I want it (again, my personal experience of live classical concerts is primarily frontal with rare or insignificant perception of wall or ceiling reflections) versus just directly measuring an HRTF from 10 m away and applying binaural pans for said virtual channels, adding simulated reflections as I please.

Anyways, going back to the OP, the point was my proposing the use of a pink noise volume pan to in effect "trace out" the line through which most traditionally panned sound sources would be positioned on your playback system, revealing to you the "actual shape" of your "soundstage" as induced by your speakers + room, your DSP, or your headphones, and likewise whether certain frequency noise bands are traversing slightly different paths or are doing so at different rates (e.g. hearing a lower noise band pan toward the center while you can still simultaneously hear an upper noise band imaging further outward). After that, maybe some additional mixing tricks could induce localization of sound sources above, below, behind, or maybe in front of that line. So in those speaker rooms where I heard nicer orchestral imaging, I would expect a pink noise volume pan from the left channel to the right channel to "show" the pink noise blob as mainly imaging a foot or two above and behind the speakers and progressing toward the right, maybe having the treble localize more directly from the tweeters at the extremities.

 

[bigshot]

Imaging is as much dependent on the room as it is on the equipment… maybe even more so. It’s really hard to generalize. You have to get the system in the room and then start solving problems. You can adjust the scale of the soundstage by adding a center channel and increasing the distance between the mains, and increase depth by adding sides and rears; but everything affects everything else. It’s about finding a balance, not following a formula.

In my theater, I worked very hard to match the aural soundstage to the visual image. There was one particular scene in Billy Rose’s Jumbo that was particularly helpful. It depicts a proscenium stage where the camera is locked. A character speaks as he enters stage right, and he slowly crosses and exits stage left. It’s really easy to hear the handoff from speaker to speaker in the front. A similar thing for the rears is the multichannel mix for The Beatles’ Good Morning. A fox hunt crosses the room from front right to rear left. If you can get smooth transitions like that, the aural image is tight.

 

[gregorio]

Of course, the stereo width and the depth cues greatly depend on the recording and mixing methods employed, whereby I was focusing on the subjective assessment that some speaker systems amid their interactions with the room seemed to lack the channel-to-channel image width I would expect of certain recordings played through my Genelecs or binaural head-tracking, but at least had the sound roughly on a line between said channels, primarily occupying the phantom center, whereas a few notable speaker systems, not too fancy, successfully created the illusion of the stereo line being positioned in space a foot or so above the tweeters and midrange drivers and a few feet behind the speakers.

Mmmm. Assuming a standard 2 channel stereo mix, then the “few notable speaker systems” did not “successfully create the illusion of the stereo line being positioned in space a foot or so above the tweeters …”, they unsuccessfully reproduced the recording. There is no height information in a 2 channel stereo mix and no intention of creating the illusion that there is. So if you are hearing the stereo line above the speakers that is due to some fault in reproduction, maybe some timing/phase or even frequency issue between the drivers or more likely, some undesirable reflections from the ceiling that your brain is interpreting as the stereo line being higher than it actually is, or it might even be completely imaginary. Sounds/Instruments being perceived as behind or in front of the speakers is something we can (and virtually always do) manipulate though.

"Out-there-ness" is the sense of the chance of some sound sources indeed imaging from beyond the front wall of the room or there being a large virtual space intersecting into rather than being confined by the physical listening space; e.g. if I were able to create a perfect personalized binaural recording of a concert hall from the middle of the fifth row and convincingly hear that imaging even if standing with my face against a wall.

OK, I get what you mean by “out-there-ness” now. We can create that effect/illusion as far as width of the stereo image is concerned (for example by applying “shuffling”, although this effect is quite rarely employed) and also for depth, using a combination of tools I’ll explain below but of course we can’t and are not trying to create a personalised binaural recording of a concert with a standard 2 channel stereo mix. All we can do is very vaguely approximate that subjectively and hope it’s going to translate well enough to a consumer’s listening environment so that their brain fills in the gaps and they imagine they’re hearing a relatively accurate representation of the orchestra in a concert hall.

I at the moment cannot set up my Genelecs to produce the illusion of imaging further than the line between the channels, though ceiling or wall reflections in my crappy setup could push some things like vocals above or leftward on the stereo line which I mainly find to be an annoyance.

Clearly you have some acoustic issues and monitors such as Genelecs that are particularly good at accurately reproducing stereo positioning/imaging are going to be particularly badly affected by room/acoustic problems. I presume you’re using some near-field Genelecs? In which case, they are designed to be used in a near-field environment, EG. A listening position about 1-2m away from the monitors and placing them at least that distance away from the rear and side walls of your room.

With my binaural head-tracking, for the grand majority of recordings that don't do weird time panning or phase tricks, I end up with a largely 1D stereo image between the two virtual channels, whereby even depth cues seem to image along that fixed line between the channels.

Again, I don’t really understand. Binaural head-tracking obviously doesn’t work using speakers, even near-field speakers/monitors, are you talking about “binauralising” a standard 2 channel stereo mix using headphones rather than speakers? If not, the results of that are going to be highly variable and could indeed be perceived as a largely 1D stereo image.

From my DSP simulation of reflections using upmixing and delays, if the delay is small enough, the sound sources end up getting panned/stretched in the direction of the simulated reflection, but I hadn't been able to simulate increased depth/distance so long as my measured HRTF ITDs are tied to the 1.5 m distance; maybe I need to play with front wall reflections. i.e. If I play my DSP while facing out into the living room, I still feel like the stereo line is expectedly no further than 1.5 m in front of me due to the distance at which I measured my HRTF and ITDs.

It seems you are trying to apply HRTF/ITD to speaker reproduction? That’s rarely if ever going to work because a standard 2 channel stereo mix already contains reflections and processing that accounts for how we hear in an acoustic space. The engineers/producer obviously have a HRTF and are subjectively mixing according to what they are hearing/monitoring. You cannot add your own to that because your speakers are not isolated from each other (as HP earcups are). Maybe I’m misunderstanding what you’re trying to do?

As such, I don't know whether I would rather attempt to play with speaker choice, placement, and room treatment to position the stage exactly where I want it (again, my personal experience of live classical concerts is primarily frontal with rare or insignificant perception of wall or ceiling reflections) versus just directly measuring an HRTF from 10 m away and applying binaural pans for said virtual channels, adding simulated reflections as I please.

The way we create depth when mixing is to manipulate volume, HF roll-off, reflections and compression. We can do this artificially, using reverb units, EQ, fader level and compressors or we can do it through microphone layouts/positioning and then panning, time adjusting and balancing or most commonly, some combination of the two approaches. Most importantly, depth is not created by these effects, it is created by the contrast of them. Eg. Merely adding reverb, HF roll-off, etc., will still result in a “1D line”, it’s the contrast of having different reverb, HF roll-off, time off-set, etc., between the different instruments/sounds in the mix which creates depth. For example, applying the same reverb, the same level, HF roll-off, time offset, etc., to say both a violin and a trumpet will result in them both being perceived as the same depth/distance (the same 1D plane) but applying more compression, less HF roll-off, less reverb and more level to only the violin, will result in it being perceived as sounding closer than the trumpet and now we no longer have the perception of a 1D plane, we would perceive “depth”.

Anyways, going back to the OP, the point was my proposing the use of a pink noise volume pan to in effect "trace out" the line through which most traditionally panned sound sources would be positioned on your playback system, revealing to you the "actual shape" of your "soundstage" as induced by your speakers + room …

Using pink noise will only trace out the line between your speakers and in some cases extreme issues with your room acoustics, it will not reveal the actual shape of your (perceivable) soundstage. This is because adding reverb to pink noise just results in slightly louder pink noise, adding pink noise to pink noise with a HF roll-off results in pink noise with slightly more bass, adding quieter pink noise to pink noise results in slightly louder pink noise. In other words, pretty much all those things that define our perception of depth in a mix (or in the real world) simply don’t work with pink (or white) noise, all we perceive is a slightly louder or quieter pink noise or pink noise with a somewhat different freq balance, not any positional depth/distance cues. Again, I’m not sure I’ve correctly understood/interpreted what you’re trying to say?

G

 

[bigshot]

Usually, you want the midrange and tweeters on level with your ears at the standard listening position. You don’t want them higher than that. Bass is fine down low near the ground because it flows along the floor filling the room. It isn’t nearly as directional.

If you’re hearing vertical content and you aren’t listening to Atmos, it’s probably a trick your brain and ears are playing on you. It wouldn’t be intentional on the part of the sound engineer.

 

[MrHaelscheir]

undesirable reflections from the ceiling

Yes, I was assuming that to be the main contributor to the raising of the stereo image. In the context of the speakers not being tall enough to image the stage where I want it should there have been ceiling absorption, then in this room, the subjective effect that disembodied the perceived stage from the speakers producing it was to me advantageous in this context, whereby I would of course ideally have more room treatment and position the speakers higher. The point is that I used to assume that the only "correct" or "good sounding" imaging must be on a line between the channels.

Clearly you have some acoustic issues...

I certainly do, as in I live in a situation where I currently have no choice/space but to position my Genelec 8341As (primarily an overkill choice for DIY HRTF measurements with the primary goal of headphone reproduction) to the left and right of my desktop's dual-screen monitors amid a computer centered at the right half of a wide living room, my sitting about 1 m from each channel; I do not yet have access to a proper listening room where I can properly position and treat things, so I was just describing how I was hearing things in this poor context.

Again, I don’t really understand. Binaural head-tracking obviously doesn’t work using speakers... are you talking about “binauralising” a standard 2 channel stereo mix using headphones rather than speakers?

Yes. For context, I am talking about https://www.head-fi.org/threads/anyone-into-crossfeed.961533/page-3#post-17950791 (post #32 and on). Think "DIY quasi-anechoic Smyth Realiser) or simulating/binauralizing stereo channels in a maximally treated room through headphones. All my "DSP" is regarding playback through headphones to simulate frontal speaker imaging. Plug-ins such as SPARTA AmbiRoomSim allow me to add channels to which I can route delayed audio within the Reaper DAW. It is in this context that I perceive the virtual speakers as being no more than 1.5 m away even if I am facing out into my living room with the far wall being well beyond 1.5 m away; either I am "imagining" this proximity, or the HRTF I captured with in-ear mics using https://www.earfish.eu/ and how I am applying it to headphone binauralization with SPARTA AmbiRoomSim (with reflection simulation turned off) and CroPaC is simulating the actual distance at which I measured the HRTF rather than projecting all binauralized sound sources out at infinity. It is one thing to have depth cues in a recording, but assuming volume, distortion, and angle are controlled for and one's head were in a vice, would it not be possible to judge whether a stereo pair in an anechoic chamber were positioned closer or further due to differences in ITD among other HRTF cues?

The DIY Earfish HRTF capture setup in question. In theory, my DSP is simulating what it would be like to listen to music using a Genelec stereo pair in my backyard with the reflections mostly truncated from the impulse response (the Earfish capture used around a 13 ms window).

Using pink noise will only trace out the line between your speakers and in some cases extreme issues with your room acoustics, it will not reveal the actual shape of your (perceivable) soundstage.

Here, I was talking about the latter case for those few speakers I heard at the show. I am not saying that the speakers were reproducing height information; I was saying that their dispersion plus ceiling reflections and other seemed to raise the stereo image to what I found to be a more favourable position given those speakers' height, whereby if I was hearing music imaging along that raised line, I would expect that a pink noise volume pan would also be perceived as imaging along that ("erroneously", as you may call it) raised line, and vice versa. Likewise, I would expect that listening to how a pink noise volume pan images through headphones would likely predict the "stage" upon which normally mixed music would image upon if played through those headphones.

One of the favourable speaker systems in question. I am not complimenting the speakers themselves or the gear so much as the collective setup happening to have produced a subjectively favourable presentation of recordings.

Now, would it be plausible for some individuals to be "immune" to intentionally mixed depth cues such that even said reverb or sound sources that were mixed to "sound" further are still perceived as being imaged along a 1D line?

Usually, you want the midrange and tweeters on level with your ears at the standard listening position. You don’t want them higher than that. Bass is fine down low near the ground because it flows along the floor filling the room. It isn’t nearly as directional.

If you’re hearing vertical content and you aren’t listening to Atmos, it’s probably a trick your brain and ears are playing on you. It wouldn’t be intentional on the part of the sound engineer.

In my case, I would only want to position the midrange and tweeters a bit higher to simulate my perspective from the fourth to seventh row of a concert hall where the strings are in fact a bit elevated relative to one's eyes and ears. And as stated, the "vertical content" being described is merely a raising of the to-me 1D stereo line which I am attributing to untreated ceiling reflections and other. The only "intent" here would regard how the speaker's owner/presenter positioned things in order to create a favourable presentation out of that room's acoustics.

 

[bigshot]

Raising the soundstage is simple. Just raise the speakers up on stands. Alternately, you could raise a center speaker a bit higher than the mains. I have this in my theater to anchor the voices to the screen. I have a high peaked roof with beams, so I don’t get any unwanted reflection from the ceiling.

 

[gregorio]

at "reference levels" the fletcher/munson curve doesnt as much tho...

There aren’t any reference levels. Assuming a level of 60 phon (60dB SPL at 1kHz) then the original Fletcher/Munson contours indicate a difference of about 20dB throughout the audible spectrum, while at 80phon it’s around 15dB. However as mentioned, there has been considerable research in this area since then, using more precise equipment than existed in 1933 and the original Fletcher/Munson contours have been revised several times. The difference at 60phon is now known to be about 50dB and at 80phon around 40dB. I’m not sure how you consider a 40 - 50dB difference to not be doing much?

you are right its a variable i forget to mention in this but its not the only variable...

Sure, it’s not the only variable but with up to a 40 - 50dB difference it’s obviously a very significant variable and in some cases easily the most significant variable but you were not aware and did not consider it at all, you instead discussed all sorts of other variables, some of which make little or no audible difference. So as I stated, an example of a relatively huge difference of which you are not aware/did not consider but you claim you can hear and seriously consider other differences of just a fraction of a dB, that are also above the frequency range of human hearing!

I’m not going to get any further into your loudness war claims, it’s not a simple matter, it’s clear you don’t really understand it and it’s completely off topic anyway.

G

 

[Ghoostknight]

There aren’t any reference levels. Assuming a level of 60 phon (60dB SPL at 1kHz) then the original Fletcher/Munson contours indicate a difference of about 20dB throughout the audible spectrum, while at 80phon it’s around 15dB. However as mentioned, there has been considerable research in this area since then, using more precise equipment than existed in 1933 and the original Fletcher/Munson contours have been revised several times. The difference at 60phon is now known to be about 50dB and at 80phon around 40dB. I’m not sure how you consider a 40 - 50dB difference to not be doing much?

Sure, it’s not the only variable but with up to a 40 - 50dB difference it’s obviously a very significant variable and in some cases easily the most significant variable but you were not aware and did not consider it at all, you instead discussed all sorts of other variables, some of which make little or no audible difference. So as I stated, an example of a relatively huge difference of which you are not aware/did not consider but you claim you can hear and seriously consider other differences of just a fraction of a dB, that are also above the frequency range of human hearing!

I’m not going to get any further into your loudness war claims, it’s not a simple matter, it’s clear you don’t really understand it and it’s completely off topic anyway.

G

its not like i have never considered it

well to make a better comparison of my point, lets take the 2-4khz where humans are most sensitive, now you can either put one single sine wave in this range or multiple .... multiple will sound "louder" specially with musical content

there are multiple variables and loudness war isnt that easy to understand

 

[MrHaelscheir]

Raising the soundstage is simple. Just raise the speakers up on stands. Alternately, you could raise a center speaker a bit higher than the mains. I have this in my theater to anchor the voices to the screen. I have a high peaked roof with beams, so I don’t get any unwanted reflection from the ceiling.

Indeed, that is the recommended way assuming access to ceiling treatment and others, whereby the point of my posts was to say this was my first time experiencing similar being achieved using the room's reflections, again not to say that that method is to be recommended. I am nonetheless curious of the limitations on speaker placement or room treatment for creating the illusion of projecting the image well beyond the 1D line between the speakers if not beyond the front wall.

 

[gregorio]

I certainly do, as in I live in a situation where I currently have no choice/space but to position my Genelec 8341As (primarily an overkill choice for DIY HRTF measurements with the primary goal of headphone reproduction) to the left and right of my desktop's dual-screen monitors amid a computer centered at the right half of a wide living room, my sitting about 1 m from each channel; I do not yet have access to a proper listening room where I can properly position and treat things, so I was just describing how I was hearing things in this poor context.

Ah, OK. I tested those out at a friend’s studio a couple of years ago. For what it’s worth, for a mid size nearfield monitor the Genelec 8341A’s are IMHO the best on the market. Certainly the best I’ve ever heard or tested, so good choice! Even in somewhat poor acoustic circumstances they can get fairly reasonable results with the GLM kit. There are better/more comprehensive room correction systems on the market but the GLM is still better than virtually all the consumer ones. Nevertheless, even the very best systems have their limits and can’t overcome particularly poor acoustics. Assuming you’re using it, looks like you’ve exceeded it’s abilities.

For context, I am talking about https://www.head-fi.org/threads/anyone-into-crossfeed.961533/page-3#post-17950791 (post #32 and on). Think "DIY quasi-anechoic Smyth Realiser) or simulating/binauralizing stereo channels in a maximally treated room through headphones. All my "DSP" is regarding playback through headphones to simulate frontal speaker imaging.

Right, yes, I did misunderstand. I’m not personally familiar with that hardware or software you mentioned, so I’m out of my depth regarding any of that.

I was saying that their dispersion plus ceiling reflections and other seemed to raise the stereo image to what I found to be a more favourable position given those speakers' height, whereby if I was hearing music imaging along that raised line, I would expect that a pink noise volume pan would also be perceived as imaging along that ("erroneously", as you may call it) raised line, and vice versa.

Not necessarily, I again raise the concern that pink noise + reflections is just pink noise with a slightly different frequency balance, rather than pink noise with discernible height or depth cues. One can of course discern lateral positional cues based on level (ILD), say outputting the pink noise to just the left speaker but not based on timing (ITD), unless you use very short repeated bursts of pink noise (say a couple of milli-secs or less). In the case of HPs, I would expect the results to be highly variable from person to person, depending on how their individual perception interprets various slightly different frequency balances.

Now, would it be plausible for some individuals to be "immune" to intentionally mixed depth cues such that even said reverb or sound sources that were mixed to "sound" further are still perceived as being imaged along a 1D line?

With speakers? Assuming they didn’t have some fairly serious hearing impairment then “no”, I don’t see how anyone could be immune to it. It is however possible that some might not be consciously aware of it, especially if the mixed depth cues are quite subtle. Some modern mixes of certain genres have significantly less intended depth than mixes of other genres or compared to older mixing styles. This doesn’t really affect orchestral recordings though. But even in these cases, someone with normal hearing can consciously discern those depth cues, although if they’re unused to listening to music recordings analytically they may need some training. I’ve had to do that in the past with brand new students who hadn’t thought about or identified reverb previously.

If you’re talking about with HPs, then that’s another story, yes, it is plausible. One can effectively be immune to depth and height cues in standard 2 channel stereo mixes, as the brain tries to make sense of positional cues designed for speaker reproduction and fails. Having a visual reference commonly solves/improves this but sound on it’s own, even with repeated exposure and training, many will still effectively be immune. Although not directly related, a potentially interesting fairly recent finding is that if we have a binaural mix designed with a HRTF that is different enough to a particular listener’s HRTF that they can’t make sense of it (sounds are in the wrong location or the soundfield is just 1D), they can, with repeated exposure and training, acclimatise to that “wrong” HRTF and make sense of it, furthermore they can also still make sense of a binaural mix designed with their own, “correct” HRTF. This has quite a few implications because it means we can store more than one HRTF “model” in our brains and switch between them. It might also make the professional testing of generic HRTFs more problematic.

G

 

[MrHaelscheir]

With speakers? ...

Mainly with speakers, and if regarding headphones, only after a personalized speaker HRTF and mix has been applied. I am not talking about literally not hearing or being able to notice the depth cues or reverb so much as despite something like an off-stage trumpet clearly sounding further, the trumpet is independent of that timbre and reverb still perceived as imaging from the 1D line between the speaker channels. Basically, it still feels like looking at a distant mountain within a 2D photo or video as opposed to the depth effects possible with VR. Basically, whether an individual can be "too analytical" so as to hear through the illusion. Else I have yet to experience a well-set-up speaker system that convincingly revealed "layering" within a recording rather than my only perceiving a 1D sonic image.

 

[bigshot]

The amount of blurring of the image because of poor room acoustics and speaker placement is a pretty wide range. I generally think in terms of optimizing the image, and how well a room and speakers can present the soundstage. That’s a narrower range.

If you’re looking for scale of the image, near field monitors aren’t as good as full size speakers in a good size room.

The Culshaw Ring cycle contains lots of off stage horns that sound distant. It’s mostly a matter of secondary depth cues, particularly reverb. That isn’t difficult to reproduce. I think the realism probably depends as much on the quality of the overall sound as it does the sound of the horns themselves. Your mind is taking the pieces and assembling them and interpreting the illusion.

If you’re looking for layers of sound indicating distance, a multichannel system is what you’re going to want to look for, not two channel. Depth is what the extra speakers add,

 

[gregorio]

I am not talking about literally not hearing or being able to notice the depth cues or reverb so much as despite something like an off-stage trumpet clearly sounding further, the trumpet is independent of that timbre and reverb still perceived as imaging from the 1D line between the speaker channels. Basically, it still feels like looking at a distant mountain within a 2D photo or video as opposed to the depth effects possible with VR. Basically, whether an individual can be "too analytical" so as to hear through the illusion.

I think I understand. What defines a trumpet as being distant/off-stage is the loss of volume, an additional loss of high frequency (due to air absorption) and the various/numerous parameters of the reflections (reverb). If you get all those right by artificial means or by mixing the right balance of mic recordings then there is no difference between that and the real thing, so the brain of course can’t tell any difference. In practice though you can rarely if ever get all those parameters exactly right, so it’s a question of how close or how convincing you can’t get, how good your listening skills and how good your monitoring system. There certainly are circumstances where I/we can hear the imperfections and therefore “through the illusion”, in fact typically quite a few such circumstances but no one at home will notice, even if they have a much better than average system. I’ve spent decades manipulating every parameter of reverb, so I can recognise them far better than pretty much any consumer, plus I work in monitoring environments much more analytical than almost any consumer and in addition, I know the practicalities of recording and mixing and therefore where I’m more likely to find weaker illusions.

Else I have yet to experience a well-set-up speaker system that convincingly revealed "layering" within a recording rather than my only perceiving a 1D sonic image.

Are you saying that you’ve only ever experienced a 1D sonic image with speakers, that you’ve never perceived any depth, presence or layering? I think I must have misunderstood this bit because even a pretty crappy speaker setup will reveal at least some depth/distance (depending on the recording of course), albeit not accurately or very well?

G

 

[MrHaelscheir]

Are you saying that you’ve only ever experienced a 1D sonic image with speakers

So far, it seems, but I have never spent a proper amount of time with a proper system. I can hear the reverb and the timbral cues, but they are largely like "distant mountains on a 2D painting". I will need more time with a system to tell whether I can for example clearly hear a sense of the woodwinds or trumpets actually being further back on the stage than the strings etc. I can't say whether this experience has anything to do with my hearing largely identical image placement and distance through virtually any headphone largely independent of cup size and driver distance, or with how the HD 800 S's tuning can't "fool" me into hearing a super "wide" soundstage beyond the drivers.