Early horizontal reflections - isn't there a price to be paid?

[ctrl]

I say that it reduces realism when reproducing recordings of classical music which in their documental approach have capture the ambience or acoustic cues of the original event. If you add your room's acoustic footprint on top you get confusion. You will get more immersiveness and spaciousness which can be benefical with close-mic'ed studio produced music.

Wouldn't say that when listening to live recordings of classical concerts there is "confusion" due to the reflections on the recording and the early-reflections of the listening room, because the delay of the reflections is too different - listening room 3-10ms, concert hall >25ms.

But you are right with saying there will be "more immersiveness and spaciousness" because of the additional early reflections.
Is this always a bad thing?

And at least in concert halls, there is another advantage of strong early-reflections - increased dynamics.
Very interesting on this topic is a study with the title Concert halls with strong lateral reflections enhance musical dynamics
It says:

The concert hall conveys orchestral sound to the listener through acoustic reflections from directions defined by the room geometry. When sound arrives from the sides of the head, binaural hearing emphasizes the same frequencies produced by higher orchestral-playing dynamics, thus enhancing perceived dynamic range. Many studies on room acoustics acknowledge the importance of such lateral reflections, but their contribution to the dynamic responsiveness of the hall has not yet been understood.
Because dynamic expression is such a critical part of symphonic music, this phenomenon helps to explain the established success of shoebox-type concert halls.

This enhancement of the dynamics by lateral reflections cannot be included in the live recording of the concert.
But by early reflections of the listening room this effect could be partially recreated and the recording could be closer to the "original"?

At least the issue of early-lateral reflections does not seem to be as simple as some like to present it by proclaiming that absorbing all early-reflections is the best way to listen.
As conclusion the study says:

The present findings suggest that communication of the dynamic information in the music to the listener is augmented in rooms that provide early lateral reflections. A room that improves the transmission of musical information (35), such as orchestral dynamics, is expected to provide a more powerful musical experience to the listener.
Concert halls that benefit most from this phenomenon include classic shoebox-shaped rooms as well as certain modern designs (16). These findings establish a connection between the success of venerable historic shoebox-type concert halls, such as Vienna Musikverein, Berlin Konzerthaus, Amsterdam Concertgebouw, and Boston Symphony hall (ref. 10, p. 496), and binaural dynamic responsiveness.

 

[andreasmaaan]

Would you agree, in theory at least, that once the electric signal has been transduced into sound waves the highest accuracy or fidelity is achieved by getting those waves to reach the listener's ears without modification?

Kind of/arguably. Could elaborate a bit further? For example, should these waves arrive at the listener's ears with or without interaural crosstalk?

Anyway, things like soundstage, spaciousness and envelopment are mostly an audiophile thing, and a matter of preference at that.

Completely agree they are a matter of preference. What do you mean though that they're "an audiophile thing"?

In my view they're a complement or a surrogate for the lack of visual cues, and also an enhancer of the listening experience.

I don't see them as a complement/surrogate for the lack of visual cues, but rather as a complement/surrogate for the deficiencies inherent to stereo recording and reproduction.

 

[RayDunzl]

Would you agree, in theory at least, that once the electric signal has been transduced into sound waves the highest accuracy or fidelity is achieved by getting those waves to reach the listener's ears without modification?

That would be nice (I suppose), but it isn't going to happen in a room.

Here's an electrical source (quick repeated log sweep) , and a microphone pickup (my Electric Ear) at the listening position.

They are time aligned, using some impulses to attempt to match "timing".

No EQ since these are JBL LSR 308 (a relatively "approved" speaker) and should be pretty flat up close.


Time alignment:

Big Picture:

Bass is late because the sweep starts below the speaker's low frequency capability.

Fortunately, I think our ears are a bit forgiving of inexactitude and our brains good at reimagining what gets presented to it.

 

[tuga]

Wouldn't say that when listening to live recordings of classical concerts there is "confusion" due to the reflections on the recording and the early-reflections of the listening room, because the delay of the reflections is too different - listening room 3-10ms, concert hall >25ms.

But you are right with saying there will be "more immersiveness and spaciousness" because of the additional early reflections.
Is this always a bad thing?

And at least in concert halls, there is another advantage of strong early-reflections - increased dynamics.
Very interesting on this topic is a study with the title Concert halls with strong lateral reflections enhance musical dynamics
It says:


This enhancement of the dynamics by lateral reflections cannot be included in the live recording of the concert.
But by early reflections of the listening room this effect could be partially recreated and the recording could be closer to the "original"?

At least the issue of early-lateral reflections does not seem to be as simple as some like to present it by proclaiming that absorbing all early-reflections is the best way to listen.
As conclusion the study says:

Spaces for live music production have different requirements to spaces for the reproduction of recorded music.

Stereo is unable to recreate the original soundfield: both direct and reflected sound come from the same 2 sound sources.
Using side walls to reflect both direct and reflected sound which should not be coming from the front of the listener in the frist place is making a mash of it.

 

[tuga]

Kind of/arguably. Could elaborate a bit further? For example, should these waves arrive at the listener's ears with or without interaural crosstalk?

Unfortunately cannot avoid interaural crosstalk by physical means unless we use headphones which in itself is a very unnatural listening experience.
A panel between the speakers (preferably with binaural recordings) is the closest one can get but not particularly natural or pleasant either.
I tried it for a few weeks with a seat cushion some years ago.

Without the panel the reflection of a speaker from the opposite wall only adds to the confusion (I am referring to the overlaying of the recorded cues with the listening room's acoustic footprint).

Completely agree they are a matter of preference. What do you mean though that they're "an audiophile thing"?

Not a single one of my "normal" friends or acquaintances is the least worried about it, even the music lovers amongst them.

I don't see them as a complement/surrogate for the lack of visual cues, but rather as a complement/surrogate for the deficiencies inherent to stereo recording and reproduction.

They are that too in a way. Not everyone enjoys them beause they're not particularly good at it. Two wrongs don't make a right.

 

[andreasmaaan]

A panel between the speakers (preferably with binaural recordings) is the closest one can get but not particularly natural or pleasant either.
I tried it for a few weeks with a seat cushion some years ago.

Actually, I tried a slightly more sophisticated version of this and thought it sounded superb (in an otherwise quite reflective room, mind you).

Without the panel the reflection of a speaker from the opposite wall only adds to the confusion (I am referring to the overlaying of the recorded cues with the listening room's acoustic footprint).

The question I was really interested in your opinion on was whether the direct sound from the R speaker arriving at the L ear (quite apart from any reflections) is necessary for the highest fidelity in your view? Or, conversely, whether the highest possible fidelity would necessitate its removal?

Not a single one of my "normal" friends or acquaintances is the least worried about it, even the music lovers amongst them.

Ah yes, that's true. But the same must be said about literally every aspect of sound reproduction, not just the question of dispersion/reflections

They are that too in a way. Not everyone enjoys them beause they're not particularly good at it. Two wrongs don't make a right.

Of course not. But reflections can subjectively mitigate some of the unpleasantness resulting from stereo recording/reproduction, IMO/IME (obvs this is not everybody's experience).

 

[ferrellms]

I have noted that when @amirm reviews a speaker that has good off-axis response - or off-axis response that helps improve the overall estimated in-room response - he writes something like, "off-axis looks good so don't go absorbing the side reflections," or "the speaker seems designed to use reflections to tame the treble so don't go crazy with wall absorption and let the speakers use the room to create a larger image."

I understand the logic of this, but in my system - admittedly just one data point - I find that while limited side-wall absorption narrows the soundstage a little bit, it also improves imaging.

I have a single 2x4 foot, 2" thick panel on each side wall, more or less at the first horizontal reflection point. After reading some of Amir's reviews, I have tried removing the panels. My immediate impression is a slightly wider soundstage, but a more-than-slight reduction in soundstage precision. Without the panels there is a bit more ambience and "air" around the music to go with that perception of wider soundstage - although I detect that mostly when I put the panels back and then I hear the sound again without that bit of added air/ambience.

The effect without the panels has a pleasant aspect to it, but I find myself missing the extra bit of tightness and soundstage precision in the mids and treble when I have the panels on the walls. Something about the imaging just seems to lock in when the panels are there. I don't experience the difference as extreme, but I would not call it subtle either - it's immediately obvious to me, with any musical style or recording I might choose, at any reasonable listening volume I might choose.

This is of course just my subjective experience, and I while I really don't think I'm imagining the difference, I am open to a well-reasoned argument for confirmation bias if someone thinks what I'm saying can't be true. I'm also open to the argument that the increased soundstage width is merely additive and that the soundstage precision is not actually reduced by removing the panels - but again, that doesn't really square with what I think I'm hearing.

At any rate, I'd love to hear others' experiences and/or insights about this.

Thanks!

I have found the same result. Early reflections do smear the image. See what Ethan

I have noted that when @amirm reviews a speaker that has good off-axis response - or off-axis response that helps improve the overall estimated in-room response - he writes something like, "off-axis looks good so don't go absorbing the side reflections," or "the speaker seems designed to use reflections to tame the treble so don't go crazy with wall absorption and let the speakers use the room to create a larger image."

I understand the logic of this, but in my system - admittedly just one data point - I find that while limited side-wall absorption narrows the soundstage a little bit, it also improves imaging.

I have a single 2x4 foot, 2" thick panel on each side wall, more or less at the first horizontal reflection point. After reading some of Amir's reviews, I have tried removing the panels. My immediate impression is a slightly wider soundstage, but a more-than-slight reduction in soundstage precision. Without the panels there is a bit more ambience and "air" around the music to go with that perception of wider soundstage - although I detect that mostly when I put the panels back and then I hear the sound again without that bit of added air/ambience.

The effect without the panels has a pleasant aspect to it, but I find myself missing the extra bit of tightness and soundstage precision in the mids and treble when I have the panels on the walls. Something about the imaging just seems to lock in when the panels are there. I don't experience the difference as extreme, but I would not call it subtle either - it's immediately obvious to me, with any musical style or recording I might choose, at any reasonable listening volume I might choose.

This is of course just my subjective experience, and I while I really don't think I'm imagining the difference, I am open to a well-reasoned argument for confirmation bias if someone thinks what I'm saying can't be true. I'm also open to the argument that the increased soundstage width is merely additive and that the soundstage precision is not actually reduced by removing the panels - but again, that doesn't really square with what I think I'm hearing.

At any rate, I'd love to hear others' experiences and/or insights about this.

Thanks!

Many prefer the tighter, more precise imaging and transient response of reduced early reflections. I certainly do, as do recording engineers. Evidently, Toole found that listeners prefer a more spacious sound and that lateral reflections are beneficial. While this may be true for some recordings and rooms, the artificial spaciousness of early reflections may be pleasing to some, but is coloration caused by the room. Ethan Winer has a good post on this in this forum from a few years back.

 

[vavan]

It's why I can't do center channels. I know the sounds is coming right from the speaker, and since I can only fit it slightly below the screen, it drives me nuts.

Just this week I tried setting up a center speaker again and hated it

sometimes additional front heights and virtual center channel uplift in AVR may be of help

 

[napilopez]

I'd still like further investigation about where room treatment is applied. Front, ceiling, back reflections more obviously and uncontroversially lead to coloration, but I'm not sure I've ever seen a room that is treated only in these regions while leaving the sidewalls bare. I suspect this would be a good compromise.

sometimes additional front heights and virtual center channel uplift in AVR may be of help

Yep, I've used that before, albeit not with a center. I recently moved so I havent set up the atmos speaker yet, but maybe that'll do the trick.

 

[tuga]

The question I was really interested in your opinion on was whether the direct sound from the R speaker arriving at the L ear (quite apart from any reflections) is necessary for the highest fidelity in your view? Or, conversely, whether the highest possible fidelity would necessitate its removal?

That is a good question.
By using headphones one eliminates that problem (interaural crosstalk and also acoustical interference) and also remove the room from the equation, and thus get higher fidelity. I wonder if the feeling of a soundstage inside your head is related to not listening with both ears to a single channel, I've never actively looked for or read any research about this.

 

[thewas]

I wonder if the feeling of a soundstage inside your head is related to not listening with both ears to a single channel

Unfortunately not only, as there easy implemented crosstalk circuits or apps which mix some variable percentage of the right channel to the left one and vise versa, which often widens a bit the perceived soundstage (also a reason why open headphones sound more spacious than closed ones, acoustic crosstalk) but the dissapointing in head localisation remains.
If you want to have a more realistic imaging in front of you you need to manipulate the head related transfer functions like with https://smyth-research.com/ or

 

[fredoamigo]

That is a good question.
By using headphones one eliminates that problem (interaural crosstalk and also acoustical interference) and also remove the room from the equation, and thus get higher fidelity. I wonder if the feeling of a soundstage inside your head is related to not listening with both ears to a single channel, I've never actively looked for or read any research about this.

hello tuga


here it is for me the most complete study of the perception of the directivity/reflection that I have in my possession ... I come back from time to time because very long to read .
http://epubs.surrey.ac.uk/855204/1/27558504.pdf

 

[Hipper]

The question I was really interested in your opinion on was whether the direct sound from the R speaker arriving at the L ear (quite apart from any reflections) is necessary for the highest fidelity in your view? Or, conversely, whether the highest possible fidelity would necessitate its removal?

Toole writes about this (Third Edition page 159 onwards). Side wall reflections can partially fill in the possible 6dB hole around 2kHz caused by this inter aural crosstalk comb filtering.

My limited experiments confirm the effects he describes using pink noise but I still find that my stereo phantom image seems stronger without side wall reflections.

 

[Hipper]

I'd still like further investigation about where room treatment is applied. Front, ceiling, back reflections more obviously and uncontroversially lead to coloration, but I'm not sure I've ever seen a room that is treated only in these regions while leaving the sidewalls bare. I suspect this would be a good compromise.

Ceiling reflections are dependent on your speaker drivers dispersion patterns (as well of course as your ceiling shape and height). Mine being ribbons have a narrow vertical dispersion pattern and indeed, when I've experimented with absorbers on the ceiling I can't hear any difference, so I don't use them.

Floor reflections are debatable. I have thick carpet and underlay so haven't noticed any problem. It's often said that we are used to floor reflections and so ignore them. I suppose one could experiment by placing glass or mirrors on the floor. It's often recommended not to have a coffee table between speakers and listener - or a desk for that matter - but I've not experimented and I've seen no measurements from anyone. Logic says it would affect things but psychoacoustics doesn't seem logical.

With my rounded baffle speakers I noticed that I got reflections off my equipment when it was on the front wall between the speakers. Placing an absorbent panel in front of the gear made that evident. Moving the equipment to the side wall also lead to reflections in my room. I now have panels on the front and rear walls.

I do prevent side wall reflections using panels next to the speakers as opposed to on the wall.

In addition I have twenty one bass traps in the four wall - wall corners and four ceiling - wall corners.

The result is probably over the top for most people but it works for me. Not only was the frequency response smoothed (although it still needed further work from EQ) but perhaps more importantly decay times were reduced and evened out across the spectrum, along with smoother phase measurements. The room is 14' x 13', so not so big and almost square which may lead to more bass issues then perhaps a more ideal longer/narrow shaped room.

In the end, there are lots of theories and some science, a lot of which seems to depend on preferences. As we are individuals there are individual preferences. My conclusion therefore is that we should study the science (for me Toole particularly but Winer too) and be guided by them, but ultimately we should use that guidance not as dogma but to experiment for ourselves using our own listening with our own equipment in our own room to find what is best for us.

 

[andreasmaaan]

I'd still like further investigation about where room treatment is applied. Front, ceiling, back reflections more obviously and uncontroversially lead to coloration, but I'm not sure I've ever seen a room that is treated only in these regions while leaving the sidewalls bare. I suspect this would be a good compromise.

In my last studio, I had the front and back walls heavily treated (4" thick 6' x 10' rockwool absorbers spaced 8" from the walls), and had additional absorbers built into the desk to partly catch the floor/desk bounce. The ceiling was around 18', so although I wasn't treating that reflection, it was delayed around 18-19 ms and low in level relative to the reflections from the sidewalls, which were bare concrete. Speakers were CD (90° horizontal) from 800Hz-16,000Hz and toed-in to a point in front of the LP. This was a very good listening room IMO - definitely the best one I've had.

Side wall reflections can partially fill in the possible 6dB hole around 2kHz caused by this inter aural crosstalk comb filtering.

Yep, it was this phenomenon that partially informed my question.

 

[tuga]

Of course not. But reflections can subjectively mitigate some of the unpleasantness resulting from stereo recording/reproduction, IMO/IME (obvs this is not everybody's experience).

I agree. It's also perhaps musical genre or recording-type dependent.

 

[tuga]

Nothing really new but Joseph D’Appolito the explains the theory in

Testing Loudspeakers: Which Measurements Matter, Part 2

Directional queues come from the first arrival response.
We judge arrival direction in well under a millisecond.
However, judging what we are hearing takes longer.

To determine spectral balance, the ear/brain combination analyzes the incoming sound typically over a 5 to 30ms interval.
This interval is called the Haas fusion zone. Within this interval we are not aware of reflected sounds as separate spatial events.
All of the sound appears to come from the direction of the first arrival.
Lateral reflections from adjacent walls help extend the soundstage beyond the physical span of the loudspeakers.
The comb filtering action of the many early reflections arriving at the listening position with varying phases adds a sense of spaciousness to the sound. (It also argues against the need for phase accuracy in loudspeakers.)

You can see that the perceived timbres of sounds in rooms are the result of temporal processing and spatial averaging of reflected sounds arriving at our ears from many angles.
In typical home listening rooms, direct sound and early reflected sounds dominate.
Late reflections are greatly attenuated.
This is clear from the measurement of RT60s in the range of 0.2 to 0.4 seconds. (Compare this to concert halls where RT60s of 3 to 4 seconds are common.)
What we hear is a function of the directional characteristics of the loudspeakers and strong early reflections from the room boundaries.

You can think of the early reflection response as a loudspeaker’s in-room response averaged over a period extending out to 30ms after first arrival. But now we have a problem.
The early reflection response is room dependent.
A designer cannot predict how the early reflection response will look in any particular room.
This will depend on the room size and shape and speaker location.
It will also be affected by the room furnishings, any acoustic treatment, and the number of people in the room.
However, we can examine a loudspeaker’s directional characteristics anechoically.
To guarantee that sound arriving at the listening position is affected only by the arriving reflections and not by any off-axis anomalies in the speaker’s response, the off-axis response curves should be smooth replicas of the on-axis response with the possible exception of some roll-off at higher frequencies and larger off-axis angles.

 

[andreasmaaan]

@tuga that's a great summary (as I'd expect from D'Appolito!).

The only thing I'd note is with regard to this:

The comb filtering action of the many early reflections arriving at the listening position with varying phases adds a sense of spaciousness to the sound.

To be a bit picky, it's not comb filtering that does this. Rather, comb filtering is a byproduct of the phenomena that correlate with perceived spaciousness, which are primarily relative delay, level, and degree of interaural cross correlation (IACC) between the direct sound and reflections, which in turn is largely a function of the angle from which the reflection arrives.

Or, to put it another way, there is a strong (negative) correlation between perceived spaciousness and IACC, and a strong (positive) correlation between perceived spaciousness and relative level/delay of the reflection, but little (or no) correlation between perceived spaciousness and degree of comb filtering per se.

 

[hardisj]

I'd still like further investigation about where room treatment is applied. Front, ceiling, back reflections more obviously and uncontroversially lead to coloration, but I'm not sure I've ever seen a room that is treated only in these regions while leaving the sidewalls bare. I suspect this would be a good compromise.

Kind of OT but when I tested the Revels recently in my HT, I was surprised to find that I much preferred them in my living room which has no treatments at all. Just the standard couch/chair/blinds. It made me think that I needed to remove the sidewall panels in my HT and listen again... I just forgot to do that. Maybe if I have time this weekend I will, before I have to send them back.

 

[tuga]

Kind of OT but when I tested the Revels recently in my HT, I was surprised to find that I much preferred them in my living room which has no treatments at all. Just the standard couch/chair/blinds. It made me think that I needed to remove the sidewall panels in my HT and listen again... I just forgot to do that. Maybe if I have time this weekend I will, before I have to send them back.

The response in your sitting room looks better than in your cinema room below 1kHz and a lot better below the transition region.
One would expect it to be the other way around...