Research on reflections

[aac]

speakers contribute the other half

At least it solves "wide or narrow" question almost completely.

 

[Chromatischism]

Thanks for having me,

I'm quite confident, Toole was right in his conjecture that stereo as a concept needs some additional thought. At least as an aftermath.

Stereo is flawed--think of just turning the head in front of the system. All rules broken. Stress rises because the (in)famous "image" isn't plausible to the untrained ear anymore. The hearing can be trained, though, as kind of a sport. There already were stereo training recordings out in the 7ties, not kidding.

My conclusion is, Toole more or less suggested to soften the rigid stereo regime to something bearable in daily life. Of course common people agree, hence prefer the less demanding speakers. More envelopment, less conflict with the idea of "some music in some room".

At least there are two individual functions of a stereo speaker, as to say.

- direct sound for directional clues
- reveberation by some mayor extent determines the timbre

I've got, for long now, a quite elaborated, narrow-ish constant directivity set-up, with at least 20dB top-quality headroom. Harman tilt, check, smoothness, check, 30Hz full level, check. I never listen "critically", let alone focusing on image (pun intended). So as to evaluate a wider directivity I just added one xtra speaker for the sum of right/left, placed of course in the middle of the two stereo lighthouses. A DML actually, with extreme wide, de-correlated dispersion. It is x'ed-over where the main speakers get succedingly narrower, namely at about 1kHz.

Yes, it eased the stereo thing quite a bit.

So, as not to bother You with ready-made-up conclusions, what do You think: seperate the two functions mentioned above, directional clues from narrow left/right and adding appropriate corrections for the desired smooth--adjustable that was, reveberant, means grand total sound field?

Objection: to pass on expensive solutions that resolve these somehow contratdictionary issues in one piece, partly.

You lost me in the last few paragraphs, but yes, Stereo was a compromise from the beginning with early 3-speaker setups. They had the same problem then as we do now: where to put the middle speaker!

But here is my take on issues that a center speaker creates that are not talked about, from my experience testing an identical, vertical speaker in the center.

My Left and Right speakers are angled to cross behind the listener. Listener can see the inside face of each speaker. This gives the most pleasing sound in terms of frequency response, imaging, and soundstage, and is the case for many speakers on the market. There is about 1.5' to the side walls. The Center speaker is pointed straight ahead, as expected.

• The center speaker will never have the same toe angle as the left and right speakers, thus giving the listener the 0° sound profile vs the 10-30° listening window for Left and Right
• The Center speaker will never have the same side wall reflections as Left and Right

The result is that the Center speaker sounds tonally different due to the different toe angles. And, its apparent image size is smaller and confined to the speaker itself, again due to a lack of reflections. It is very localizable, whereas the Stereo effect is more spacious and sounds like it is coming from the screen (if you have one). So what we thought we would get with 3 identical speakers, is not what we are getting in practice. Of course this situation can be improved with a projection screen and an elevated center speaker firing through it rather than settling for a speaker below a flat panel. But that's not most homes.

 

[Tangband]

Hello,

interesting discussion. There are many aspects to consider.

One important aspect is the listening distance, position and the room. By adding the directivity of the speaker to the equation you get the important ratio of direct sound to later reflections at the listening position.
You have envelopment and exactness, which are difficult to achieve at the same time, since you need more direct sound for exactness and strong diffuse later reflections for envelopment.

So the perfect speaker is room dependent.
Here are some thoughts about it.

Best
Thomas

Doing semiproffesional recordings of acoustical instruments, all I can say is that every 2 channel recording is only an illusion of the real event. Multimic recordings even more so.
And youre absolutely right about the perfect loudspeaker is room dependent. It depends very much on the listening room, what kind of loudspeakers functioning the best. Sometimes a narrow dispersion is prefered sounding the best, sometimes not.

 

[fineMen]

You lost me in the last few paragraphs, but yes, Stereo was a compromise from the beginning with early 3-speaker setups. They had the same problem then as we do now: where to put the middle speaker!

But here is my take on issues that a center speaker creates that are not talked about, from my experience testing an identical, vertical speaker in the center.

My Left and Right speakers are angled to cross behind the listener. Listener can see the inside face of each speaker. ...

Apolgies, language barrier. My question is quite simple. It is not about a so called 'center speaker', as that would point to a technical realization, before the original problem was clearly stated.

First, a room solely dedicated to stereo would contribute to the total cost of ownership a lot. That's not feasable, right? Me thinks, Toole took that as a starting point. One might argue, that even with a 'perfect' room stereo doesn't actually work. Move the head: both, time related and spectrum related directional clues would be affected by just a little turn. The change would naturally appear as implausible to the hearing, and the illusion is for sure broken. To ignore these flaws can be trained, though. As always, some are more successful with this than others.

At least for recreational purposes, Toole identified two distinct tasks for the loudspeaker(s). To first generate a well balanced sound field, reverberation included. Second to provide equally balanced directional clues from at least two separated speakers. These tasks are interconnected by a stated demand for quite early reflections as opposed to the summed up long-term reverberation.

There is only one comment from Toole addressing this. Q: Do early reflections blurr the so called phantom image? A: No, if so only mildly.

I would like to strech this further. The early reflections don't need to originate from those two speakers, that provide the directional clues for the stereo imaging. There is no relevant connection between early reflections and the phantom image as such. Only that early reflections provide clues regarding the actual listening room, which facilitate--magically, to overcome flaws that are inherent to stereo as a concept.

My personal conclusion is, that the directivity index of the common two stereo speakers isn't crucial. As long as, by whatever means, some early reflections of whatever origin are provided. While, of course keeping the overall spectral balance of the reveberation field intact. Those early reflections don't need to be phase-correlated to the directional clues.

What hinders us to accept the most simple solution to the directivity issue? An additional, monophonic speaker that adds early reflections, and by that reveberation in a frequency range, where the main speakers become more directional? The more narrow the dispersion of the mains, the wider opportunity to dial-in the actual proportions of early / late reflections and directional clues left/right? Reiterated, technically this should be, once set, equivalent to a two speaker set-up with predefined properties.

 

[fineMen]

Reiterated, technically this should be, once set, equivalent to a two speaker set-up with predefined properties.

Practically I've got speakers with a directivity index of about 6(dB) @ 300Hz gradually narrowing down to about 9(dB) @ 10khz. These are toed-in by 45°. Due to the actual radiation pattern this alters the effective directivity index to, say 5 and 7 respectively. A DML with 'chaotic' phase rendition but very wide dispersion is added behind the baseline of the former two. Since the DML is a dipole, it was possible to minimize the output towards the listener. It was integrated by feeding it with a summed left+right signal, crossed over @ 1kHz with a quite shallow 6dB/oct. The level (measured in the reverberation filed), referred to the mains (both) was not critical at all. The difference between -10dB up to 0dB was gradual, not game changing. Don't say this is the holy grail, but it catches my interest still. Visitors like it, and my cat too.

 

[Kvalsvoll]

Much progress has been done in later years, what this is about is the acoustic properties of the room and the speaker radiation pattern.

Some interesting points:
- We are doing sound reproduction, not sound production.
- No, early reflections - especially discrete - are not beneficial.
- Effects of vertical and horizontal radiation pattern are different - as some have also described in this thread.
- Increasing indications that one specific radiation pattern is the best - and generally preferred.
- We now know how to fix the room.

 

[fineMen]

Much progress has been done in later years, what this is about is the acoustic properties of the room and the speaker radiation pattern.

Some interesting points:
- We are doing sound reproduction, not sound production.
- No, early reflections - especially discrete - are not beneficial.
- Effects of vertical and horizontal radiation pattern are different - as some have also described in this thread.
- Increasing indications that one specific radiation pattern is the best - and generally preferred.
- We now know how to fix the room.

May I cite Your website:

"The F205 will not present a diffuse sound filling the room with lots of reflections. If that is what you desire, there may be other speakers that are more suitable."
and
"This minimizes reflections from surfaces close to the speaker, making the speaker less affected by placement and improves clarity and spatial location of instruments in the soundstage."

You first give the medicine, and only after investigate the desease. The success justifies Your efforts taken for a special treatment. A schema often seen in audio. Claim some technical means and connect them to some perceivable outcome, described in steampunk poetry. "The 3 by 5 entangled bamboo root fibres in the cone give a detailed depth to the soundstage; You will be frightened to fall into it!"

You advertise a mechanical cardioid combined with horn tweeters, right? Somebody else will sell the apropriate room / treatment. But who would take care of a customer, who, at some day would stop to listen critically? But rather than to check the justification of the purchase over again, she would start to actually use it for sheer enjoyment? Stop listening to salesman and speakers, but what actually did the sound engineer negotiate with the musicians? Where's the trick that makes me like the sound of music?

Last point addressing the first point You made: just by technical means alone, sound can never be "reproduced". It is not possible. Physics denys it. Many people take the term of "sound reproduction" to literally. A less specific mission shall be accomplished. To convey an acoustical sensation to a human. Again, the specifics are left open, by the infamous 'circle of confusion'. The how to. In this regard Your proposed solution isn't the only one. It doesn't fit the newly set quasi-standard materialized by e/g Genelec.

 

[Kvalsvoll]

May I cite Your website:

"The F205 will not present a diffuse sound filling the room with lots of reflections. If that is what you desire, there may be other speakers that are more suitable."
and
"This minimizes reflections from surfaces close to the speaker, making the speaker less affected by placement and improves clarity and spatial location of instruments in the soundstage."

You first give the medicine, and only after investigate the desease. The success justifies Your efforts taken for a special treatment. A schema often seen in audio. Claim some technical means and connect them to some perceivable outcome, described in steampunk poetry. "The 3 by 5 entangled bamboo root fibres in the cone give a detailed depth to the soundstage; You will be frightened to fall into it!"

You advertise a mechanical cardioid combined with horn tweeters, right? Somebody else will sell the apropriate room / treatment. But who would take care of a customer, who, at some day would stop to listen critically? But rather than to check the justification of the purchase over again, she would start to actually use it for sheer enjoyment? Stop listening to salesman and speakers, but what actually did the sound engineer negotiate with the musicians? Where's the trick that makes me like the sound of music?

Last point addressing the first point You made: just by technical means alone, sound can never be "reproduced". It is not possible. Physics denys it. Many people take the term of "sound reproduction" to literally. A less specific mission shall be accomplished. To convey an acoustical sensation to a human. Again, the specifics are left open, by the infamous 'circle of confusion'. The how to. In this regard Your proposed solution isn't the only one. It doesn't fit the newly set quasi-standard materialized by e/g Genelec.

I choose not to comment on this. I appreciate that you respect that.

 

[fineMen]

I choose not to comment on this. I appreciate that you respect that.

Sure; it's only so, that on Your website You state to follow a scientific path. In my understanding that opens a door for unbiased discussion.

You directly contardict the findings of Toole, namely the preference for a vivid soundfield. Sources for such conclusions would be intersting indeed. As far as I get it, You rely on a proverbial self-evidence of "less reflections => less distortion => more real => more quality => more enjoyment". (distortion meant as deviation from what is considered trustworthy "reproduction")

Anyway, I accept Your choice. No bad feelings.

 

[fluid]

You directly contardict the findings of Toole, namely the preference for a vivid soundfield. Sources for such conclusions would be intersting indeed.

I don't think that it does

 

[fineMen]

I don't know think that it does

View attachment 162743

Thanks!

Conclusions are derivative, though. I cannot quote it directly, so I rephrase: science revealed that musicians hear different, period. All the conclusions start there, but are no longer supported by the original investigation. Words like "satisfying amount of" very much read like an afterthought of the commenter. The Memo, second, only states as a may-be, that pros have individual preferences.

Both implicitely and explicitely don't address the common wo/man. To the contrary one might conclude, that a specific group of listeners is an exception to the otherwise confirmed rule.

 

[fluid]

Both implicitely and explicitely don't address the common wo/man. To the contrary one might conclude, that a specific group of listeners is an exception to the otherwise confirmed rule.

In terms of preferences for directivity and reflection patterns there is no definitive rule. The percentages of people that prefer speakers with wider directivity and more early reflections in blind testing may be higher than those that prefer less, but the exact reasons for that are also somewhat derivative. Choice of program, mono or stereo, multichannel and the persons audio history certainly play a part. There is no point in assuming that you or any other person fall into one group or another or even that your preference will be consistent. There is Japanese research in Toole's book that found that some engineers preferred less early reflections when working but more when recreationally listening at home.

When I see people form such strong opinions about it one way or the other and quote Toole as the source I have to wonder if they have actually read the whole book as this is not the only time that doubt is cast on the idea that everyone likes more reflections. I have tested it for myself and I certainly do not although sometimes on some program material I can see the attraction.

 

[fineMen]

In terms of preferences for directivity and reflection patterns there is no definitive rule. The percentages of people that prefer speakers with wider directivity and more early reflections in blind testing may be higher than those that prefer less, but the exact reasons for that are also somewhat derivative. Choice of program, mono or stereo, multichannel and the persons audio history certainly play a part. There is no point in assuming that you or any other person fall into one group or another or even that your preference will be consistent. There is Japanese research in Toole's book that found that some engineers preferred less early reflections when working but more when recreationally listening at home.

When I see people form such strong opinions about it one way or the other and quote Toole as the source I have to wonder if they have actually read the whole book as this is not the only time that doubt is cast on the idea that everyone likes more reflections. I have tested it for myself and I certainly do not although sometimes on some program material I can see the attraction.

Thanks again, and I utterly agree this time. Statistics doesn't tell what an individual outcome shall be. Regarding Toole, my objective is not to normalize people.

As to regain focus let me explain where I come from. I cannot stand stereo for longer than a few seconds. The artifical soundstage is to laboursome for me. It flipps away for the better, for the worse it challenges my senses with contraditory sensations. I cannot keep my head straight. I have to curiously move it to investigate the sound source deeper--and then it all collapses. It comes as natural to me as breathing, impossible to suppress.

You may have trained it. A pro for sure has trained it to actually focus on his work. He steers the process of generating the infamous soundstage.

Metaphor: when driving a yacht myself I easily stand the boat's movements. As soon as somebody else takes over, I die from sea-sickness.

I actually felt enlighted, when I encontered Toole's book. It describes my problems to the point. Reflections help--times a lot, to cope with the inherent flaws of stereo.

 

[Chromatischism]

You directly contardict the findings of Toole, namely the preference for a vivid soundfield

Where does Dr Toole state this?

As far as I know it all goes back to the speaker shuffler experiment, which was not definitive on that aspect.

 

[fineMen]

There's a chapter in his book that suggests to go for at least lateral reflections, as long as they aren't exagerated. Whatever that means, again, read chapter "7".
I personally think that the urban legend of reflection=>bad is debunked.

Another aspect, more on topic, still bothers me. How to evaluate the expected soundfield of an in-wall speaker. Or more common a bookshelf speaker in a bookshelf. It for one would be devoid of the most colored portion of the reflections. Second it wouldn't show the desired tilt exactly due to that. To lift the lower part as to meet the Harman target would also affect the direct sound.

With my proposal of an additional mono speaker, carrying left+right summed up behind the main speakers, the total sound field would become more bright than without. But it would add to the amount of reflections.

 

[Chromatischism]

You'd lose imaging though. Much stereo music and immersive audio would suffer.

 

[fluid]

As to regain focus let me explain where I come from. I cannot stand stereo for longer than a few seconds. The artifical soundstage is to laboursome for me. It flipps away for the better, for the worse it challenges my senses with contraditory sensations. I cannot keep my head straight. I have to curiously move it to investigate the sound source deeper--and then it all collapses. It comes as natural to me as breathing, impossible to suppress.

I actually felt enlighted, when I encontered Toole's book. It describes my problems to the point. Reflections help--times a lot, to cope with the inherent flaws of stereo.

I think I can understand where you are coming from. For a long time at home I couldn't care less about imaging and preferred it to be blurry as it otherwise annoyed me.

A few personal experiments have really changed my mind and preference. I have reduced a lot of early reflections including the lateral ones. Early is such a relative term. My initial aim was to try and keep the first 15ms below 20dB if possible. I pointed the speakers directly at the listening position and re equalized for that change. I also use other filtering to make the two channels more equal in level and frequency response. Overall I am now quite amazed at the imaging specificity and actually seek out or include music that I might not otherwise enjoy for the imaging it presents. Removing those reflections does expose the change in timbre from phantom centre images but there is equalization that can be used to reduce that, which then introduces it's own issues so I can see why some people are such verdant supporters of multichannel. Reflections both help and hinder they are neither all good or all bad.

Someone else might try that and say that they hate it.

I personally think that the urban legend of reflection=>bad is debunked.

Another aspect, more on topic, still bothers me. How to evaluate the expected soundfield of an in-wall speaker. Or more common a bookshelf speaker in a bookshelf. It for one would be devoid of the most colored portion of the reflections. Second it wouldn't show the desired tilt exactly due to that. To lift the lower part as to meet the Harman target would also affect the direct sound.

With my proposal of an additional mono speaker, carrying left+right summed up behind the main speakers, the total sound field would become more bright than without. But it would add to the amount of reflections.

I think there are two urban legends, one that reflections are great and one that they are bad, as I said above I don't think either is true. In Floyd Toole's Third edition book he devotes a few paragraphs in trying to explain the above.

An in wall speaker can behave as if it is in an infinite baffle which changes the baffle step and diffraction signature and also the distance and angle to where the reflections will come from. Without some thought this is not a simple exercise to get right.

A speaker in a bookshelf could potentially be the same or it could be a speaker placed inside a cavity and will sound like it with resonances, very early reflections and diffraction that will colour the timbre.

There are a number of methods for increasing reflections at times when they are helpful but as said above preference plays a large part in which option someone may like more.

Earl Geddes approach is to eliminate reflections within 10ms but then to increase latter reflections through reflective room surfaces to end up with an overall quite reverberant room. To angle the speakers to make use of time intensity trading and the directivity of his design to minimize under 10ms and maximize later reflections from the opposite walls and other surfaces.

Some recording studios and listening rooms used a Haas kicker to put a bump in the reflected energy around 25 to 30ms and increase the later arriving sound making the room sound bigger and more pleasant.

The same thing can do done with speakers placed behind the listener and various decorrelation methods, filtering and aiming of the speakers can be used to fake it.

Some add rear facing ambience tweeters and the list could go on.

 

[Kvalsvoll]

Sure; it's only so, that on Your website You state to follow a scientific path. In my understanding that opens a door for unbiased discussion.

You directly contardict the findings of Toole, namely the preference for a vivid soundfield. Sources for such conclusions would be intersting indeed. As far as I get it, You rely on a proverbial self-evidence of "less reflections => less distortion => more real => more quality => more enjoyment". (distortion meant as deviation from what is considered trustworthy "reproduction")

Anyway, I accept Your choice. No bad feelings.

This I can try to answer.

First, a company website presenting products is not a scientific publication. It is not meant to be so. What you see here, will merely be the results from research - or "research" - going on behind the scenes. This goes for my website, and for most others as well - they show products, and do not go into depth with technical analysis and discourse that most customers never find interesting nor relevant for them.

Even my technical articles I have tried to write so that many can read them, and get through with my message. And let people experience and try out things for themselves - like the amplifier test, instead of listing results from scientifically viable controlled tests, I encourage people to listen to the original sound sample and compare it to the one recorded from the output of an amplifier - how big is the difference. If you have only read the typical reviews, this would be quite surprising, and much more convincing.

I do not see much conflict between my conclusions so far, compared to Toole. It is more of a different approach, that builds on results both from others and my own earlier experiences and findings.

 

[fineMen]

Some add rear facing ambience tweeters and the list could go on.

Thanks a lot for the valuable input. I'm going to digest it. And of course I'll try out what You suggest with my digitally controlled modular system. Happy me to have the opportunity.

 

[fineMen]

This I can try to answer.

... It is more of a different approach, that builds on results both from others and my own earlier experiences and findings.

Yep, as You said. Advertising--at You website, isn't meant to be a scientific publication. I still think this state of affairs is a pitty. Fair enough, though. Amps, I always thought this case is settled. I personally cannot imagine to question my humble AVR, if my goal is to just fancy me with interesting musical presentations. It does the trick, so what else to expect? More detail? Not really ... More volume would entertain my neighbours next street ...

I think our little conflict is settled. As I, for the time being are not that much into "imaging" with stereo, I can't appreciate audio systems that claim for themselves to be superior in this respect. For whatever reason. Would You mind to link some scientific sources here anyway?

Last, to tell that Your subwoofers are technically "horns" is a bit of a hyperbole. Just my impression. Performance still is pretty much o/k for the formfactor.