The Objectives of a Loudspeaker in a Small Room

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
More thoughts

My ongoing disagreement about sound reproduction with Lynn Olsen gets me thinking. How can two reasonable and knowledgeable people have such different opinions on a common subject within their expertise? I then noted a comment he made which made me think.

Lynn, and many, view the sound reproduction of an orchestra as the end goal - they like classical/orchestral music. This is not my preference and I see things differently. Consider this:

If we have a playback room, as we all do, and that room is pretty small, as virtually all of us have, is it reasonable to think that the sound of a huge ensemble of instruments playing in a very large space could ever be convincing? You see, I don't see this as feasible at all. Any acoustic remnants of the small room will totally confound the perception of the acoustic of the larger one - this is unavoidable, the ear cannot be so tricked. Thus as a first requirement for this "symphonic" playback, we need a virtual anechoic chamber - no local acoustic. But the large space acoustic is multi-path with sound arrivals from all directions, so two channels is never going to do the job. It just does not seem feasible to me to ever get a really convincing playback of a symphony over two speakers in a small room, and I have yet to hear a playback that even comes close. Lynn and I seem to agree on this point.

But you see, I prefer to listen to small ensembles with a few instruments. These performances don't have much acoustic in them (and when they do it tends to be of a smaller space) and my playback room adds or emphasizes the acoustic in a very real manner. I have done personal recordings of a friend who is a concert pianist (Dickran Atamian) and played them back for him and myself. The imaging and realism was astounding. It was just as if the piano was in the room with us. So in this case the ideal of an acoustic playback that mimics the original is indeed very possible, achievable and even achieved.

We must be very careful in these kinds of discussions to outline what our expectations are. What are we trying to achieve? The ideal of recreating the analog of one being in a large hall listening to 100+ instruments over two speakers is probably not achievable (this seems best done with headphones where I have had this experience). But the ideal of creating the impression of a solo vocalist or small ensemble in ones own listening room is not only achievable, but I believe it is here today I have found in my own listening that the more instruments that there are in the recording, the less impressive I find the playback to be - even with good recordings. This is a generalization of course, and much depends on the recording, playback, an infinite number of variables, but as a general rule I find that small ensemble recordings sound more realistic when played back in small rooms. If all you want to listen to is a large orchestra in a large auditorium, then I fear that you will forever be disappointed in a small listening. I go out for this kind of performance.
 
gedlee said:


Thats the one. Thanks so much for posting this.
....


You're Welcome!
Thank you for making all these documents available...



I'm far away from being as experienced and educated in audio stuff as you or Lynn, or many others in this forum, but I think your last post put's it very nicely - Stereo reproductions has got it's limits, and if you set your objectives too high, you will hunt for that holy grail for the rest of your life, never reaching a point of satisfaction - or end up in the traps layed out by the high end industry...

Maybe it's just a matter of time until the reproduction of the Berliner Philharmoniker will be possible in my 30 sqm livingroom, but I'm afraid that will take much more than a CD-Player, Stereo Amp and two speakers.... Research in being done and solutions will be found, but most probably way beyond the questions of dipole, monopole, point source or multiway etc.

cheers. LC
 
Re: Gedlee

You write:
If we have a playback room, as we all do, and that room is pretty small, as virtually all of us have, is it reasonable to think that the sound of a huge ensemble of instruments playing in a very large space could ever be convincing? You see, I don't see this as feasible at all

but isn't something different but equally realistic and attractive (and equally not easy to achieve) feasible?

don't You think that the problem of realistic scale is really a problem of realistic of perspective?

what You cannot certainly achieve in a listening room is a front row perspective of a large auditorium, that is beyond question
indeed an anechoic chamber would be neccesary

but imagine Your listening room as a kind of a box in an opera house
in other words - imagine Your listening room transported to the opera house - the front wall of Your listening room is taken out and You have a kind of realistic perspective
perspective from the plane where the stereo microphone (or a main microphone) is placed
isn't it achievable?
 
The issue again comes back to objectives. My objective is "reproduction of the original". Sounding "good" or "bad" is irrelevant, as is personal taste. What the producer intended is what I want to hear - maybe HIS taste is bad, or maybe the recording is pathetic, if so then that's what I want to hear.

My only point in the above was to question what level of accurate reproduction can we expect from two channel stereo? I think that is a relevant question - one that needs to be discussed. But preference is another issue altogether and I don't want to go there.
 
Thanks Lovechild, that's it.

gedlee said:
What the producer intended is what I want to hear

That might be difficult to ascertain without being present at the final mix and possessing near perfect audio memory. But I understand what you are saying. The difficulty is implementing that with 2ch in a acoustically small room ;).

What I found fascinating about the study is not so much the modal region performance of the different formats, since that is very room dependent, nor the fact that the ported JBL scored the highest in the bass region - I'd wager that 90% or more of the testers owned ported box speakers themselves. It's the fact that whereas the Summa (as expected) outperformed the JBL (most likely) due to it's highly optimized horn and resulting mid-upper frequency performance, it was at most the equal of the Gradient.
The Gradient using what has to be considered an ever more compromised horn than the JBL - an active speaker cone with a rolled surround termination, poor throat with VC attachment, etc, etc.
Why was the treble and mid performance of such a non-optimal horn, made up of good, but far from stellar drive units, at least the equal of the SOTA driver/OS waveguided Summa?
Was the Gradient extensively measured? Was the controlled directivity the dominant factor for the Gradient even when the JBL fared so poorly?

cheers,

AJ
 
Re: Gedlee

I would like to clear up my position -
"reproduction of the original" is also my objective

what I intend to write above had nothing to do with personal preferences, with >>sounding "good" or "bad"<<
Perhaps I shouldn't use the word "attractive" as it appeared to be misleading, all I wanted to say was that it could be "an ideal equally worth pursuing"

sorry if my english is clumsy
my command of Your language is weak and rather passive

You write:
>>What the producer intended is what I want to hear<<

of course the problem is that "the original" can be defined in two ways i.e. as "the orginal musical event" or as "what the producer intended"

Isn't it that if You say that "the original" is "what the producer intended" than Your REAL"original" is what could be heard from the producer's studio monitors in his control room and nothing other?
If so than don't You think that domestic loudspeakers should mimic as closely as possible the sound of studio monitors and our listening rooms should mimic as closely as possible the acoustics of control rooms?

Don't You think that this is the only way to hear "what the producer REALLY intended"?

BUT - are studio monitors and control rooms sufficiently standarized to serve as a model?

As to the question of limitations of accurate reproduction from two channel stereo may I ask You to answer to my question already stated above:
don't You think that the problem of realistic scale is in fact a problem of realistic of perspective?
Do You think that such realistic perspective is unachievable in domestic environments?
If You think so - why?
What is wrong with the metaphor, that I proposed above, of listening room as a box connected through the "lifted" front wall to the recording spaces?

Moulton suggests that if we supress the reverb of the listening room after 50 ms the listening room becomes just a carrier of information about acoustics of the recording space:
He says:
>>But if you take a look at what's really going on in recordings, playback rooms are generally small and the early reflections happen very quickly-whereas in a recording space (or simulation of a recording space that we do with artificial reverb), those reflections are much, much later in time.
What happens is that the early reflections of the playback room carry information about the recording room quite well<<

Is anything wrong with what he states?
 
Re: AJinFLA

You wrote: >>The difficulty is implementing that with 2ch in a acoustically small room<<


the real problem lays elsewhere IMHO

after all the producer is also producing with only two channels in an acoustically small control room :)

so two channel and an acoustically small room is not the real problem, e can easilu replicate that conditions

the real problem is with lack of studio quality of the electronics and the loudspeakers (in the first place) and dissimilarity of acoustics of our living rooms with that of control rooms

the real problem is also how to reproduce "the original" in the other meanig i.e. "the real musical event in real (or somehow artificially created) space" in an acoustically small room and using only two channels
 
graaf said:
Re: Gedlee

You write:
>>What the producer intended is what I want to hear<<

of course the problem is that "the original" can be defined in two ways i.e. as "the orginal musical event" or as "what the producer intended"

Isn't it that if You say that "the original" is "what the producer intended" than Your REAL"original" is what could be heard from the producer's studio monitors in his control room and nothing other?
If so than don't You think that domestic loudspeakers should mimic as closely as possible the sound of studio monitors and our listening rooms should mimic as closely as possible the acoustics of control rooms?

Don't You think that this is the only way to hear "what the producer REALLY intended"?

BUT - are studio monitors and control rooms sufficiently standarized to serve as a model?

As to the question of limitations of accurate reproduction from two channel stereo may I ask You to answer to my question already stated above:
don't You think that the problem of realistic scale is in fact a problem of realistic of perspective?
Do You think that such realistic perspective is unachievable in domestic environments?
If You think so - why?
What is wrong with the metaphor, that I proposed above, of listening room as a box connected through the "lifted" front wall to the recording spaces?

Moulton suggests that if we supress the reverb of the listening room after 50 ms the listening room becomes just a carrier of information about acoustics of the recording space:
He says:
>>But if you take a look at what's really going on in recordings, playback rooms are generally small and the early reflections happen very quickly-whereas in a recording space (or simulation of a recording space that we do with artificial reverb), those reflections are much, much later in time.
What happens is that the early reflections of the playback room carry information about the recording room quite well<<

Is anything wrong with what he states?


We can only judge "what the producer wanted us to hear".

The rest of your discussion is a very real question and not a new one (check Floyd Tooles discussion of this same issue). We are not going to resolve these problems here. But the bottom line to me is that we must be seeking to have loudspeakers with better and better "transparency" (trueness to the original signal) in both the mixing room and our own listening spaces. This is why standarizing on a particular speaker is a problem - it doesn't allow for improvement. Its a never ending circular issue - improve monitors > improve playback systems -> further improve monitor -> further improve playback systems.

What you quote Moulton as saying is quite incorrect. The rooms early reflections dominate the perception of "space" and image and nothing from the recording comes through in that period of time. After 50 ms the small room's acoustic is basically gone and the recodring acoustic will come through, but at that point our hearing system has already made its decision on the acoustic space - its a done deal - and the contradictory acoustics between the early and late periods only confusses it.
 
now I see
thank You so much for the answer :)

BTW it is really sad that someone like Moulton (who seems to be a serious engineer and researcher) is disseminating informations that are "quite incorrect" :(

anyway thank You very much for admiting that the question I put forth is real even though we are not going to resolve these problems here

I hope that someday we will :)

best regards,
graaf

ps.
only one more question
What's Your opinion on the Stereolith three-channel stereo?
It is so unusual but it appears to work as promised, though the way it recreates the soundspace somewhat differs from that of standard stereo set up: http://magazine-audio.com/essais-hi-fi/essais-enceintes-hi-fi/22
 
Administrator
Joined 2004
Paid Member
graaf said:
What's Your opinion on the Stereolith three-channel stereo?


"Bidule" indeed! Trucage and gadgets.

Actualy - it looks interesting. No details on the crossover or an phase tricks that might be going on are given in the article. Should be easy enough to try, tho.

As to the matter of large music in small rooms, yeah, it's hard to do. But I have heard it done, and done well. And I don't think there is anything wrong with doing "better" than the producer did. Most of them work in somewhat limited conditions. Having heard some recordings reproduced in a way that would blow your mind, I wish the engineers and producers could have been there. They would have been pleased and proud.

Some large systems in large rooms don't do symphnc music very well either. They can seem flat and without the liveliness of the recorded venue. Why exactly, I don't know.

I've been working so hard at getting my speakers to sound good on big orchestral music that they now do that better than the small stuff. How did that happen? :confused:
 
"bidule" or not ;) the conclusion of this review is very promising:

>>nous avons eu un plaisir infini à l’écoute de ce système, nous nous sommes affranchis de nos apprioris, de nos doutes, de nos certitudes, de nos références pour ne nous concentrer uniquement sur la musique, c’est à ce prix que la magie a opérée<<
 
I vision a diversified opinions on the subject with no conclusions. But it would be fun to see what all comes out of this. Might end up like those cable threads.:D

Since the room has no less influence on sound than probably design of the speakers themselves, perhaps a standard room characteristic be assumed to that the focus can be more on the Loudspeaker?:angel:
 
Re: soongsc

listening rooms seem to be acoustically not so diverse, they are quite standarized, that is - quite typical
after all they are in most cases our living rooms
"living room" has functions and architects are supposed to take those functions into account when they work
it has to be not to small and not to big and to have proportions most suitable for all of its functions to be best realized
typical sizes, typical proportions, and not very diverse acoustics because it is a function of furnishing and decor - typical furniture, carpets, curtains etc.
"living rooms" of typical size and with typical furnishing and decor (that is most of them) are very similar from the point of view of fundamental acoustical characteristics i. e. the RT60

if I'm wrong than correct me, comments welcome! :)
 
AJinFLA said:
Thanks Lovechild, that's it.

It's the fact that whereas the Summa (as expected) outperformed the JBL (most likely) due to it's highly optimized horn and resulting mid-upper frequency performance, it was at most the equal of the Gradient.
The Gradient using what has to be considered an ever more compromised horn than the JBL - an active speaker cone with a rolled surround termination, poor throat with VC attachment, etc, etc.
Why was the treble and mid performance of such a non-optimal horn, made up of good, but far from stellar drive units, at least the equal of the SOTA driver/OS waveguided Summa?
Was the Gradient extensively measured? Was the controlled directivity the dominant factor for the Gradient even when the JBL fared so poorly?


I think that we are talking about different speakers. The Gradients that we used did not have horns. They were a readily available "Hi-End" loudspeaker of some rupute, which made them a good basis for comparison. But they were entirely direct radiators.

Unfortunately I don't have any measurements from the Gradients.

The Gradients were very good sounding speakers, granted, but here is the thing. They were 12 dB LESS efficient than the Summas with a least 20 dB less Max SPL capability. Thus, while they were good sounding, they would never even come close to be acceptable at high volumes such as in a Home Theater playing a live performance of Cream (for example). Its a whole lot easier to get good sound with limited headroom than it is to get the same sound with 20 dB more output.
 
graaf said:
Re: soongsc

listening rooms seem to be acoustically not so diverse, they are quite standarized, that is - quite typical
after all they are in most cases our living rooms
"living room" has functions and architects are supposed to take those functions into account when they work
it has to be not to small and not to big and to have proportions most suitable for all of its functions to be best realized
typical sizes, typical proportions, and not very diverse acoustics because it is a function of furnishing and decor - typical furniture, carpets, curtains etc.
"living rooms" of typical size and with typical furnishing and decor (that is most of them) are very similar from the point of view of fundamental acoustical characteristics i. e. the RT60

if I'm wrong than correct me, comments welcome! :)


Well, I think that you are wrong. My listening room is anything but typical and its NOT a living room.

This is the whole point of this thread. Are we trying to squeeze some speakers into our wifes living rooms and call it a "sound system"? (I'm not.) Or are we trying to find the ideal solution for really exceptional sound reproduction (again as I am). Because if our "goals" are not the same then its highly unlikely that our approachs or our results will be comparable.

As I have said before, if you are limited in what you can do (because of space or whatever) then I sorry for you - I wish you the best. But this stuff is my life and I take it very seriously. When I discuss things here it is from this "extremist" point of view and you have to keep that in mind. I am seeking "the very best that can be achieved". If I have to knock out a few walls to do that then so be it - I've done that before. But basically its just easier to start from scratch with a new room and design it right from the ground up (I do that a lot.). Thats what I specialize in. (I don't install Bose cubes in living rooms.)
 
soongsc said:
I vision a diversified opinions on the subject with no conclusions. But it would be fun to see what all comes out of this. Might end up like those cable threads.:D

Since the room has no less influence on sound than probably design of the speakers themselves, perhaps a standard room characteristic be assumed to that the focus can be more on the Loudspeaker?:angel:


I am sorry, but I have drawn many conclusions in this regard thank you. To me its not "diversified opinions". You will find that most "experts" agree (Floyd Toole and I agree on virtually everything - some minor differences, but on the whole we see eye to eye.)

There is no way to standardize the rooms, but there is a way to make virtually any room acceptable to excellent. Its all a matter of design, but you have to know what to do. Much of the "diversified opinions" out there are simply incorrect. Granted, sorting out the "wheat from the chaff" can be a difficult thing to do , but read some of the classic texts, like Kutruff, or Floyd Tooles stuff and you will come to understand the realities.
 
Re: Gedlee

and what about that unusual Stereolith thing?

and as to "diversified opinions" -
You write that opinions are not so diversified amongst real "experts" and that You and Floyd Toole agree "on virtually everything " but don't the results of Toole's research sugest that early lateral reflections are good?

for me this seems to be one of the conclusions of this abstract: http://www.linkwitzlab.com/Acoustics and Psychoacoustics of Loudspeakers in Small Rooms.doc

isn't it inconsistent with what You say and consistent with what Moulton says "quite incorrectlly"?

clearly I don't understand something :(
if it is a misunderstanding could You clear it up please? :)
 
gedlee said:

I think that we are talking about different speakers. The Gradients that we used did not have horns. They were a readily available "Hi-End" loudspeaker of some rupute, which made them a good basis for comparison. But they were entirely direct radiators.
Unfortunately I don't have any measurements from the Gradients.

Hi Earl,

The Gradients tweeter is mounted coaxially at the throat of the midwoofer - hence my "non-optimal" horn comment. The cone does indeed provide horn loading and directivity control.
Here are some measurements
Gradient Revolution
Here is the Seas drive unit (or an extremely close relative) Seax coax
http://www.ele.tut.fi/~artoko/audio/speakers/coax/coaxial measurements.html
http://www.ele.tut.fi/~artoko/audio/speakers/coax/coaxial%20polar.pdf
Clearly the tweeter horn loading is not as refined as the Summa, with reflections of the surround and a moving cone around the throat, yet it seemed like this listening panel was not bothered by it. Was the directivity characteristics of the Gradient the dominant factor?
Here is the raw tweeter response of a similar Seas coax T18RE


The Gradients were very good sounding speakers, granted, but here is the thing. They were 12 dB LESS efficient than the Summas with a least 20 dB less Max SPL capability. Thus, while they were good sounding, they would never even come close to be acceptable at high volumes such as in a Home Theater playing a live performance of Cream (for example). Its a whole lot easier to get good sound with limited headroom than it is to get the same sound with 20 dB more output.

No arguments there, the Summa uses SOTA pro drivers, the Gradient, a modest Seas home unit. I think it further illustrates my point as to the very different speakers resulting in a statistical tie. As I noted before, the Gradient is probably not intended for as high SPL output as the Summa, but there is no reason why the Gradient platform could not be used with pro units, such as a B&C coaxial, etc. if the goal was live SPL capability.

cheers,

AJ
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.