I'm never going to hear a megaspeaker in a good room am I?


Was thinking about something. There’s a thread about good $40K speakers which made me think that honestly despite hearing a lot of them at shows, I’ve never heard one in a decent environment. Now, perhaps we can argue:

If it doesn’t sound good anywhere, including a hotel room, is it really that good a speaker?


But let’s not go that route. But I am thinking to myself, in well treated rooms the best speakers I’ve heard were merely mid-range Wilsons and Magicos. I say "merely" because they were under $40k, not because of performance. The two best speakers I’ve heard, in medicore rooms were the SF Stradivari and Snell A/III, and top of the line Vandersteen.

All the $40K + speakers I’ve heard have been at shows, and either very badly treated rooms, or in halls so big the first reflection point was like a mountain echo. Am I ever going to get to listen to $40K+ speakers in great rooms anywhere again??

As a result, I’ve developed a severe bias against the performance of mega speakers, because I only ever hear them in terrible rooms and have not heard one I’d spend money for, and honestly that's unfair to them.

erik_squires

Showing 7 responses by audiokinesis

Erik posted:

" I don’t mean to nit pick but you listed this as beneficial:

1. Increased time delay between the first-arrival sound and the onset of the lateral reflections, and a generally increased decay time.

"Did you mean the last part? I thought generally in home rooms we want to decrease the decay time? Maybe I’m not understanding what ideal is here."

Excellent question!

Reflections done right are your friends!! And "done right’ STARTS with the reflections having the same spectral balance as the direct sound, or nearly so... a little bit of rolloff in the highs is normal. But not too much - note that Peter Snell included a rear-firing tweeter in the Type A to fill some of the top-octave energy which would otherwise have been missing from the reverberant field.

So everything I’m going to say in this reply is said with the ASSUMPTION that the off-axis energy’s spectral balance is essentially correct (which in practice tends to be the exception rather than the rule). This is not going to be an all-encompassing answer because this is a big subject.

By way of background, in home audio, I consider the transition between detrimental "early" and beneficial "late" reflections to be about 10 milliseconds. This figure is not arbitrary. Researcher David Griesinger finds the ear to be especially sensitive to aberrations between 700 Hz and 7 kHz, and according to Earl Geddes the mechanism by which the cochlea perceives sounds within this region implies that "you need a 10 millisecond reflection-free window if you want no coloration or imaging effects.”  

(Early reflections are not entirely detrimental - they DO increase the soundstage width, or "Apparent Source Width" [ASW], to use Floyd Toole's term.) 

ALL reflections convey spaciousness, but the earliest ones do so at the expense of clarity and imaging precision. Early reflections may also cause coloration, with arrival angle playing a role (the closer to the same direction as the direct sound, the worse). This would seem to imply that the floor and ceiling bounces are especially detrimental, but in practice they are not (imo the reasons why may be fairly complex).

Late reflections convey spaciousness without the detrimental effects which early reflections can have, so we want to preserve them as long as they are not too loud or last for too long, which is seldom the case in a normally-furnished home listening room. The wider the speakers’ radiation pattern, the greater the chance that we will indeed have too much in-room reverberant energy - but imo it’s probably more likely that too much reverberant energy is having its detrimental effects due to excess EARLY reflections rather than excess LATE reflections.

Spectrally-correct late-arriving reflections also help to convey rich timbre. This is one of the things the MBL Radialstrahlers do exceptionally well, especially when positioned far enough from the walls to avoid significant horizontal-plane reflections within the first 10 milliseconds. Spectrally-correct reflections also help convey liveliness.

So early reflections are a two-edged sword, but later reflections (done right) are virtually always beneficial. Therefore in my opinion we want to minimize the early reflections but encourage the late ones. And if we use a room treatment approach which is aggressively aimed at reducing the decay time, imo we are killing off some of our beneficial late reflections... that is, "beneficial" assuming they were "done right" (spectrally correct) in the first place.

Duke
The three speakers Erik identifies as having sounded good in "mediocre" rooms are all speakers which have relatively smooth off-axis response as well as other characteristics which generally result in good in-room response: The Vandersteen Model 7, the Sonus Faber Stradivari, and the magnificent Snell Acoustics Type A. In other words, these are not three random speakers that Erik’s ears picked out - they are three speakers which start out doing some important things right when it comes to room interaction.

If placed in a larger and more theoretically ideal room, the main beneficial differences would be:

1. Increased time delay between the first-arrival sound and the onset of the lateral reflections, and a generally increased decay time. This results in less degradation from early reflections as well as less "small room signarture", so we hear more of the acoustic signature on the recording, and less of the playback room.

2. The bass would be smoother because the larger room would result in greater modal density, which results in more numerous and smaller (and therefore less audible) room-interaction peaks and dips.

Wouldn’t it be nice if these attributes could be grafted into our typical smaller room? How would we go about doing so?

1. Let’s take the fairly wide and uniform radiation pattern of the speakers Erik likes, and chop it in half. Let’s aim half of the energy (now in a considerably narrower pattern) at the listening area, and aim the other half well away from the listening area, such that it bounces off the room boundaries a time or two before arriving. This does two things: It reduces the amount of energy in the earliest (most likely to be detrimental) reflections, and also pushes the "center of gravity" of the reflections back in time, approximating what would happen in a larger room.

2. Various techniques for improving the in-room performance in the bass region exist, so I’ll leave it at that for now.

Room interaction matters a lot to me so I am commercially involved with speakers which have these sorts of attributes. The ones which would qualify as "megaspeakers" (using the $40k yardstick) are SoundLab fullrange electrostats.

Anyway I do not subscribe to the school of thought which says that it’s up to the homebuyer to own a megaroom and up to the acousticians to make the megaroom work for the megaspeakers. In my opinion good room interaction starts at the loudspeakers design stage, whether those speakers be "mega" or otherwise. Kudos to Richard Vandersteen, Franco Serblin, and Peter Snell for paying attention to room interaction.

Duke
Erik wrote:

"don’t most listening rooms have too long of an RT 60 to begin with, not to mention, it is usually pretty uneven.’

In general when we move into a larger listening room the RT60 will correspondingly tend to be longer, but the larger room STILL sounds better! So clearly RT60 is NOT telling us the whole story.

Making the RT60 more EVEN is of course desirable, BUT I have a question about the unevenness you mention: Is it based on measurements using loudspeakers whose off-axis energy is uneven? If so, is that uneven off-axis energy what’s showing up as "uneven decay"? If that’s the case, then the problem originates with the speakers, not the room.

The problem with a room acoustic approach which focuses on RT60 is that it does not target those reflections which are most likely to be detrimental (the early ones), but instead tends to have a heightened effect on the beneficial later reflections ("beneficial" assuming we’re talking about spectrally-correct reflections). This is because the early reflections will only be attenuated by the absorptive material ONE time, while the later reflections will bounce around the room enough that they may well strike the absorptive material MULTIPLE times, especially if there is a lot of it.

Imo minimizing detrimental early reflections is best accomplished by loudspeaker design and set-up, and if we still need to address the in-room decay smoothness, we can do so without the additional challenge of trying to compensate for the speaker’s uneven off-axis response.

Duke

Hi Erik,  

I certainly don't expect you to simply accept my claims, and I thank you for hearing me out. 

Best wishes, 

Duke
@erik_squires, in a big room like an auditorium, with listeners being far away from the speakers, the direct-to-reverberant sound ratio is MUCH lower than in a home audio setting, where we are much closer to the speakers. In a big room the reverberant field conveys so much more energy than the direct sound that intelligibility is an issue. The faster we can get that powerful reverberant field to die away, the less "noise floor" the direct sound has to overcome.

Also, in a large room the reverberant sound field is "statistical" - that is, the level of reverberant energy is effectively identical throughout the room. There are SO MANY reflections coming from SO MANY directions that WHERE we place our acoustic panels makes little difference on their net effect.

Concert halls, and larger ones in particular, have issues with excess reverberant energy, which makes seats near the back of the hall (where the direct-to-reverberant ratio is lowest) audibly inferior to seats near the front. The sound goes from "immediate and attention-grabbing" near the front of the hall to "muddled and less engaging" near the back of the hall. The cause is more complex than simply the direct-to-reverberant sound ratio (the earliest reflections are the worst even in a good concert hall!) but that’s still a major factor.

In our small rooms, in the absence of obvious issues like slap-echo, we have the opposite situation: Our direct-to-reverberant sound ratio is typically MUCH higher than in any performance venue. This is largely because we are typically within maybe eight to twelve feet of the speakers. The reverberation times in our small rooms are also much shorter because the reflection path lengths are much shorter, meaning that within a given time interval the sound will have been attenuated by room boundaries or room furnishings many times more often.

Ime there seems to be a "sweet spot" as far as how loud the reverberant sound is relative to the direct sound. The larger the room (the longer the time delay before the "center of gravity" of the reflections), the louder the reverberant energy can be before clarity suffers. Hence most good seats in a concert hall have a direct-to-reverberant sound ratio which would be WAY too low for home audio, but it’s okay because of the much later arrival times of the reflections.

You mentioned uneven decay times - how are you going to find out whether your room has uneven decay times, so that you can figure out what to do about them? You have to make measurements. And if the measurement process INCLUDES your loudspeakers, then inevitably it INCLUDES their off-axis response. Yes there are omnidirectional loudspeakers designed specifically to measure decay rates in large halls, but I doubt very many audiophiles have access to such.

Based on conversations with acousticians, it is much easier to fix the room with acoustic treatments than it is to fix the room AND the loudspeakers at the same time.

Duke
@erik_squires wrote: "So in a small room, I can’t imagine actually wanting to enhance the reverberant amplitude or timing per se. To my ears, the congestion in clarity, especially dialogue suffers too much."

Might that "congestion" be due to an excess of early reflections? Quoting Dr. David Griesinger:

"When presence [clarity and immediacy] is lacking, the early reflections are the most responsible."

If Griesinger is correct, then chasing a reduced RT60 is like trimming the tail of the dragon... it helps, but your main problem is at the front end.

I don’t doubt your observations, but unless I am mistaken, you are making them based on using speakers which have a wide enough radiation pattern that early reflections are virtually inevitable, and those reflections probably are not a particularly good spectral match with the direct sound. Please don’t take this as an insult of your speakers - their designer and I simply have different ideas about "what matters most", and he may be right and I may be wrong, or we may both be wrong.

I would much rather make my adjustments to the direct-to-reverberant sound ratio at the loudspeaker, rather than by using room treatments. I’m NOT down on room treatments - diffusion can be delicious - but imo there are issues which are better addressed elsewhere in the chain.

I probably should have make it clear that I’m coming at this from a very different and unorthodox direction, loudspeaker-wise: I’m starting with a narrow-pattern, controlled-directivity speaker which will sound overly dry (but have great clarity) in just about any room, then adding just enough reverberant energy to enrich the timbre and spatial qualities without degrading clarity. And in my experience "it sounds right" seems to be a fairly specific and consistent point on the continuum in a given room.

Duke
@erik_squires, likewise I thank you for the thoughtful engagement!  Much more satisfying than a squabble.  I'll try to avoid those in the future. 

"I don't see a reasonable model of room treatment where early reflections are not controlled, echoes removed, and reverberation is NOT naturally reduced." 

Nor do I! 

That's why, in my opinion, the preferred starting point is the loudspeaker, and specifically its radiation pattern.  Of course if the speaker sucks who cares what its radiation pattern is, so this is just ONE aspect that imo is worth paying attention to. 

Best wishes, 

Duke