These days, most of the better room-correction systems give you the ability to set an upper limit for the frequencies being “corrected.” And for my money, it’s not only the most useful feature of such systems but also the most misunderstood.
The exact terms for this feature vary from company to company. Anthem’s ARC Genesis room-correction software refers to this setting as the “Maximum Correction Frequency.” The Audyssey MultEQ Editor app dubs it the “MultEQ Filter Frequency Range.” Other room-correction systems use their own nomenclature, but the principle is the same: when you engage this setting, you’re effectively standing on the target frequency response curve of your room like Gandalf on the Bridge of Khazad-dûm, telling your room-correction system, “You shall not pass.” In other words, “Don’t filter the output of my sound system above this frequency.”
Digging back through both my personal and professional e-mail inboxes (as well as my social media DMs) for the past few years, I’ve found that the vast majority of questions readers have asked me have been about this setting. That’s probably because in every A/V receiver or preamp review that I’ve written in recent years, I’ve either spelled out my max filter frequency settings (with room-correction systems that allow for such) or lamented the lack of this feature on room-correction systems without such capabilities (such as Yamaha’s otherwise-good YPAO R.S.C. Multi-Point system).
All of this of course raises a question: why would you want to impose limits on your room-correction system to begin with? After all, aren’t these systems designed to listen to the sound of your speakers in your room, determine how the room itself is coloring the sound before it reaches your ears, and reverse as much of that coloration as possible? Indeed, they are. But to understand why you may be better off “correcting” only certain frequencies, you first need to understand that not all sound waves behave the same in your room. (If, by the way, you’re familiar with all of this and simply want to know how I find the right maximum filter frequency for a given room, feel free to skip down to the Dirac Live screenshots below.)
Simply put, below a certain frequency, your room primarily acts as a resonator. And above that frequency, your room begins to behave more as a reflector. Or, if you’d prefer a rather hackneyed simile, bass-frequency sounds in your room behave more like the water in the deep end of the wave pool at your local water park, whereas midrange- and high-frequency sounds behave more like the balls on a pool table.
The German physicist Manfred Schroeder first documented this phenomenon back in the 1950s, and we now refer to the crossover frequency between these two acoustical zones as the Schroeder frequency. And the reason we’re so concerned with finding this Schroeder frequency for any given room is that below that point, the biggest bugbears in terms of room acoustics for any listening space are standing waves, which are caused by sound coming from your subwoofer or speakers interacting with that same sound reflecting off the boundaries of the room.
When these waves are in phase, they reinforce one another, resulting in increased amplitude, or sound pressure levels (SPLs). When they’re out of phase, they effectively cancel each other out, resulting in decreased amplitude (decreased SPLs). To hear this phenomenon for yourself, simply send a low-frequency test-tone to your subwoofer and walk around the room. As you move from spot to spot, it will almost certainly sound like someone is mucking around with the volume knob on your receiver or preamp.
When low-frequency sound waves from your subwoofer or speakers (red) interact with their own reflections from the boundaries of your room (blue), the resulting interference causes significant increases and decreases in amplitude (green) that are highly dependent upon frequency and seating position.
Above the Schroeder frequency, the quality of the surfaces within your room (primarily how reflective, absorptive, and/or diffusive your walls and furniture and decorations are) have more of an impact on sonic colorations than do room dimensions and speaker placement.
But what does all of this have to do with room correction? Well, this is a bit of an overgeneralization, but I think most room acoustics experts would agree that while better room-correction systems can do a pretty good job of dealing with standing waves, even the best of these systems can do more harm than good when it comes to combatting higher-frequency problems.
And then there’s the question of whether or not we even need to “correct” problems in the mid and upper frequencies at all. I tend to side with Dr. Floyd Toole, whose book, Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms, contains one of my favorite quotes on the subject: “Two ears and a brain comprise a powerful acoustical analysis tool, able to extract enormous resolution, detail, and pleasure from circumstances that, when subject to mere technical measurements, seem to be disastrous.” In other words, if I may be so bold as to paraphrase such a renowned researcher, we tend to “hear through” most acoustical problems affecting mid and high frequencies, since our brains are perfectly capable of “correcting” them.
That said, neither my brain nor my ears are too happy when subjected to boomy, bloated, one-note bass. So I’m a big fan of room correction, but only to a point. The question remains: how do we find that point?
Getting back to the Schroeder frequency, which is essential to finding the right max filter frequency, there’s a standard formula for calculating this transition point in a room: 2000 times the square root of T/V, where T equals the room’s reverberation time in seconds and V is the volume of the room in cubic meters. The problem is, that calculation is really only valid for larger concert halls and auditoriums, where the reverberation time is significant. In smaller spaces, like your average den or media room, acoustical asymmetries (like the location of doors and the relative rigidity of different walls) mean that we really have to rely on measurements and a keen eye to find the Schroeder frequency.
Thankfully, many of the better room-correction systems will allow you to do this, though none of them hold your hand through the process. Let’s start by looking at some measurements of my main media room taken with Dirac Live a few years ago on my old Emotiva RMC-1 preamp-processor. (These screenshots are from an older version of Dirac Live, but the principles remain the same in new versions, as well as in room-correction systems like ARC Genesis. Just make sure you look at all of your measurements, not merely the average spectrum).
As you can see, there is a pretty clear delineation between the two primary zones of audio in these graphs. Below 200Hz, the peak-to-peak variations in the actual measured output of my surround speakers are significantly larger than the peak-to-peak variations above 200Hz. Not only that, but below 200Hz there’s much less-consistent overlap between the frequency responses of each speaker as measured from different positions, and much less consistency between the average response of two identical speakers placed symmetrically in the room, equidistant from the main seating position.
Those are my visual clues that the Schroeder (or crossover) frequency of my main media room is right around that point, which I’ve marked with a red line. Incidentally, it’s probably pretty close to the same in most mid-sized media rooms, so if you don’t want to do the work of figuring out where your own room’s crossover frequency is, 200Hz is a super-safe guess.
And if you’ve kept up to this point, you may have logically concluded that this is where I would set my maximum filter frequency. But not so fast. Keep in mind that this is a crossover frequency, not a brick wall. The transition between the “resonant” region and the “ray” region of a room extends roughly two octaves above its Schroeder frequency. Each octave corresponds to a doubling of frequency, so I can safely assume that by 800Hz (200Hz doubled and doubled again), standing waves have pretty much completely ceased to be an acoustical gremlin in my room. And so—roughly speaking—that’s where I’ll set my maximum filter frequency.
Why roughly and not exactly? Well, remember that we’re eyeballing our Schroeder frequency here, so there is a baked-in margin of error. What’s more, I prefer to set my filters such that my target curves transition smoothly to the unfiltered output of my speakers above that point. If I set my maximum filter frequency right at 800Hz, I end up with a different target curve for my left and right surround speakers between 200Hz and 800Hz, which looks something like this:
Whether or not I would be able to actually hear such a mismatch between the target curves of my left and right surround speakers is highly debatable, but it would bug me knowing it’s there. So in a case like this I would bump my maximum filter frequency up or down ever so slightly until I arrived at a consistent target curve for both speakers—something like this:
If the resolution of these screengrabs isn’t sufficient to reveal the differences, in the second image I merely bumped my max filter frequency up from 800Hz to ~810Hz, which resulted in a more consistent target curve between my two surrounds.
You can see the results of setting such a max filter frequency in the graph below:
But it may be more instructive to simply look at the final average spectrum of my surround speakers after room correction:
Chances are, some of you are left positively apoplectic by the mismatch between the smooth frequency response below 810Hz (ignoring a few significant dips that Dirac avoids filling completely so as to avoid amp clipping) and the relative chaos above that point. But speaking from experience, this approach to room correction results in a far better-sounding system overall, with tight and well-controlled bass but without the deadening of sound or diminishment of soundstaging that can result from applying full-spectrum room correction in my room.
In some systems, by the way, full-spectrum room correction can lead to the exact opposite problem, overemphasizing or introducing harshness in the upper frequencies. But either way, the result can be a perceivable change in the timbre of your speakers—which, presumably, you bought because you like their timbre. (To be fair, Dirac Live does much less harm to mid and upper frequencies than most mass-market room-correction systems, and it even allows you to draw a target curve that hugs the in-room response of your speakers. That said, I still prefer the sound with this zone unfiltered).
As mentioned above, many of the room-correction systems that will allow you to set a maximum filter frequency (and not all will) also give you access to a graph of all your measurement positions. But one of the most popular such systems—Audyssey’s MultEQ Editor app for iOS and Android—gives you only a rough, averaged graph of all the measurements for each speaker, like so.
Looking at that, you can still approximate the Schroeder/crossover frequency by looking for the transition from larger peak-to-peak variations in frequency response to smaller ones, again, centered right around 200Hz. (By the way, don’t freak out about big spikes and dips like those you see around 40Hz and 60Hz. By the time you apply subwoofer crossover settings in your receiver or preamp, your main speakers won’t be putting out appreciable energy at those frequencies anyway.)
But I’ve run Audyssey in rooms where the peak-to-peak variations weren’t quite so stark. What I do in these instances is whip out my image editor (PaintShop Pro is my preference, but Photoshop or free alternatives like GIMP work equally well), pull up screenshots of measurements for two identical speakers positioned as symmetrically as possible in the room, erase the black background from one of the graphs, change the color of the frequency response, and merge them. The process takes only a couple of minutes, and the results look like this.
Keeping an eye out for the spot where the in-room response of these two identical and symmetrically positioned speakers start to converge (or at least become more similar), I would eyeball the crossover frequency of this room as somewhere close to 210 or 220Hz, resulting in an upper boundary of the MultEQ Filter Frequency Range between 840 and 880Hz. And frankly, that’s close enough for horseshoes, hand grenades, and room correction alike. At least for our purposes.
Again, though, as long as your listening room is average sized (somewhere in the neighborhood of 2500 ft3), you’re probably safe assuming a ~200Hz Schroeder frequency and setting an ~800Hz maximum filter frequency for your room correction system. If you’re still plagued by egregious acoustical problems above that point, they’ll almost certainly be better treated by judicious application of absorptive or diffusive room treatments, or just careful placement of bookshelves and/or draperies and moving any large reflective surfaces like mirrors or framed artwork.
If you have a much larger dedicated home theater room (say, 4500 ft3, give or take), your Schroeder frequency will likely be closer to 150Hz, meaning you’ll probably get the most benefit from a max filter frequency closer to 600Hz. Then again, if that’s your room, chances are you’re paying someone else a wad of dough to make these sorts of decisions for you.
. . . Dennis Burger
dennisb@soundstage.com