Last time out, in Can you hear what you’re doing? (Part 1), I talked about setting up monitoring to maximize its ability to “get out of the way” and allow better access to the recording itself. This time, we’ll take it to the next step and talk about addressing the room itself, whether a studio control room, a mastering room or a home listening room.
Those who have experienced a listening space with properly treated acoustics will understand why I said monitoring is more than just the speakers. It is the room in which one listens. It is where in the room the speakers are located. And where in the room the listening position is located. And where just about everything else in the room is too.
By determining speaker (and listening position) placement first, before anything else is added to the studio or listening room, we can maximize the ultimate potential of the monitoring by minimizing the contribution from the room. This is the starting point that determines how effective acoustic treatment can be when going to the next step in monitoring accuracy, which is getting the room out of the way.
The two main things to consider when addressing the acoustics of a typical studio or listening room are resonant modes in the bass and early reflections in the treble. A quick search of the Internet or the printed literature will reveal a wide variety of strategies.
Before going into what I’ve found that works, it is important to mention the idea of so-called electronic “room correction”. The basic idea behind room correction (sometimes also called “speaker correction”) is to measure the frequency response of the system in the room and then apply complementary alterations to the signal itself. In other words, special test signals are played through the system while a microphone captures the sound. The microphone output is sent to a device (nowadays, a computer) that will analyze the signal and provide a graphic representation of how the combination of the system and room responds at different frequencies.
In a typical frequency response measurement, all frequencies are fed to the device or system at the same intensity. Ideally, that device or system will then deliver all these frequencies at the same intensity. This would be called a “flat” frequency response. If the device or system caused certain frequencies to be exaggerated, the response would show bumps or peaks that correspond with the amount of exaggeration. If the device or system caused certain frequencies to be diminished, the response would show troughs or dips that correspond with the amount of diminution.
With electronic “room correction”, if the response shows a peak at a frequency of say, 100 Hz in the bass (i.e., the frequency of 100 Hz is being exaggerated), the input to the system will be altered so that it is fed a correspondingly smaller amount of 100 Hz. In theory, the end result will be equalized — the correct amount of 100 Hz is delivered by the system.
So much for the theory. I know a lot of folks who love the results of using such an approach to dealing with their rooms and I would never argue with whatever brings anyone their listening pleasure. I’ve listened to a number of designs for this electronic tack and to my ears, while the results always sound very different, they don’t sound better. They often sound quite a bit worse. After a closer look at the assumptions being made with the electronic approach, I found this isn’t at all surprising.
We’ll start with the assumption that the microphone hears the same way a human being hears. This might be true if the human had an extraordinarily narrow head with a single ear in the middle of their face (or perhaps on top of their head). Many electronic recipes attempt to deal with this by taking several measurements with the microphone in a different position for each, then averaging the results to arrive at the final response.
A larger assumption is that what the listener hears is the sum of the direct sound from the speakers combined with the sound of the room. It would be easy to get fooled into thinking this is the case if one looks at the sound in only one dimension — that of frequency response. But sound cannot exist independent of time. Without time, there is no sound. In fact, the direct sound from the speakers, having the shortest path to the listener’s ears, always arrives before any contribution from the room. The first thing we hear is the direct sound from the speakers. Then we hear the room responding. They are two things, not one.
Let’s look at what this means. With perfect speakers and a room resonance at say 100 Hz, the electronic system will drop the level of 100 Hz in the signal fed to the speakers. The formerly perfect speakers will now deliver a direct sound that has a dip at 100 Hz — the sound will be thinned, losing fullness and body. Since there is less of the frequency that excited the room, there will be less of the resonance from the room. This would be fine if the sound from the speakers and the sound from the room summed algebraically into a single entity and were not, in fact, separated in time.
When I listen to such a system, instead of the problem of the room being fixed, I hear instead, the creation of an additional problem. The electronic mode of addressing the room has taken one issue and turned it into two issues. The first arrival, the direct sound from the speakers, determines the overall character. In this case, I hear the loss of fullness and “meat” resulting from the diminution of 100 Hz in the input. Next, while somewhat lessened, the room is still resonating at 100 Hz.
This leads to another assumption, which is that those peaks and dips in the frequency response are the problems with the room. Here again, looking at the issue only in terms of frequency response is to miss the bigger picture entirely and that is this: room issues are time-based, not amplitude-based. They occur over time and are not merely changes in the level at which different frequencies are delivered to the listener. When the room resonates at 100 Hz, it is “holding onto” that frequency, “ringing” at that frequency, after that frequency has already stopped in the signal. The peaks and dips in frequency response are symptoms of room issues, not the issues themselves.
In the attempt to address the symptom, the source of the issue remains and an additional problem is created. While the level of 100 Hz in the example above might be diminished, as long as the resonance is being excited it will cause the time-based issue to exist. There will be less 100 Hz but what is there will still continue to ring after this frequency has already stopped in the input signal. The ringing fills in the quieter parts of the signal, obscuring low level detail and disturbing the pace of the music.
As I see it, attempting to address a time-based problem by using an amplitude-based “solution” is like trying to fix a broken arm by wearing a different hat. Much better, in my view, to address the problem at its source: the room itself. Earlier in this entry, I said that when addressing room acoustics, we need to consider resonant modes in the bass and early reflections in the treble. (Another thing about the electronic method is that it doesn’t even attempt to address early reflections. It is possible to imagine how this might be accomplished but it would involve even more degradation of the direct sound from the speakers.) The most effective means of acoustic control in my experience have been the modern day iterations of the “functional sound absorbers” engineer Harry Olsen first proposed more than half a century ago. These cylindrical “traps” are available as commercial products, Art Noxon’s excellent Tube Traps from ASC (Acoustic Sciences Corporation) or you can build variations on the theme yourself with instructions that can be found on the Internet. Cylindrical traps can have additional benefits in the treble range, which we’ll cover next.
The example above mentioned a single room resonance. Typical rooms will have several. There will be a fundamental resonance frequency for each room dimension (length, width and height). There will also be harmonics at double each of these frequencies and, depending on room dimensions, at quadruple these frequencies too.
“Controlling early reflections is as simple as putting absorbent material at the points where the reflections occur. There are two such points, one for each speaker, on each wall as well as on the ceiling (and uncovered, hard flooring). Imagine the walls and ceiling of the room are mirrors. From the listening position, you’d see a reflection of each speaker on each of the walls and on the ceiling. If the floor is not covered with a carpet or rug, you’d see a reflection of each speaker here too. An easy way of finding these points is to enlist the aid of an assistant who will hold a mirror up against each wall while you sit in the listening position. With the mirror at your (seated) eye level, the assistant moves the mirror along the wall until you can see the reflection of one of the speakers. When you can see one of the speakers in the mirror, you’ve found the point on the wall to place the absorbent material. Have the assistant continue moving the mirror until you see the other speaker. Now you’ll have the second place on that wall that will need the absorbent material. Do this for each wall as well as the ceiling if possible. A carpet or rug will work well to prevent early reflections from the floor.”
“While there are many types of foam sold for this purpose (and they are better than nothing at all), a much better solution lies with the modern iteration of the “functional” traps mentioned above. The best commercial designs as well as the better DIY (do it yourself) variety will be cylinders that have half their surfaces covered with a material that is reflective in the treble while the other half is fully absorbent. Too often, the use of foam to control reflections results in a dead feeling in the listening space that is neither natural sounding or comfortable. The common mistake is in the perception that if a little is good, more must be better. In fact, what is needed is absorption of the early reflections without affecting the later ones. Further, diffusing or scattering these later reflections contributes to the naturalness and comfort of the room. Cylinders with reflective halves allow the sound to be tailored as each cylinder is rotated. When placed at the early reflection points, the absorbent half can be oriented toward the speaker it is nearest, while the reflective half will help diffuse the sounds arriving later in time, in this way maintaining the natural ambience of the listening space.”
In addition to addressing room modes in the bass and early reflections in the treble, diffusion is the third element in room treatment. Many photographs of studios and listening rooms published in magazines or on the Internet show diffusors installed near the speakers and some manufacturers even advise their customers to do things like this. Placing diffusion near the loudspeakers (or behind the listener’s head or on the middle of the ceiling) will only ensure the early reflections that should be absorbed (and thus prevented from reaching the listener) are instead scattered, guaranteeing they will reach the listener’s ears. This hardens the sound, shrinks the soundstage and defocuses images. To avoid these, diffusion should be used only for late reflections. Place diffusors so they are facing away from the speakers. Other photographs will show studios with foam or other absorbing material completely covering the walls. To avoid a “dead”, closed-in sound, absorption should be used only at the early reflection points.
With cylindrical type sound treatments, I orient them so the absorptive half of the cylinder is aimed at the nearest speaker, thus, the “live”, reflective half is aimed away from the speaker. Traps that are equidistant from both speakers are oriented so their absorptive half faces the center of the room. These cylinders perform bass trapping, early reflection absorption and diffusion, all in a single device. With some placed at the early reflection points, the soft side performs the absorption and the “live” side only gets sound that has been around the room already.
By the way, there is another application for these which can work wonders in smaller recording studios. Using a loose ring of traps around what is being recorded, with the soft sides facing the center of the circle, the diffusive sides keep sending reflections from the room back to the room. When this is properly done, the mic(s) inside the circle “think” they’re in a larger room than they are really in.
Lastly, a few ideas for super inexpensive room treatment. These won’t work to anything like the same extent as the cylinders mentioned above but they will provide an idea of what proper treatment can do:
- Gather at least a dozen cardboard boxes (the light brown cartons in which items are delivered to supermarkets or in which mail order items are shipped). Ideally, these will be approximately 16” x 16” x 24” (~40 cm x 40 cm x 60 cm) in size. Two dozen boxes would provide an even better idea.
- Fill all of the boxes as tightly as you can with individually crumpled sheets of newspaper.
- Tape the boxes closed and stack them to create columns at least 6 feet (~ 1.8 meters) tall.
- Place a column in each corner of the room. If you have gathered more boxes, also place a column at the half-way point along each wall.
- To avoid reflections from the columns, cover them with thin towels, blankets or other absorbent materials. For a neater appearance, cover them with any soft, absorbent fabric you find attractive.
- The 5 steps above will (begin to) address the room’s resonant modes in the bass. To address the early reflection points, find them using the information from earlier in this entry and cover them with thin towels, blankets or other absorbent materials. Here again, an absorbent wall hanging may have more eye appeal.
- Now just listen to some music and notice the increased pitch definition in the bass, the increased sense of “punch” and tightness of timing, smoother treble, more easily audible low level detail and (if the recordings contain it) the improved sense of three-dimensionality and focus.
With speakers and listening position optimally placed, treating the acoustics of the room by addressing resonant modes, absorbing early reflections and diffusing later reflections, helps get the room out of the way, making it even easier to hear past the monitoring all the way to the recording itself.