Few of us are in a position to fit our houses with false walls, suspended ceilings and floating floors — so how can we create a reliable monitoring environment? Paul White explains that it's not as daunting as it seems.
We've all been there — the mix sounds great in the studio, but as soon as it's played back on another system, the balance is all wrong, the bass is wildly out of control, and what seemed like a masterpiece at the time is now just plain embarrassing. A high proportion of the readers' phone calls I take are from people who are concerned about their choice of monitors, so there's obviously a degree of awareness of the problem, but there's more to monitoring than simply buying a good pair of loudspeakers; working in a room with bad acoustics can have a very profound effect on what you hear while monitoring.
What are we actually listening to when we sit down in front of our consoles to mix? Obviously, we hear the sound coming directly from the monitor speakers, but added to that is a significant amount of reflected sound, as our latest mix bounces off every reflective surface in the room. The reflected sound arrives at our ears later than the direct sound, and because some frequencies are absorbed more than others, the reflected sound is also highly 'coloured'. An instinctive first response might be to suggest mixing in a totally dead or acoustically absorbent room, which should remove all reflected sound, leaving us with just the pure sound of the speakers. In theory, this is fine, but in practice, we can't ever create a completely absorbent room because of the need to have equipment in the studio, and all equipment has reflective surfaces, especially the mixing desk. More significantly, we are used to living in a moderately reflective acoustic environment and the vast majority of people find totally anechoic (non‑reflective) rooms psychologically disturbing. On a more practical note, to create a truly anechoic room which absorbs energy right across the audio spectrum requires the walls to be lined with an absorbent material, such as Rockwool, to a depth of many feet, which is impractical in most commercial installations, let alone home studios and MIDI suites. Even if it could be achieved, you'd need a very powerful monitoring system because the perceived sound level is much lower in a totally dead room.
But I've left what I consider to be the most powerful argument against anechoic monitoring environments until last: most people listen to their music either in a domestic living room or in the car, neither of which is remotely anechoic. If music is to be played back in a normal room, it makes some kind of sense to mix it in a room with similar characteristics. This argument breaks down if you need to use a very large monitoring system or one with an extended bass response, but for the person working from home with small or medium sized monitors, a reasonable monitoring environment can be created with very little effort or expense.
Monitor Placement
A properly set up hi‑fi system will have the speakers arranged symmetrically with no obstruction between the speaker and the listener. In the studio, however, we invariably find a mixing console occupying the intervening space. The current trend is to position near‑field monitors on top of the meter bridge of the mixing console, but this is really courting acoustic disaster. The near‑flat surface of a mixing console makes it an almost perfect reflector of sound, so if you put your monitors on the meter bridge, what you'll hear is the direct sound from the monitors followed closely by a strong reflection from the desk. This reflection follows the original so quickly that you don't hear it as an echo; instead it combines with the direct sound to produce a tonal change similar to weak phasing.
The solution is quite simply to put the monitors on stands a little way behind the mixing console. Some sound will still be reflected from the console's surface, but at so low an angle that it will miss the engineer's ears. Figure 1 shows the direct and reflected sound path for both console‑top and stand‑mounted monitors.
With the speakers moved a little further back, it's also more likely that the distance between the monitors and the engineer's chair will be correct. Ideally, the speakers should form a (roughly) equilateral triangle with the listener which, in most home studios, places them between three and five feet from the listening seat. A further consideration is that monitors should not be placed too near corners, as this will produce misleading bass 'hot spots', where some bass notes will appear to be louder than they really are.
In a room with imperfect acoustics, near‑field monitoring offers, by virtue of its close proximity, the advantage that the listener hears a greater proportion of the direct sound from the loudspeaker than the sound reflected from the walls, floors and ceilings of the room. This is an important advantage, because in an untreated room the reflected sound won't have the same characteristics as the direct sound — it will, in effect, be 'EQ'd' in an unpredictable way by the physical properties of the room and any furnishings or equipment within the room. It's also true that most untreated rooms perform least accurately at the low bass end of the spectrum, so using small or medium sized monitors that don't generate very low bass neatly sidesteps that problem. This doesn't mean that you have to put up with thin‑sounding monitors; you can can still use speakers that go down to 50Hz or so — just don't expect to be able to reproduce 20Hz organ pedal notes with any degree of accuracy.
The Monitors
Monitors themselves have an important role to play in accurate mixing, and though the paper specification of just about any monitor you might consider buying will almost certainly look good, the truth is that there are greater differences between different monitors than between just about any other component in the audio chain (with the possible exception of microphones). A well designed monitor will have a reasonably flat frequency response between, say 70Hz and 18kHz; its bass response should fall away gently rather than suddenly; it will have low distortion when used within its power range; and it will have good dispersion characteristics. This latter point is very important — it isn't sufficient that the on‑axis sound should be true and accurate: the off‑axis sound should also have a reasonably flat frequency response. If this is not the case, then off‑axis sound reflected from the room surfaces will be further coloured by the off‑axis peculiarities of the speaker. If you have a pair of monitors that seem to sound radically different when used in different rooms, the chances are that their dispersion characteristics are to blame.
Staying with dispersion for a moment, all multi‑way loudspeakers (those with separate tweeters and bass/mid drivers) produce an acceptable sound over a wider angle in the horizontal plane than in the vertical plane, due simply to the physical geometry of the driver positions. If you place these monitors on their side, the acceptable listening angle is more than halved, which means that you only hear an accurate mix if you're sitting exactly mid‑way between the speakers. It follows from this that mounting the speakers upright is a much better bet if you have the space to do so. If you're forced to lay the speakers on their side, angle them in so they're pointing directly at the engineer's chair and place them with their tweeters outermost to give the widest possible stereo image.
More Room Problems
So far we have a pair of decent monitors and, hopefully, an amplifier capable of providing at least as much undistorted power as the speakers can handle. The monitors are sensibly positioned behind the console, away from corners, and the engineer's chair forms the apex of an equilateral triangle with the monitors. In a small room, this often means setting the speakers up against the longest wall simply to avoid getting the speakers too close to the corners (and by too close, I mean ideally not closer than 18 inches). But we're not out of the woods yet, because room reflections can still interfere with what we're hearing.
A common problem is flutter echo caused by sound bouncing back and forth between two parallel surfaces, the outcome being an audible 'ringing'. Try clapping your hands in a room with hard, parallel walls and you'll hear quite clearly what I mean by ringing. Professional studios are built with non‑parallel walls, but in a typical house, moving the walls is rarely an option. Fortunately, as you'll be spending most of your time in the engineer's chair, you only have to worry about curing flutter echo in that position. The easiest approach is to fix a square yard (or metre if you like) of acoustic foam tile to the wall directly to each side of your normal listening position and centred at your (seated) head height. As well as reducing the flutter echo, this treatment should also clean up the stereo imaging quite noticeably. Low cost, 2‑inch foam tiles will work quite adequately, though a similar thickness of fire‑proof, open‑cell upholstery foam will do almost as well; a thin fabric covering will make it look more acceptable. You can easily tell if the foam is the open‑cell type; if it will soak up water, it's open cell.
Don't be tempted to try carpeting all the walls instead, because most carpet absorbs only the very high end of the audio spectrum, leaving the bass and mid‑range to bounce around uncontrolled. The result is a boomy, boxy‑sounding room. Figure 2 shows a practical monitoring layout with sound absorbing tiles used to damp out flutter echoes.
The Final Frontier
The remaining major trouble spot is the wall directly behind you, because any sound from the monitors that doesn't hit you is going to bounce off it and come back to you a short time later, seriously compromising the quality of your monitoring. If the wall is more than ten feet or so behind you, then you don't really have to do very much. You could, for example, put a soft settee along the bottom of the wall and fix a heavy curtain, folded into drapes, above it, spaced two or three inches from the wall. Alternatively, you could break up the flat geometry of the wall by fitting it with shelves for your tapes, manuals, records, CDs and so on, to help scatter the reflections — though some absorptive material (such as the settee) is still recommended.
If you have an even smaller room and the back wall is only a few feet behind you, the ideal solution is to make the wall 'disappear', as far as audio is concerned, by covering it entirely with absorptive material. A 6‑inch depth of rockwool slab fixed between battens should be adequate for use with near‑field monitors, and a simple porous fabric such as hessian is fine for covering the finished 'trap'. However, if you don't want to go to these lengths, consider hanging a heavy curtain over most or all of the rear wall and then fix a row of acoustic foam tiles to the wall, behind the curtain, at head height.
On Reflection
The last point to cover is the floor of your monitoring room. If at all possible, a wall‑to‑wall carpet should be fitted, as this helps kill floor‑to‑ceiling reflections, at least at the high end of the spectrum where ringing might be a problem. Symmetry should extend beyond the monitor placement, so try to match reflective surfaces on one side of the room by a similar area of reflective surface on the other. However, avoid having flat reflective surfaces directly facing each other, as this reintroduces the danger of flutter echoes. If you have a window, try putting vertical blinds up which, if left half open, will reduce high frequency reflections by a significant amount. And because glass isn't a particularly good isolator of low frequency sound, you won't get much low frequency reflection either — it'll go straight through. If you have equipment racks on one wall, balance them as much as possible by shelving on the opposite wall, and if you have room for more soft‑furnishings, this will help damp down room resonances. In this respect, putting absorbent furniture in the rear corners of the room probably has the most beneficial effect. It follows that a typical bedroom studio with the bed still in it will probably make quite a good mixing room with very little modification.
Finally, get used to your monitoring system by playing your record and CD collection in the studio and keep a few test CDs handy for comparison purposes when mixing. And — I make no apology for repeating this — take a break after recording before you come to mix. Give your ears a chance to recover and you're far more likely to come up with a good mix.
Outside Influences
After a hard day in the studio, it can sometimes be difficult to hear just what's wrong with a mix, and though I've mentioned this technique once or twice before, it bears repetition because I count it as one of the most important aids to achieving a good‑sounding balance.
Once you've got your mix up and running, open the studio door and stand outside in the corridor or in the next room. Now, instead of your ears struggling to make sense of both the direct and reflected sound, you'll hear a single, homogonous, largely mono, reflected sound — and, for whatever reason, any imperfection in the balance sticks out like a sore thumb. Now you'll be able to hear if the hi‑hat is too loud or the vocal line too quiet, even if it's only by half a dB. Make any necessary adjustments, then come back out of the studio to check the balance. Once you're happy with the mix as heard from next door, the chances are that it'll sound fine in the control room too.
The other check most professionals make is to play a cassette copy of the mix over a typical car system. Make a cassette copy without noise reduction for this purpose — even if your car system has the option of Dolby B or C, there's a risk that differences in machine alignment will cause an overall change of tone, so it's better to put up with a little more hiss and get a more honest overall tonal balance.
Why Not Equalise The Room?
Why bother with acoustic treatment? Why not simply put a graphic equaliser in the system and adjust it until a combination of the direct and reflected sound seems satisfactory? Well, that used to be an accepted way of working, and even today equalisation is used to compensate for minor room deficiencies, but realistically, equalisation is of very little practical use in this context. Firstly, our ears don't just average out the direct and reflected sound — they have the ability to 'focus in' on the direct sound and so take less account of the refllected room sound than you might think. Secondly, and perhaps more importantly, a poor listening room sounds bad because its natural reverberation time is wildly different at different frequencies. Room reflections and re‑reflections are time‑domain problems, in that they occur after the original event, and no amount of EQ can change the reverberant characteristics of a room. Indeed, a room can be considered as a kind of mechanical equaliser which can only be adjusted by mechanical means — juggling with the area and type of reflective and absorptive surfaces until the reverberation time is more or less even over the whole audio band. Put simply, if you have good speakers in a bad room, at least the direct sound is correct, even though the reflected sound is wrong. If you then equalise the monitors in an attempt to correct the room, the direct sound from the monitors will now be wrong — and the reflected sound will still be wrong!