We all want to improve the quality of our recordings, but there's often no need to rush out and buy new equipment. You just have to take a close look at the way you're using what you already have.
It's natural to always want to make better-sounding recordings, but in order to understand how to do that, it's necessary to separate equipment-related issues from the way you record. The problem could be the sound source itself, any one of a number of stages in the signal chain, or that one or more parts of the recording process are being handled in a less-than-optimal way. Finding the correct solution involves taking a hard, honest look at what you're doing and the way you're doing it, before rushing out to buy that esoteric valve microphone or 24-bit/96kHz converter.
The best place to start the examination is at the signal source. Judging by the reader enquiries we receive, vocals are still at the top of the list, but you don't need any special equipment or elaborate techniques to record good vocals. You do, however, need a good vocalist and a reasonable environment in which to record. Virtually any carpeted domestic room is fine, provided that you keep the mic away from the walls, but, at the same time, not in the exact centre of the room. The choice of mic will make a difference to the timbre of the recording, but a good singer will sound fine whether recorded using a dynamic stage mic or a high-end capacitor model. What's more important is to keep the singer fairly close to the mic to prevent the room acoustics intruding on the sound too much, but not so close as to allow proximity effect to make the sound bass-heavy.
A mic distance of six to nine inches is usually OK, with a gauze pop shield between the mic and the singer to prevent 'popping' on plosive 'B' and 'P' sounds. Record the result either flat or with a little compression and all you'll need to do at the mixing stage is add a hint of reverb to flesh out the sound. You may be able to improve the sound further by using a little EQ and perhaps more compression when you mix, but if the basic sound isn't good, you should be looking for causes rather than fixes. Boxiness is usually due to recording in too small a room or placing the mic too far from the singer, but it's easy to damp the acoustic by hanging sheets or blankets around the room — note that acoustic material is most effective when hung slightly away from the wall rather than being in direct contact with it. It's also important that the singer feels comfortable, which means setting up a good headphone monitor mix with enough reverb on the vocals to help accurate pitching.
When miking acoustic instruments, you need to be a little more careful with the choice of microphone, as some instruments are difficult to record adequately with dynamic microphones. Dynamic mics are fine for close or loud sounds, but lack the sensitivity and high-end response to record more delicate sounds. Even though you can get away with a dynamic mic to record an acoustic guitar if you're careful, you'll certainly find the result sounds clearer and cleaner if you use a capacitor or back-electret microphone.
Assuming you have a suitable mic, the most important areas to check after the quality of the source are the mic position and the influence of the room. Instruments are often recorded with greater mic distances than vocals, so consequently the room has more of an influence on the result. Nearby hard surfaces can add flattering reflections to the sound, but a room with hard, bare walls all the way round can sound boxy, especially if it is small.
Good results can sometimes be obtained in otherwise difficult rooms by positioning the player close to one reflective wall (ideally facing away from it), and either rolling back the carpet or placing a reflective material such as linoleum or hardboard on the floor under the player. This provides strong early reflections to reinforce the sound of the instrument, often making it seem closer and brighter.
Mic position is also vitally important, as acoustic instruments produce different sounds from different parts of their structure. You have to get the mic far enough away for all these components to integrate properly for the most natural sound, but, at the same time, not so far away that spill from other instruments compromises the recording or unflattering room acoustics dominate. In the case of instruments with a sound box, such as the acoustic guitar, you also have to be careful not to let the sound coming from the sound hole dominate the recorded sound, otherwise the tone may be noticeably boxy or resonant.
I have devised a very rough rule of thumb for recording acoustic instruments which should get you into the right ball park as far as positioning the mic goes. Even so, you should always take the time to experiment with moving the mic position rather than using a formulaic approach. Part one of this rule is: estimate the longest dimension of the sound-producing part of the instrument and use that as your initial mic distance. For example, most of the sound from a guitar comes from the body, which is around 18 inches long, so an initial distance of 18 inches would be a good starting point.
Part two of this rule is not to point the mic at the loudest component of the sound, such as the end of a wind instrument or the sound-hole of a guitar. On occasions, breaking this rule can produce good results, and many sax recordings are done with the mic pointing at the business end of the instrument, but in most cases you'll get a more accurate sound by backing off and aiming the mic away from the 'loud spot'. A further tip is that, if all else fails, put the mic close to the player's ears, because most acoustic instruments sound right from the player's perspective.
Throughout this article I've referred to the quality of the source sound and, after all, that is the event that you're making a record of, so it is clearly the most important element. The majority of unsatisfactory demos involving real, as opposed to sequenced, musicians are let down by timing and tuning problems, neither of which can be fixed in the mix (other than less serious monophonic pitch errors which can usually be patched up using a dedicated processor).
In the case of guitars, tuning problems aren't only caused by old strings or by not checking the pitch against a tuner before every take, but also by basic intonation problems resulting from incorrect setting of the bridge saddle positions or by nut slots that aren't cut deeply enough. These problems need to be sorted out before you record, not during a session, and if you can't do the job yourself, get a guitar tech to sort the instrument out for you — it's well worth it. Tuning issues can also be caused by pressing too hard with the left hand when fretting notes, particularly chords, or by inadvertently dragging strings over to one side slightly. These flaws may go unnoticed at gigs, but they'll stick out very obviously on a recording.
Sadly, timing can't be fixed by mechanical means alone so the only solution is to rehearse extensively and listen very carefully to your recordings to see that everything is sitting together properly, especially the bass and drums. Don't assume that adding overdubs or effects will hide timing problems either, because they seldom do. And rather than mess around editing and copying sections to patch up the song, you'll probably find it's much quicker and better sounding just to play it again, provided that you notice the problem at the session and not some days later when all the gear is packed away!
When recording electric guitar amplifiers, you can afford to break a few rules, because you're after the artistically 'best' sound, not necessarily the most accurate. Even so, the basic mic rule works. A typical guitar speaker is around 12 inches in diameter, so setting up the mic 12 inches from the front of the cabinet will generally produce a workable sound. However, you can also get a good sound by moving the mic right up to the speaker grille, and you'll find that you get a useful tonal variation by moving the mic between the middle of the speaker and its edge. Miking close to the edge gives a softer tone than miking the centre of the cone.
Going to the other extreme, it's equally valid to say that if the amp sounds good in a particular room then moving the mic further away to capture some of the room sound is also OK. This will produce a more spacious sound, but at the same time it's likely to be less upfront than close miking, so the choice of which is best depends on where the guitar part will sit in the track and how that sound relates to the other instruments. The new generation of modelling guitar amps has taken the pain out of guitar recording as you always hear exactly what you're going to get as you record. Nevertheless, there are situations where miking a real amp produces noticeably better results, so don't sacrifice quality for convenience.
Drums are often a big problem for home studios, not just because of the noise they make, but also because many domestic rooms are too small to record them to their best advantage. The close-miked sound of the individual drums won't suffer too much from being in a small room, but the ambience mics (overheads) might, because they are further away and hence receive a greater proportion of room sound. For this reason, it's worthwhile spending some time experimenting to find the best position and distance for the overheads, and you might even consider using boundary-effect mics (PZMs) fixed to the walls or ceilings as an alternative to traditional overheads.
In my experience, though, the most influential element in drum recording is the drum kit and the drummer. A bad kit will always sound bad and a good kit played badly will still sound bad. Only a properly tuned kit, played by a good drummer will sound right. I've been involved in too many sessions where the drummer simply doesn't know how to hit a kit and, no matter how much EQ or compression you apply, the tom fills still sound like a rat scampering across a heap of inflated hot-water bottles!
Having established that your source and your mic positions are as good as you can get them, the next thing to check is gain structure. In the digital world, which is where most of us live these days, that means getting as much level into the A-D converters as possible, while not allowing the peaks to clip. Leaving a little headroom spare as a safety margin is often wise, but the lower you set your recording level, the more resolution you lose as fewer bits are used to represent your signal. That's one reason analogue compression is often applied to a signal prior to the converters — it helps prevent overload but at the same time maximises the audio level feeding the converters.
When using a sequencer that includes plug-in processing, it's usually only necessary to ensure that the channel and master meters are reading sensible values, but if you have outboard effects or processors, you'll also need to check that their input and output levels are set to make the best use of the available dynamic range.
We all know that we're supposed to pay attention to accurate monitoring, but there's also a degree of the 'if I ignore it, it'll go away' syndrome attached to the subject. This is very risky, as a monitoring system that lies to you is certain to compromise the quality of your mixes, especially if (as most project studio operators seem to) you also do your own mastering. However, it's really not that difficult to set up an adequate monitoring system, as most furnished domestic rooms with carpets make passable monitoring environments provided that you work reasonably close to the speakers and don't choose monitors with too great a bass extension. Also be aware that the smaller the room, the less accurate the bass end of your monitoring is likely to be — in these situations, choose smaller monitors that roll off at around 50 to 60Hz and resist the temptation to add EQ to bass sounds to make them sound bigger. More often than not, if you leave the bass end of your EQ set flat, the bass balance will look after itself.
There are numerous two-way speakers that will produce good results in a project-studio environment, so all you need to do is mount them on solid stands in a symmetrical arrangement, three to five feet from your mixing position and at head height. Avoid putting them on the meterbridge of your mixer and also avoid placing them too close to corners, as the proximity of reflective surfaces will compromise the sound, resulting in an unpredictable rise in bass end.
Because you're working in a room that was probably not designed with audio in mind, you may need to add a few bits of acoustic treatment, but in most cases this means an area of acoustic foam (around one square metre is usually OK) or heavy drapes directly either side of your mixing position plus something to break up any flat, reflective surfaces close behind you. Shelves, curtains and equipment racks usually do the job just fine. We've covered monitoring in detail on numerous occasions, so I won't go into more detail here except to reiterate the importance of getting to know how your speakers sound when you play commercial recordings through them. The most important component in any monitoring system is the hearing of the engineer, so think of this acclimatisation process as a means of calibrating the engineer to the monitors and the room.
Electronic sources such as synthesizers may seem easy to record, but there are issues that can compromise the sound quality. In some instances, the instruments themselves may be noisy, especially some of the early digital synths. There's little you can do about this noise at source other than maximise the output level of the instrument (though not setting it so high that you get distortion on loud chords), but if you have access to any kind of analogue noise-reduction box, these can sometimes be very effective. A more realistic scenario for computer users, though, is to record the synth onto a separate audio track and then use a denoising plug-in to clean up the recording before mixing.
The other problem that besets electronic instruments is the old ground-loop hum gremlin, and that's not always easy to fix. However, I usually find that making up a special cable to feed unbalanced instruments into balanced mixer channels usually does the trick. This has been described many times in these pages, but is worth repeating. Essentially you wire the balanced end of the cable normally, using twin cored co-axial, but at the synth end you use a regular mono jack with the 'hot' cable core wired to the tip, the 'cold' core wired to the earth tag of the jack, and the cable screen left disconnected. Don't remove the earth cable from the mains plug of an instrument to cure hum, as this could seriously compromise your safety. Ground loops are less likely to be a problem if everything that can be wired balanced is wired balanced and if mains cables all go to the same distribution board rather than plugging into outlets right around the room.
Virtual instruments are the easiest of all to record, as they don't have cables to go wrong or cause ground-loop hum. However, if the channel into which they are inserted is turned up too high, they can clip, and then they will sound disgusting. Keeping an eye on levels is pretty much all you need to do.
The burning question for many readers is whether working at 24-bit or 96kHz resolution will make their recordings sound better. My own view on this is that if your mixes already sound as good as the very best commercially available CDs, then it might be worth using a higher resolution, but working at 24 bits is far more significant than using a 96kHz sample rate, especially if you're going to have to convert back to 44.1kHz for CD production. Recording at 24 bits enables you to leave a little more 'headroom', while still maintaining more than enough dynamic range to dither down to 16-bit when you come to mix, but even 16-bit audio has a maximum dynamic range of 96dB, and I think you'll find that the ratio of signal to ambient background noise in project studios is typically worse than 65dB. What this means is that if you are working on pop music, which has an inherently limited dynamic range, you're going to be very hard pushed to hear any difference between recordings made at 24 bits and recordings made at 16 bits. Probably more significant is the choice of converter — a cheap 24-bit converter will almost certainly sound rougher and more 'digital' than a good-quality 16-bit converter. You may also find that a digital system comprising multiple components sounds significantly better when clocked from an external word clock generator, as daisy-chaining devices can lead to excessive jitter, resulting in increased noise and distortion.
By all means experiment with higher resolutions, but if your existing recordings are sounding noticeably 'digital', then the chances are either that you have a system with poor-quality converters or that your interconnection system is introducing jitter. It also follows that high-resolution audio only really works when you use a central word clock generator, unless your system is almost entirely computer-based and self-contained.
We've covered mixing before on numerous occasions and we've also discussed the importance of arranging your song so that the various voices and instruments complement each other. If your individual tracks sound great, but they sound a mess when mixed, then examine your arrangement and your choice of sounds. Don't be fooled into thinking that better equipment alone will make a lot of difference.
You'll know if your recording technique is good if you can set up a mix with no EQ and no effects other than a hint of reverb, and it sounds great. This is the ideal starting point for a mix and one you should always aim to achieve, even if you don't always quite get there in practice. After that you can use EQ to fine-tune sounds and you can apply more processing or effects where needed, but if the mix doesn't sound good before you do this, the chances are that either something isn't recorded as well as it should be or your choice of sounds/arrangement could be better.
While adequate recording equipment is necessary for high-quality recording, the main quality differences are brought about by how that equipment is used. Poor wiring and intermittent patchbays can wreck all the good work done by the high-quality devices in your system, as can inappropriate recording techniques or lack of attention to gain structure.
As I said at the outset, it's only by thinking very carefully about your system and the way you use it that you'll be able to identify ways in which to make it sound significantly better. Just throwing money at esoteric preamps or high-sample-rate A-D converters is unlikely to make much of a difference, as most problems are much more fundamental than that. If you don't believe me, consider instant cameras. A professional will always get a better picture using a pocket camera than the inexperienced amateur using a top-of-the-range Leica and all the trimmings. It's what's in front of the camera, how it's lit and how you frame it that really counts. The rest is just window-dressing.
The most important part of any session is getting the best from the performers, so listen to their concerns. For example, bands usually feel happiest playing together, even though this is not ideal if you want all your tracks neat and separate. However, you can balance these needs by recording the main backing in one hit with a scratch lead vocal, which you can later overdub. You may also be able to DI some instruments onto different audio tracks at the same time as miking them up, so that you can play around with the sound afterwards. When you come to overdub, take the time to set up a good headphone mix, and respond if the performer asks for any changes.
The quirks of live performers can be frustrating at times (such as singers who need to be facing away from anyone who can see them), but being able to get great performances is what separates a good producer from a great one.