With so many low‑cost recording products coming onto the market every month, you might be tempted to ask what, if anything, is to be gained by buying more expensive models. Paul White attempts to answer.
When recording was in its infancy, everything was expensive — there was no home‑recording market, so there was no economy of scale. Now a significant percentage of musicians have their own home studio setups and a whole industry has sprung up supplying the equipment needs of these people — and, because the worldwide market is large, equipment can be sold at near‑consumer prices. But problems sometimes arise when semi‑pro musicians buy very low‑cost equipment, then expect it to produce professional results in a hard‑working environment. Sometimes you're lucky, and an inexpensive piece of gear will behave the way you want it to, but as a rule you need to buy above the low‑budget bracket if you want something really worthwhile. The purpose of this article is to look at some of the more common categories of budget equipment and to examine in what way they are compromised compared with their professional counterparts.
The mic preamps found in most mixing consoles are pretty competent, but sending a signal to a recorder via a console means passing the signal through a lot of circuitry, not just the mic amp. A cleaner way of working is to use a separate, higher‑quality mic amp so that you can pipe your signal directly to your recorder.
Cheap mic amps often have a limited audio bandwidth and introduce distortion; and although they may have impressive noise figures quoted at maximum gain, in the middle of the gain range — where you're likely to be working most of the time — the noise figure might actually be quite poor.
Esoteric mic amps tend to offer bandwidths in excess of 40kHz so as to avoid phase artifacts, have an acceptable noise performance over the entire gain range, and introduce very little distortion. This isn't just down to good micamp circuit design, though: the power supply needs to be especially clean, and a lot of care has to be paid to the design of the phantom power circuitry. The difference between an average mic amp and a really good one is a revelation. The best analogy is to say that it's like comparing a dirty window with an open window — you get a much clearer and more natural sound where detail and spatial information really shine through.
Preamps involving valves are also popular, the idea being to give non‑valve mics something of a valve‑mic character. However, not all valve circuits are created equal, and many impart a very obvious thickness to the sound rather than adding subtle low‑end warmth and gently bringing detail to the front. Some circuits use the valves at very low voltage, and — with the exception of Aphex, who've patented a system to make a low‑voltage valve behave like one being run at high voltage — these low‑voltage stages have a different tonal character to traditional high‑voltage valve circuits. Valve circuitry can be noisy, too, so check the spec and then confirm it with your ears.
Finally, circuits incorporating high‑quality mic transformers tend to perform well, but good mic transformers are expensive. Many engineers claim that transformers, like valves, add a benign character to the sound, but beware of cheap transformers, as these can actually make things much worse.
A microphone is a simple device — you put sound in one end and an equivalent electrical current comes out of the other. So why are there literally thousands of models to choose from, ranging in price from a few tens of pounds up to a few thousand? And, more importantly, what is the practical difference between a £50 mic and a £1000 mic?
It would be nice to be able to say that the more you pay for a mic, the quieter it is, the more accurate it is, and the more sensitive it is — but it's not quite that simple. For dynamic models, the specifications can be very confusing, because on paper many models seem to be very similarly specified. Even many cheaper models tend to be reasonably well engineered, but the subjective difference in sound can vary enormously from model to model.
Most dynamic vocal mics have a presence peak in their frequency response to help them project, and because different designers use boost of different amounts and at different frequencies, some tonal difference is inevitable. In some ways, the user's job is to pick a mic that suits their voice, but what sets a good mic apart from a bad one is harder to pin down.
I've often used the analogy that a microphone is like the lens of a camera — it defines the quality of what comes into the system, and any imperfections in the mic are carried right through the recording process. A poor mic may have a wide bandwidth, a reasonable‑looking frequency plot and adequate sensitivity, yet still sound bad. This may be due to narrow peaks and dips in the frequency response that have been conveniently averaged out by the measuring process used to create the frequency plot, or there may be undesirable phase anomalies caused by the porting of the capsule, or even by the design of the basket surrounding the capsule.
The best way to describe it is to say that poor dynamic mics often suffer from a honky or nasal quality; or, in the other direction, they may lack detail and projection. If you try to use such a mic, you'll find that you're always fiddling with the EQ trying to fix the problem, but this rarely works. Microphone problems are too complex to fix properly with equalisation and, to make matters worse, if you can't afford a decent mic, you probably can't afford a decent equaliser either!
Part of the problem of choosing a mic is that unless you hear two mics side by side, you quite often can't pinpoint what's wrong. If you don't know exactly what model you want, insist on doing a side‑by‑side comparison with a known reputable model, such as the ubiquitous Shure SM58. Even mics chosen for non‑vocal uses should be capable of producing a fairly natural vocal sound.
Other things to watch out for with budget mics are excessive handling noise, lack of sensitivity, and proneness to popping. Popping and handling noise are less of a problem in the studio, where the mic can be stand‑mounted and a pop shield placed between the mic and the singer, but a low sensitivity can mean excessive noise if you're not working with close or loud sources.
Capacitor mics and back‑electret mics tend to be somewhat more expensive than dynamic models, but their extra sensitivity and extended high‑end response make them the preferred choice for the serious recording of vocals and acoustic instruments. Expect to pay high prices for industry‑standard models or any mic with valve circuitry, especially sought‑after vintage models. As with dynamic mics, you should look for a natural, smooth character that sounds both warm and detailed. Some capacitor mics sound detailed and accurate, but somehow manage to lack warmth and weight, resulting in a cold, clinical sound, and again this isn't always evident from the frequency response plot.
There are several current valve mics that offer excellent value — you've probably read the reviews in SOS ‑‑ and, just occasionally, there's a budget bargain when a low‑ to mid‑price capacitor or back‑electret mic happens to perform really well. It's even possible to buy Russian‑made mics that come very close in sound to their Western counterparts, but the trade‑off here is generally poor build quality and slightly more background noise.
What you lose out on by paying less is mechanical robustness and convenience.
Large‑diaphragm capacitor mics are generally chosen specifically for their sound character rather than for their accuracy, but they should still sound comfortable and natural — the whole idea of a mic with character is that it enhances the natural aspects of sound: it shouldn't introduce coloration that's perceived as being unnatural. When testing a mic, listen for signs of raggedness at sibilant frequencies, and try to weigh up whether the sound feels comfortable. A comfortable sound is usually a natural sound; though it may be coloured in some way, the colouring will be benign rather than aggressive.
The noise performance of capacitor mics varies greatly from model to model, and how important this is depends on how you intend to use the mic. If you're working with close vocals, noise is rarely, if ever, a problem, but if you're working with acoustic ensembles, choirs, or particularly quiet instruments, then you may run into difficulties. A number of extremely accurate and quiet microphones are available from top manufacturers for discerning jobs such as orchestral recording, but — as you might imagine — these tend to be fairly expensive.
As a rule, the most expensive mics are the ones that have the best combination of sensitivity, noise performance, wide bandwidth, and smooth frequency responses; if colorations are present, they must be both subtle and flattering. But if you only need some of these attributes, you may be able to save quite a lot of money by choosing carefully. For example, you may be able to find a great‑sounding mic that is a little insensitive or slightly noisy, but for close studio vocals this is something you can live with. However, always buy the best mic you can afford — once the signal has left the mic, the sound can only get worse, not better!
The whole compressor market is a bit of a minefield, because a lot of users buy compressors for their side effects rather than just their ability to control gain. It is possible to build compressors that control gain in a fairly unobtrusive way (this is something that American company Aphex are keen on pursuing), and in situations where you just want to stop a vocal level from wandering around in a mix, unobtrusive compression is exactly what you need. However, if you want to add character at the same time, you need a compressor that also adds coloration, both dynamically and possibly in terms of harmonic distortion too. Several vintage compressors are cited as having classic sounds, and this is often due to the use of photocell gain‑control elements, FET gain‑control circuits or our old friend the valve. Modern designs have adopted all these technologies alongside the more technically advanced VCA.
Whichever type you choose, a good compressor should be able to control the gain of a signal without dulling the transient detail, and without making the sound seem squashed and lifeless. If the compressor adds coloration, the sound should still remain natural and transparent. You'll find that an up‑market compressor will do all these things effortlessly, whereas you might have to spend a lot more time fine‑tuning the controls of a budget model, and even then the result won't have as much life and presence as it should.
Equaliser designs can sound hugely different, and I've never found the EQ on any low‑ to mid‑price recording console to be completely satisfactory. The same is true of virtually all budget graphic equalisers and low‑cost outboard EQ boxes. Adjust a cheap EQ and the sound either becomes dull, boomy or nasal, depending where you apply the cut or boost. But it shouldn't — EQ should behave as a volume control that works over only a specified part of the audio spectrum; when you boost the EQ, all the frequencies in that part of the spectrum should be lifted. Sadly, cheap EQs can also introduce noise, distortion and undesirable phase shifts, which is why the sound takes on unnatural characteristics, often described as nasal, boxy, scratchy and so on.
A high‑quality equaliser will facilitate precise control over the areas of the spectrum being treated, and the circuitry is likely to have a very wide bandwidth, lots of headroom and very low distortion so as to maintain integrity when high levels of cut or boost are used. The result is quite different to what you'd hear when using a cheap equaliser — if you add treble with a well‑designed EQ unit, the sound gets brighter as you'd expect, but it should still sound natural rather than peaky or forced. Conversely, if you turn up the bass end, you end up with more low‑frequency energy, but the overall character doesn't change to become boomy or out of control. And why should it, when you're only turning up gain in one part of the spectrum, after all?
Valve equalisers command high prices, as do most valve‑based circuits, but there are plenty of engineers who feel that the gentle enhancements provided by a well‑implemented valve design are worth the extra cost. To test an equaliser, use a well‑recorded voice or acoustic instrument as the sound source, and confirm that you can make quite radical tonal changes without spoiling the natural character of the sound. What you should find is that budget EQs are OK when used to add just a decibel or two of adjustment, but a high‑quality EQ can be used to make much bigger changes without your having to worry about undesirable side effects.
Reverb units are very interesting boxes, because their purpose is to fool us into believing that what we're hearing is occurring in a real three‑dimensional space. How convincingly this works has little to do with frequency response, distortion and noise, and everything to do with how cleverly the reverb algorithm is devised. Furthermore, the more sophisticated the reverb algorithm, the more powerful a digital processor is needed to run it, so the more expensive the box. Of course it's important to have low noise and distortion too, but what makes the reverb sound either real or not is the algorithm used to create the effect.
Most budget reverbs sound acceptable when creating medium‑ to long‑decay hall and plate sounds, though compare these with a top‑end unit and you'll probably be surprised at how much more convincing the better units are. It's also true that with the really good reverb units, you can add an enormous amount of reverb to the original signal without its seeming to become swamped or overwhelmed — with a cheap reverb, on the other hand, the sound usually deteriorates into a reverberant cacophony.
Some expensive studio monitors sound significantly worse than well‑chosen budget hi‑fi speakers.
There are many factors that make a good reverb algorithm, including the details of the early reflection patterns, the density of the ensuing reverberation, and the shape of the decay. It's also important that the algorithms don't 'ring' (aside from plate settings, which are supposed to sound a little metallic), but you'll find that some of the cheaper units have metallic overtones, no matter what type of room you choose. This kind of problem is particularly evident on percussive sounds — vocals are rather more forgiving of ringing.
While long reverb times might be impressive, the true test of a reverb unit is to see whether it can recreate convincing small‑room and ambient‑space environments. Most budget units become hard‑sounding and metallic at short decay times, whereas a really good unit such as the top‑end ones from Lexicon or TC Electronic will conjure up a convincing mental vision of the type of room being emulated. This is particularly important in contemporary pop music, as long reverb times seem to have given way to cleaner, more ambient treatments.
Loudspeakers are fascinating things, and what's more fascinating is that some expensive studio monitors sound significantly worse than well‑chosen budget hi‑fi speakers. The problem is that a studio monitor has to try to do several incompatible things at the same time, so compromises have to be made; how well those compromises work out is inevitably linked to cost. Ideally, a speaker will cover the entire audio range, will go incredibly loud without distortion, will have a perfectly flat phase and frequency response, and has to be affordable — and pigs will apply for landing clearance at Heathrow!
In most cases, designing a loudspeaker system to reproduce very high sound levels compromises the other parameters needed to keep the distortion within acceptable limits, and the problem is compounded if you want high levels of low‑frequency energy, because then you need larger drivers, bigger boxes and more amplification. And, with cost being a factor, mass‑produced lightweight drive units tend not to perform as well as precision‑built devices with more substantial magnetic systems.
Cheap EQs can introduce noise, distortion and undesirable phase shifts.
Fortunately, most project studio monitors are either near‑ or midfield types, so an immensely deep bass response isn't necessary. Indeed, unless you have a large acoustically designed studio, high levels of bass end simply result in a misleading, very inaccurate sound.
Given the same design budget, a 2‑way speaker is likely to work better than a 3‑way model, which is why most home studio speakers are 2‑way systems. Three‑way systems can offer a wider frequency range, but the cost of three drivers and a more complex crossover network is not insignificant. What's more, most of the sonic problems associated with monitor systems are focused around the crossover points, and in a three‑way system there are two crossover points, as opposed to the single crossover point of a two‑way system. Because the responses of the drivers on either side of a crossover point overlap to a significant degree, it is vitally important that the dispersion and phase characteristics of the two drivers are in step.
The subjective weaknesses of a poorly designed speaker system manifest themselves as a ragged, fatiguing quality around the upper mid‑range, and possibly an uneven or inaccurate low‑end response — if the cabinet has been tuned for quantity rather than quality.
Active speaker systems (those with separate amplifiers for the different drivers) can be built to sound much better then their passive equivalents, because the designer has far more precise control over exactly what signal is fed to each driver. Furthermore, because there are no crossover components between the amplifiers and drivers, the effective damping factor of the amplifier is much higher — in other words, the amplifiers are better able to prevent the speakers from overshooting due to mechanical inertia. Active systems are also better able to send more power to the least efficient speakers, so the designer does not have to attenuate the input to the most efficient speakers by means of resistive crossover components.
So what should you listen for? First you need some familiar test CDs, ideally with high‑quality vocals and some acoustic instruments. Many CDs are not as nicely recorded as you might imagine, so take care to select well‑recorded material that hasn't been processed to death. At sensible listening levels, sit between the speakers in the usual nearfield listening position, close your eyes, and listen carefully — not to the music itself, but to the sound, especially of the voice. Firstly, does the overall tonality seem okay: not too toppy and not too bassy? At the bass end, notes should be even in level, and the sound should be smooth, not boxy or tubby. What's more, the sound should seem controlled — it shouldn't disintegrate into a low‑frequency jumble.
In the vital mid‑range, you should be able to pick out individual instruments clearly, there should be no fatiguing roughness, and vocals should seem to hang in the air between the two speakers rather than stubbornly refusing to be placed. At the high end, there should be precision and detail, but without any annoying sizzle. After listening to the speakers at a comfortable monitoring volume for 10 minutes or so, you shouldn't be feeling an urge to turn them down — if you do, there's a likelihood that high‑frequency distortion is irritating your hearing system.
Finally, try to listen to the stereo imaging. Do you hear sound coming from two speakers, or do you get a sense of each instrument and sound having its own position in space? Can you hear front‑to‑back perspective in the mix, or does everything seem flat? Once you're satisfied with tests, you're well on the way to picking a good set of monitors — but don't skimp on amplifier power, or you'll not get the best from them.
As a rule, you get what you pay for, and in this era of digital recording and highly‑specified budget mixing desks, any weak links in the audio chain soon show up.
As a rule, cheap monitors will either be inaccurate, incapable of playing very loud or both; they're also more likely to incorporate cheap drivers that won't stand up to studio abuse. In the middle price bracket, there are some great‑sounding monitors and some less great‑sounding ones. The trade‑off here is that the best speakers may be less electrically efficient than their less accurate counterparts, requiring more amplifier power — but this isn't always the case, by any means. A lot depends on how much money the manufacturer has spent on a fancy box and how much is left for the drivers and crossovers. You may also find that the best monitors don't go as loud as the less good ones — or at least they may not seem as loud. That's because distorted sound always sounds louder than clean sound, and some perfectly good monitors have been condemned for being too quiet simply because they don't hurt your ears!
Spending a little more gets you into the realms of active monitors, and here you should notice a much tighter sound. Again, the sound can vary a lot, not least because of fashions in different parts of the world, but if you do all the above tests and pick a speaker that feels comfortable to work with, you won't go too far wrong.
As a rule, you get what you pay for, and in this era of digital recording and highly‑specified budget mixing desks, any weak links in the audio chain soon show up. Having said that, there are occasional gems that cost far less than you might think, and of course there are premium‑priced turkeys out there too. After re‑reading what I've just written — and don't expect me to make a habit of it — I'd sum up by saying that, once you've set aside reliability, after‑sales support, and all the obvious technical stuff such as noise and distortion, what you're really paying for is a natural sound that's either completely uncoloured, or subtly coloured in such a way as to make it sound that little bit larger then life. In the case of monitors, I'd always advise going for the most accurate pair you can get your hands on, but when it comes to microphones or outboard gear, there's a strong case for buying both the most accurate, and whichever produces the most convincing illusion of naturalness. Whatever you decide, it's usually cheapest in the long run to buy the best you can afford, rather than buying cheap and having to upgrade every few months.
Most project studios using digital multitrack tape opt for either the Alesis ADAT or Tascam DA88 (or DA38) machines and, in all honesty, both these machines offer superb value insomuch as their sound quality rivals that of far more expensive 16‑bit digital tape recorders. What you lose out on by paying less, other than the expectation of instant response servicing, is mechanical robustness and convenience. Whereas a professional digital multitrack may use purpose‑built hardware and even a specially‑designed tape format, semi‑pro machines rely on inexpensive, easy‑to‑obtain tape cassettes based on commercial video formats. Occasionally, you will get a dropout or a jammed tape — and because the error monitoring on semi‑pro machines is so rudimentary, you don't really get any clear indication that the machine needs cleaning or servicing. A semi‑pro machine is likely to be less rugged than a professional machine, so regular servicing is essential for a relatively trouble‑free life.
On the convenience front, modular digital multitrack machines take a certain amount of time to lock up when two or more are used within the same system, and if you're trying to patch up a vocal track that requires extensive punching in and out, this can be very frustrating. A practical way around this is to make a rough mix from each multitrack onto one track of the tape being used to record the vocals, then work with just this 'slave reel' until the vocals are finished. When all the punching in and out is done, you can then load up the original tapes, mute or re‑use the guide tracks on the vocal tape and mix as normal. Admittedly, this is a bit of a chore, but it's pretty inconsequential compared with the vast cost‑saving over pro gear. What's more, if you have a commercial system based on three modular digital multitrack machines, it's not unreasonable to keep a properly‑serviced spare.
Hard Disk Recorders
I may be wrong, but the way I see it is that semi‑pro hardware‑based tapeless recorders tend to suffer from being brought to market before all the bugs have been ironed out of them. At one time this was a 'feature' only of software‑based products, but now it seems to afflict anything that involves digital circuitry; EPROM updates are becoming a way of life. When it comes to sonic quality, low‑cost tapeless recorders are limited mainly by the accuracy of the converters and the implementation of the circuitry in which those converters are used. Even so, low‑cost converters have got a whole lot better over the past couple of years, so unless you go to more than 16 bits, the difference in sound quality between a pro system and a semi‑pro system might be less than you think.
Computer‑based hard disk recording systems suffer from being — well, computer‑based systems! The problem with computers is that, in a private environment, people expect more of them than simply running MIDI and audio software. They put on screen savers, games, spreadsheets, an internet connection and so on, and then wonder why everything starts to behave in a somewhat flaky manner.
The other major problem is traceable back to software that isn't debugged properly — but instead of fixing what already exists, commercial pressures force manufacturers to add more features, and hence more bugs, every time a revision is made. Add to this the non‑standard nature of computers, especially so‑called PC clones, and it's no wonder that problems arise. The most professional thing you can do is buy a complete system ready configured, and then leave it alone! It might cost more, but if there are problems you only have one supplier to deal with, not a group of buck‑passers.
The sound quality of computer‑based systems is determined not only by the A/D and D/A converters, but where those converters are. If they're on a board inside the computer, the noise performance is likely to be significantly substandard by professional terms; if they're in an external box, or if you communicate with your soundcard digitally, then the sound quality can be excellent. Because the systems are in the main relatively cheap, the manufacturer is hardly likely to jump to attention every time you call up with a problem, though some are better than others in this respect.
Service And Support
One factor that most people fail to appreciate until they need it is that of service and support. If you've just spent £30,000 on a professional digital multitrack recorder and it goes down in the middle of a session, it's not unreasonable to expect the manufacturer to send a service engineer to fix it the same day. However, if you've bought a semi‑pro digital multitrack for about half the price of the VAT you'd expect to pay on the professional machine, you can't expect the same treatment, even if your career does hinge on it. More often than not, you'll have to send your machine away and wait days or even weeks for it to be fixed — and it does little good to protest that you're in the middle of a life‑or‑death session, because the bottom line is that the machine you've bought doesn't make enough profit for the manufacturer or dealer to be able to offer that kind of support. If you're working with semi‑pro gear and you can't afford to have down‑time, then budget for a spare machine, or try to arrange for a fast‑response service contact. There are also a number of excellent freelance studio service engineers who will fix things on site, but relatively few of these have the specialised equipment needed to fully align and test digital multitrack machines. These comments on service apply to virtually all budget equipment, so next time you beat a retailer down in price to the point that they're making virtually no profit at all, don't be surprised if they don't treat your service problems with as much urgency as you do.