The Limitations of Two-Track

The beauty of multitrack recording is the degree of control it allows you in being able to build up a musical arrangement part by part. This means having individual record/play access to each track of a recording. A normal hi fi cassette recorder may have two tracks, but it is generally only possible to record or playback from both tracks simultaneously; you can't record on track one and then subsequently play it back whilst simultaneously recording on track two. The ability to do so is at the heart of the whole process.

There are a good number of stereo reel-to-reel machines that do allow individual track record/playback access but these aren't considered multitrack. It is possible to use them in a multitrack fashion in that you might record a mono piano part on track one and then, having got that right, record a vocal on track two whilst listening to the playback of the piano on track one. This is obviously very limited in that it can only produce mono recordings and only allows you two shots at building up an arrangement.

Most machines of this type will also offer 'sound-on-sound' facilities. This greatly expands the possibilities and is a far more common approach for home recording. In this case you would record your piano on track one as before, and then using the tape recorder's simple built-in mixer, you would mix the playback of track one with the output of your vocal mike, and record the mix onto track two. If, during this process, your voice goes haywire, you still have the piano part intact on track one and thus have no problem in re-doing the vocal until you get it right.

You may now feel moved to wheel in your DZ-106 synthesizer and lay down a keyboard part or two. This is easily achieved by repeating the above process, this time mixing the output of track two (piano and vocal) with the output of your synth and recording the combination onto track one. Notice that at this point you lose the original piano track, and so if you subsequently decide that you could have done it better or it should have been louder with respect to the vocal, there's nothing you can do about it. You will also notice that at this stage the piano has been re-recorded twice, each time losing a little quality and adding a little noise to the sound. This method is known as 'sound-on-sound' recording and the process of mixing parts back and forth between tracks is commonly referred to as 'bouncing'. You will find that with each successive bounce the original parts become increasingly muffled and distorted and so you are very limited in the number of parts you can add. It's also very difficult to anticipate how loud you want each part to be before you've recorded everything else, and even if you do have a pretty good idea of levels when you record them, the bouncing process itself always plays havoc with the intended balance because of the progressive sonic degradation of the parts.

Portastudio v Four-Track Reel-to-Reel

Therein lie the limitations of two-track.

The next logical step after two-track is four-track. Four-track reel-to-reel was a well established home recording format well before the development of the portastudio. There can be no doubt that, simply by virtue of its wider tape, its faster tape speed and its superior transport, a well maintained ¼" machine will produce higher quality recordings than the cassette-based system. For home recording demo applications, however, it is still probably an illogical choice. Apart from the fact that a decent system will cost you considerably more to put together, when involved in building up relatively complex musical arrangements track by track, it is generally of greater importance to have more tracks at your disposal than to gain a couple more dB less noise or marginally less distortion. For a relatively small amount more than you'll have to pay for a good quality four-track you can buy a Fostex eight-track.

Quite obviously it all depends on how much money you have but also on how much of that money you want to allocate to recording hardware. After all a 64-track digital recording complex is no good without musical instruments, and it may serve your purposes better to concentrate resources on mikes, guitars, synths, drum machines, bits and pieces of percussion and a few basic signal processors/effects units rather than on a fabulous recording system and still be stuck for good quality sound sources.

This logic works for most people involved in the production of Pop/Rock music, because four-track will be exclusively a demo format. If, however, the vast proportion of your instrumentation is synth/sequencer based, it could be that four tracks are enough to turn out masters. In such a situation the instrumental arrangement would be 'recorded' into a multitrack sequencer of some kind (probably MIDI) and one of the four tracks would be used to record sync code to enable the sequencer, and thus all the synths and drum machines etc, to be synchronised with the tape. This would leave three tracks to record vocals and any acoustic instruments before mixing down to stereo from a combination of the outputs of the sequenced instruments and the three tape recorded tracks. This method is increasingly common and produces excellent results because the bulk of the music doesn't see tape, and thus isn't degraded by the recording process, until the final mixdown. For most professional applications, though, eight tracks tend to be the minimum even for this method.

The main advantages of the portastudio over the ¼" reel-to-reel format are:

1 Less expensive
2 More compact
3 Simpler to use
4 More convenient lower cost tape format
5 It may be able to double as a stereo hi fi machine and thus you could exchange ideas with friends who have domestic machines.

Portastudio v Domestic Cassette Machine

Though it does stray a little from the original Philips cassette track format (Philips invented the standard hi fi cassette that we all know and abuse) the Teac track layout is effectively compatible with it in as much as the four tracks in each case are similarly positioned across the width of the tape.

The most obvious difference between the formats is that in the case of the normal hifi cassette there are two pairs of tracks recorded in different directions whilst the portastudio uses all four tracks in the same direction. The original portastudios ran exclusively at a speed of 3¾ips which is twice that of the standard hi fi machine. The faster speed yields better recordings but makes it impossible to have any interchange between the formats. Some later designs run at 1⅞ips and some are switchable for both speeds. This means that they can be used as part of a normal hi fi system using tracks 1 and 3, ignoring 2 and 4; some machines even have a switch to configure them for such uses.

There may be a problem here with noise reduction systems in that Dolby B is still the most commonly used NR for domestic machines whilst many four-tracks use Dolby C, dbx or their own in-house systems. Even if the NR matches, it has to be said that the tracks won't align absolutely exactly and thus reproduction won't theoretically be quite as good. However, the results aren't at all bad and will probably be satisfactory for most people's purposes. You can easily decide yourself by taking a pre-recorded cassette with you when you go trying out machines. Whilst it may save money to be able to double up functions like this, remember that if you don't have a separate cassette or ¼" stereo machine you will have to borrow one for mixdown. But even if you never want to use it as a hifi machine it may be very significant for you in that it allows writing partners or other members of the band to give you (possibly even by post) basic recordings of ideas made on their domestic machines that you can then develop further on your four-track. The lower speed does mean slightly lower quality, but in many cases the advantages as discussed have the greater weight.

Modular v All In One

The original Teac Portastudio Model 144 featured a four-channel mixer and four-track cassette deck in a single compact unit. This is still a popular configuration but there are also a number of manufacturers who have brought out modular systems comprising stand-alone cassette decks with matching mixers and patchbays plus racks in which to combine everything into a single system. The main reasonings behind such systems are that if you already have a reasonable mixer for live PA work, why should you be forced to waste money on another one; and also that modularity introduces greater flexibility in that you are then able to use each component separately, eg you can use your mixer for PA whilst back at home your flatmate uses the deck as part of the hi fi.

The extra metalwork, transportation costs for the manufacturer, duplication of power supplies and possibly metering etc plus the need for leads, means that you have to spend more money on a complete modular system (especially if you're going in for a rack to put it all in) to get the same level of facilities. Also, my experience is that plugging and replugging gear is a pain in the neck and that part of the beauty of borne recording is being able to go straight to it when the inspiration strikes. If you're using your 12:2 PA mixer, you will not only have to continually replug during the session to allow it to work with a four-track tape deck, but it will also take up a relatively large amount of room. Even with the purpose-built desk the modular system will also almost certainly be more bulky than the all-in-one design and the need for interconnecting leads introduces another source of noise and system breakdown.

Then there's portability: if you're going round to your mate's house to use his acoustic piano or whatever, it's a simple matter to pick up a single box, stick a mike in your pocket and stroll off down to the bus stop or jump into your car. A mixer, deck and accompanying leads is generally a bit more of a hassle. If recording is an important and constant part of what you do my opinion is that you'll probably be better off with an all-in-one design. If you do have a sophisticated PA console with superior eq and auxiliary sends etc, an all-in-one system will still have the necessary input and output sockets to allow its use when convenient or important, but you won't be forced to heave it out every time you want to record something. It is also a fact that most PA consoles will work at a higher line level (signal voltage) than your portastudio and so matching may be a bit of a problem.

As the market stands at the moment, I tend to come down in favour of the all-in-one design for most applications, although special circumstances may point you in the direction of a modular system. Here are some basic points to bear in mind:

Pro All-In-One Design:
Generally less expensive for same facilities/quality; more compact; greater convenience and portability; no connecting leads; no interfacing problems.

Pro Modular Design:
No duplication of your existing facilities (PA mixer); greater flexibility.

Noise Reduction

If you play a tape with nothing recorded on it you will find that there is a residual 'noise floor' of hiss. The 'signal to noise' figure (S/N) quoted for a machine refers to how loud that noise is with reference to programme (music) recorded on it. A hissy recording will have a poor S/N. The main job of a noise reduction system is to improve the system's S/N, although it can also reduce print-through and distortion figures too.

The fundamental requirement for producing a good S/N is to get as high a signal level on tape as possible. The residual tape noise will always remain constant and so the higher the signal level recorded, the greater the S/N. If you were recording a signal with an ever constant level, this would be a simple matter of turning up the record gain so that it was just below tape overload (and thus distortion). Problems arise from the fact that music gets louder and softer and thus what is a good strong record level for a lightly played acoustic guitar becomes far too high when the whole band come in with drums, brass, keyboards and electric guitars.

Conversely, setting the record level to suit the loudest dynamic point will leave the acoustic guitar horribly under-recorded, and thus the S/N will suffer accordingly. Even with individual performances levels can change dramatically, even if it's just the pauses between notes where any hiss will become very evident. If you were fast enough and had a good enough memory it would be possible for you to keep your hand on the record level control and turn it up when the playing was quiet and down when it got loud. Then on replay you could do the exact opposite with the playback level control in order to restore the original dynamics of the performance.

The first half of this process is called compression because it compresses the dynamic range of the programme so that the difference between the loudest and softest notes is reduced. The second half is called expansion simply because it's the opposite to compression. A compressor, as used all the time in the recording process, is concerned exclusively with the first part of the process in that it is intended to reduce the dynamic range of a performance in order to keep it present and iron out inconsistencies in performance technique. The NR system has to expand the compressed material from tape because it should ideally be transparent in that it should make no audible difference to the nature of the recorded sound except in as much as it won't be accompanied by so much residual tape noise. The combination of compression and expansion is commonly termed compansion and all noise reduction systems work with this principle in some more or less sophisticated way. One of the main undesirable side effects experienced with simple compressors is pumping: any electronic audio system will have its own noise floor, similar in nature to the residual tape noise, though electronically generated. As the level control is turned up so the noise is increased, and thus if a lot of compression is being used (ie the amount of gain reduction is large) a pumping effect will be heard as the level goes up and down with the strength of the programme. If the compression/expansion characteristics aren't well matched for some reason going to and from tape (and this may be the fault of the tape recorder and not the NR system itself), the same pumping effect maybe heard; and remember, the level detector can't tell the difference between low level programme and no programme. Thus when mismatches occur the noise level tends to be exaggerated in gaps between playing.

The NR systems most commonly found on portastudios are Dolby B, Dolby C and dbx.

The dbx found on portastudios is, in principle, the same as that used in some major multitrack recording studios (mostly in America), although a difference in the quality of components used in more up and down market units seems to have a noticeable effect on the transparency of the NR system. It applies a straight 2:1 compression to everything and anything recorded and an exact mirror 2:1 expansion on playback. It's very simple and justly claims a very high 30dB of noise reduction.

Until very recently I was rather against dbx on portastudios simply because experience had shown me that the system was very prone to pumping and breathing with transient sounds such as drums or spikey synths. The first notable exception to this was the Tascam Porta One, costing around £421 (minus discounts) and featuring dbx which works extraordinarily well. One explanation for this unprecedented level of performance from dbx at this price is that both the accuracy of the tape system and the quality of the dbx components have been improved for this relatively recent model. For whatever reason, it's certainly a very good machine and seemed to mark the start of a new lease of life for dbx with a number of more recent machines using it to great effect. The rule, then, is simply listen before you buy.

Dolby B is to be found on some older less expensive models and as it is more or less standard on domestic machines you will probably be quite familiar with it. It doesn't tend to travel well between machines in that programme recorded on one machine and played back on another will often suffer. Even when working properly it is not as effective as newer dbx or Dolby C. Dolby C can be seen as two Dolby B systems in tandem, but the important thing is that in practice it provides very good noise reduction levels and is very tolerant of transient sounds such as drums etc, at its best producing almost no pumping effects.

The Mixing Console

Whether housed in a single box or in separate modular form for the purposes of discussion any portastudio system can be divided into two basic component parts: the mixer and the tape deck.

Mixing consoles are known by many names including mixer, desk, console, board and practically any combination thereof, although there is no actual difference implied in any specific term. In the context of sound recording the mixer acts as a combining, processing and distribution centre for audio signals. Hence they are all about inputs and outputs: an input is a point at which a signal can enter a piece of equipment and an output is a point at which a signal can leave a piece of equipment. Signals come out of outputs and go in to inputs. Outputs plug into inputs. This may seem childishly simple to you but there are many people who get very confused over these basic issues. There are generally two types of inputs on a mixer: Microphone and line.

As the names suggest microphone inputs are designed to accept the low level outputs produced by mikes, whilst line inputs are intended for the stronger signals received from line level outputs. The term line level used to refer to a very strict set of electrical characteristics as laid down by the BBC, but today electronic circuits are far more tolerant of each other and the term is applied to a number of different standards that can generally be persuaded to talk to each other simply by plugging them in and adjusting their levels with the knobs and sliders provided.

A degree of caution is still advised, however. Four-track cassette systems will generally work at a nominal line level of -10dB or occasionally at -20dB or -10dB/-20dB switchable, whereas professional and PA equipment will be more likely to operate at the higher levels of 0dB or +4dB. The term 'nominal' means that the figure is given as an optimum average level to aim at, although actual programme will obviously fluctuate around that figure as it gets louder and softer. On a -10dB mixer's VU meters a reading of 0VU will be produced by a -10dB constant tone. The amount over that figure you can go without causing distortion is referred to as the mixer's 'headroom'. Better mixers will have greater headroom and will therefore be less prone to distortion in the event of sudden peaks in programme.

The tape deck will be set up so that a voltage of -10dB will produce an optimum record level on tape, and even if the mixer has plenty of headroom and can stand substantially more than that, the tape will start to 'saturate' and distort if it is driven much beyond the stated figure. Thus it's important to have a mixer whose line level matches that of your tape deck, or else you must be aware that the mixer's meters will be either over- or under-reading. Using a mismatched mixer/deck will also result in generally poorer S/N and distortion and for convenience isn't advised.

Equalisation is simply another term for tone controls. The eq section is a very important part of a mixer and can obviously make a big difference to the degree to which you can control the sounds being recorded. There are generally only two types of eq to be found on portastudios: Two-band fixed and two-band sweepable.

The first is to be found on older and less expensive models and is basically bass and treble offering so many dB of boost or cut at two fixed frequencies such as 100Hz and 8kHz. The second type also has to gain controls to boost and cut but each of these bands also has another control to determine the centre frequency of the band, eg they might be 'sweepable' between 60Hz and 1.2kHz for the LF band and 700Hz and 10kHz for the HF band. The second arrangement is vastly preferable, providing a much higher degree of control.

Auxiliary sends can also make a large difference to the flexibility of your recording system. Each auxiliary send allows a sub-mix to be made of whatever's going through the input channels. This submix is then generally used for one of two things: foldback (headphone monitoring for whoever's playing) or as an effects send, to eg a reverb or echo device. Foldback sends tend to be best taken from a point in the circuit just before the main channel fader (pre-fade) so that you can mess around with record/playback levels with out altering the monitoring balance. Effects sends, on the other hand, are generally taken post fade so that as you fade an instrument out, the effect fades with it, rather than staying on like a ghost image.

During recording, foldback mixes are likely to be necessary, whilst during remix, only effects sends will be required. Thus there will sometimes be a pre/post-fade switch on the auxiliary send(s) to allow them to be used for either purpose.

Certain of the more expensive portastudios will have balanced mike inputs which, under some circumstances, offer a notably superior performance as compared to unbalanced inputs. When an audio signal is sent down a piece of wire it is prone to picking up what is termed airborne interference. This can take the shape of hum, radio breakthrough and all sorts of spurious, unwanted noise. To keep this to a minimum, cable used for both microphone and line level signals is screened with a wire wrap or braid. This braiding is similar to that found on the coaxial cable used for TV aerials. For short lengths, at line level and under normal circumstances, as long as it is connected to earth at one end, it should do its job effectively. However, because of a microphone's lower signal level, such interference is bound to be comparatively higher and more obtrusive, especially where longer cable lengths are used (over 5m). Most line level inputs and, on less sophisticated equipment, mic level inputs will use an unbalanced line.

If you can get an indication as to whether a line is unbalanced by looking at the cable connections: if there is a single core with a screen, it is an unbalanced line. The standard alternative is to use a balanced line, and this is indicated by the fact that the cable will have two cores plus a screen, and all three will have separate connections at both ends. Sometimes two-core cable is used for unbalanced lines, but this is fairly unusual outside of professional installations. The way in which a balanced line is connected allows it a far greater immunity from noise and interference, and at mic level, it is undoubtedly an advantage. An unbalanced line can be connected to a balanced line as long as the connectors at each end are wired correctly, but it will then operate as an unbalanced line.

Metering is another area worth consideration. The actual characteristic of the meter, ie whether it's called VU or peak reading, is a bit academic at this level because most of the meters you come across will fall somewhere between the two. The main choice will be between moving coil meters (normal looking ones) and LED bargraph meters. For me the latter arrangement is better because it's easier to see from a distance and at a glance roughly what's going on. If you do go for moving coil (and some very good machines do feature them) check that they're illuminated — reading unlit VUs in low light is very tiring overlong periods.

So those are all the component parts that go to make up portastudios great and small. The fact that these machines are more or less complete self-contained studios means that they are not too difficult to put through their paces, and so the final and possibly most important suggestion is that once you've narrowed it down to a few likely models by wading through the next few pages, you do a tour of your local dealers and actually try recording a few bits and pieces just to see how you like the eq, and the noise reduction etc, etc. Here are the machines that you'll be choosing from...