One year on from the arrival of Tascam's revolutionary 1" 24-track tape recorder sees the release of an alternative version of the machine — designated the MSR24S, it features Dolby S in place of the dbx noise reduction employed on the original model. Although I do not wish to further perpetuate the rather unjustified criticism that the dbx system has received in the past, there can be no doubt that the availability of an alternative noise reduction format will further enhance the sales of this remarkable machine.

Dolby S is a derivative of the state-of-the-art Dolby SR professional system, developed primarily as a domestic system aimed at improving the performance of cassette recorders. Its appearance on this machine will certainly change one or two preconceptions about the system, whilst the resultant audio performance will, I guarantee, not fail to surprise those who are able to audition it at first hand. At the risk of pre-empting my conclusion, the MSR24S really does sound sensationally good!

To recap a little first, the MSR24 was the first 24-track machine not to employ 2" tape. Nothing so revolutionary in itself, for track widths have been narrowing over the years to the point where we had already achieved a very acceptable 16-track performance from 1/2" tape, but 24-track had always seemed to be some sort of milestone — the last bastion of the professional formats. The astonishing thing about the MSR24 was that, for the price of a second-hand, second-rate 2" machine, it offered an excellent, modern, motion-sensing transport, with sophisticated microprocessor control facilities on board and a host of other up-to-date facilities, in a compact machine which not only gave you 24 tracks, but sounded great as well. For the private studio owner, not greatly concerned about compatibility with the outside world, using 1" tape simply meant lower tape costs, and a more compact machine. In the commercial world, the decision is obviously more difficult — but with the mess of competing digital formats having fragmented the consensus more than ever before, the opportunity is surely there for cost-effective narrow-gauge 24-track to establish itself as an alternative, serious professional format. The MSR24S fits the bill perfectly.

CONSTRUCTION

Remarkably, the MSR24S is little larger than its 16-track (MSR16) predecessor, although this has been achieved by employing an external power supply unit (PSU). This is a heavy 4U rack-mounting device, laden with supply monitoring LEDs, which connects to the main chassis via a substantial multicore cable with large, locking 'D' connectors. Removing the PSU from the chassis has also kept the weight of the MSR24 within convenient limits — it is feasible for one person to lift it without undue strain. The main chassis can be rack-mounted, and indeed this seems to be the most popular method of installation, in my experience, employing a free-standing rack-mount trolley, with the PSU mounted below.

The MSR24S is constructed on a main deck-plate, substantial enough to inspire confidence, with a solid-looking head assembly, and the only incongruity perhaps being that plastic is used for the cosmetic head block cover. The 10.5" reels are located on to very positive retainers, using an expanding spring-locking principle which provides a very solid and reliable fixing. The simple, symmetrical transport layout uses a standard pinch wheel/capstan assembly, with roller bearings on all the tape guides, and employs brushless DC reel motors and phase-locked-loop capstan drive. The tape tension arms at either end of the transport path are firmly damped against excessive movement at the ends of their travel, further adding to the impression of a well engineered piece of equipment which has received its due share of attention to detail. The transport gives every sign of having been built to withstand heavy usage, and I am most impressed by its performance.

Tape handling is very good on these machines (in the past, tape handling was guaranteed to set professional machines apart from the semi-domestic formats). The transition from fast wind modes into play, and vice-versa, is beautifully smooth, which will help minimise the level of routine tape degradation inherent in the amount of shuttling that goes on during a heavy overdubbing session.

Below the head block (which has a proper retractable/pop-up head shield) and transport assembly are all the controls and meters, mounted on what at first appears to be a removable panel — virtually the whole of which is actually duplicated on the remote control unit dedicated to this machine. In fact, it doesn't actually dismount from the unit, but merely hinges down from the bottom to give easy access to the line-up presets behind it. This should certainly encourage regular maintenance of the electronic side of the machine. However, regular 'tweakers' will be disappointed by the fact that the machine employs a two-head design, with just an erase head and a combined record/play head. Given that a substantial part of the development costs of this machine must have been invested in the 24-track heads themselves, a three-head version would undoubtedly have cost a great deal more, so I am sure this was unavoidable from an economic point of view. Having only the one active head means that record alignment becomes a two-stage process — one pass in record to adjust, and one in play to view the result. Twenty-four lots of that might get a bit tedious!

Round the back, the user is confronted with an abundance of connectors and interfaces, with all audio interconnection being by means of normal, domestic, phonos. Nobody will particularly welcome the necessity of interfacing via 48 phono connectors, but it is difficult to see what else Tascam could have done given the extremely compact dimensions of the MSR24S chassis. Being phonos, all signals are unbalanced and at the -10dBV Tascam operating level. However, as even the larger and more expensive mixing desks these days seem to make simple, user-configurable provision for operation at -10, this should not prove to be the source of difficulty that it might once have been. Usage with other than 24-group desks has sensibly been allowed for, with back panel switchable input parallelling available, to offer the maximum operational convenience with either 8 or 16 group outputs. Track 24 is omitted from this arrangement entirely, on the assumption that it will be dedicated to sync code, although a simple factory mod or a little DIY paralleling easily overcomes this, if that really doesn't suit your kind of work.

In common with most of Tascam's recent machines, the MSR24S is well prepared for any synchronising need that might come its way. The three control interfaces cover standard SMPTE/EBU, RS232, and a port dedicated to Tascam's own RC424 remote control device. The audio-visual and multi-media worlds seem to have adopted narrow gauge multitrack more readily than any other market, and it is surely a far-sighted move on Tascam's part to have made their present generation of machines so readily interfaceable. Full externally controlled machine sync is available, with an External setting on the varispeed function, Fix and Variable being the other options. A varispeed readout is available by means of the Display switch, which toggles the main counter display between 'real time' tape position, or varispeed percentage (+/-15%). The tape counter seems pretty dependable, even over a prolonged cycling test, indicating both accurate detection and that tape slippage is minimised by the gentle tape handling qualities of the transport.

TRANSPORT FUNCTIONS

Transport functions abound on this machine, with two instantly accessible locate points, in addition to zero. Auto-play is available as an overriding function (it will enter play mode whatever point you locate to), and cycling between the locate points is easily set up. Auto drop-in is available, complete with a Rehearse function to ensure that you have set the In and Out points accurately, and giving you (or the performer) the opportunity of practicing the drop-in part exactly in context.

When you finally make the drop-in, the MSR24S staggers the timing of the turning on of the heads to facilitate a 'gapless' ('glitch-free') result. Normally the erase and record currents are activated simultaneously, causing a section of unerased tape (equal to the distance between the heads) to be over-recorded. Delaying the current to the record head (under microprocessor control) by the appropriate amount for the speed in use avoids the overlap 'glitch' that often restricts the success of the tightest of drop-ins. Apparent drop-in/out 'speed' is further enhanced by the fact that the heads are located fairly close together, so that at 15ips the inevitable drop-out delay is minimised. Of course, you still can't do the impossible and drop in unnoticed in the middle of a note, but anything you would normally expect to be able to do on a quality machine, you can do very well indeed on the Tascam MSR24S — it is very fast and free of audible vices, whether dropped-in manually, automatically, or via the footswitch facility on the rear panel.

There are several useful edit modes on the MSR24S, accessed via the dedicated Edit mode switch. Manual Edit is the normal mode, disabling the reel brakes and retracting the tape lifters so that you can hear the signal as you move the tape manually across the head. Pressing and holding either of the fast wind buttons will do this for you automatically, if you need to move the tape a long way. Since the tape stops as soon as you release the switch, it really is quite controllable. There are manual and automatic dump modes for ditching sections of tape, and a useful Spot Erase facility for very controlled and accurate clean-up jobs. This is facilitated by being able to enter Record whilst in normal Edit mode — if the relevant track is pre-selected to Ready, you can enter and leave Record whilst moving the reels slowly by hand. It works pretty well in practice, although there must surely always be a risk of generating a click on tape by exiting Record when the tape is not moving fast enough. Even though the fast automatic drop-in facility of the MSR24S is likely to be employed for most clean-up tasks, it is nice to know that Spot Erase is there for anything too transient to be removed by conventional methods.

Also worthy of mention are the Load facility, which can be used to define the usable area of the tape and therefore prevent the machine from accidentally winding off in either direction, and the Spool function, which permits winding at a reduced speed to ensure neat tape storage.

All in all, this is a very good functional transport, with more than its share of onboard automation, but which is nevertheless fundamentally good at fulfilling its most important role — handling tape efficiently and kindly.

MONITOR MODES

On the MSR24S, monitor logic is governed by three switches designated All Input, Auto Input, and Insert. 'All Input' is self-explanatory, whilst 'Insert' sets up the normal overdub situation, ie. monitoring follows the recorder status, giving sync output until a track enters record, whereupon it switches over to input. 'Auto Input' is a very welcome addition, formerly very much the province of 'pro' machines only (it also appears on the MSR16S, incidentally).

Auto Input gives input monitoring at all times other than when the tape is playing, and is not in record. Effectively, it means that when recording or overdubbing, and monitoring via the tape return path rather than the mixing desk output, you will hear the microphone signal whenever you are not able to hear the tape signal, thus keeping you in contact with the performer(s) during all wind and stop modes, without the need for constant switching on the part of the operator. Auto Input, like automatic talkback, is one of those functions that is so obvious, and so natural in use, that you only really appreciate it fully when you are subsequently deprived of it!

Tracks are set to Record Ready via switches beneath their appropriate LED column meters. The switches are clearly marked with durable engraved track numbers, and there is a status LED located just above each one which flashes to indicate Ready mode, then illuminates fully when Record is actually entered. Level metering is via 12-segment green LED bargraphs, which turn to red at the nominal 0VU point, and are calibrated to +10dB at the top of the scale. The machine seems to be quite happy with the meters being driven right to the top of their scale, like the meters on many mixing desks these days. I can't say that I am ever terribly comfortable with this — if you are regularly taking an LED meter right to the top, you have no way of knowing if you are actually exceeding the top level!

Given that with the dbx-based MSR24 you are effectively no longer fighting tape noise at all, and should therefore primarily aim to avoid driving the tape into non-linearity, a little more visual conservatism with levels perhaps ought to have been encouraged. However, with Dolby S working primarily on the quieter elements of the signal (rather than on the whole signal, as dbx does), the possibility of tape saturation is altogether less critical, thus still allowing you to use some of the creative effects of 'abusing' an analogue recording, such as deliberate tape compression, without fear of unwanted side-effects.

Noise reduction, whether it be the Dolby S on the MSR24S or the dbx of the original model, can be switched out if desired, in banks of eight tracks or solely on track 24 (to dedicate that track to SMPTE/EBU timecode or other sync codes). The Sync Lock function also conditions track 24 against accidental erasure — a very useful safety feature these days, given that so many applications depend on synchronisation with external equipment of one kind or another.

DOLBY S

Dolby S has been described as the consumer version of Dolby's professional SR noise reduction system, although one could perhaps characterise it with just as much accuracy as a development of Dolby C, with an LF band, plus the addition of one or two of the more sophisticated functions of SR. Indeed, I understand that the current terms of the licensing agreement for Dolby S actually prohibit manufacturers from using the system on equipment deemed to be 'professional'.

The triumph of Dolby SR lay in its unique ability to make optimal use of the characteristics of the human ear, whilst simultaneously embracing the overriding Dolby principle of 'minimum processing' — if the ear is incapable of hearing noise in a particular part of the spectrum, then signal in that area is not processed. Dolby S, although inevitably less sophisticated than its illustrious professional counterpart, inherits this legacy of 'the least treatment is the best treatment' and thereby suffers none of the modulation artifacts or potentially compromised transient performance of simple broad-band compander (compressor/expander) systems.

Dolby S consists essentially of five active elements, dividing into high frequency (HF) and low frequency (LF) stages, with a single fixed band operating in the less critical area below 200Hz, whilst the other four stages all operate from 400Hz upwards. Both fixed and sliding bands are provided, facilitating optimal matching of the system characteristic to the spectral content of the incoming signal. The HF bands are then further subdivided into high and low level stages, with two 12dB compander stages operating with staggered thresholds, giving the relatively high figure of 24dB of noise reduction when both stages are fully activated, but without the need for the unfavourably high compression ratio that would otherwise be required.

The well known Dolby A system relied on dividing the signal into four discrete bands, employing compression and expansion in each one, to achieve between 10dB and 15dB of noise reduction (dependent on frequency). One of the most significant improvements achieved by SR lay in its ability to conform much more closely to the actual signal spectrum than is possible with just a fixed four-band system. Dolby B and C systems both adapt their processing to the HF content of the incoming signal to some extent, so that noise reduction is only carried out on any area of the signal where the high frequency content is insufficient to 'mask' the tape noise. Dolby S takes this principle further, with the two sliding-band HF filters (separate High and Low level circuits, with a threshold 25dB below 'Dolby level'). In addition to the sliding bands, which adapt their action to affect only higher frequencies in the presence of strong signal levels at lower frequencies, fixed bands are also employed. These combine with the sliding bands via a technique called 'action substitution'.

It is an inevitable characteristic of a sliding-band system that the noise reduction effect is reduced at lower frequencies as the band slides up towards the HF region — action at high frequencies will, of course, be maintained. It is similarly inevitable that a fixed band system will display loss of noise reduction at frequencies above the frequency of a loud signal within its band, whilst showing significantly less loss below. By employing both fixed and sliding bands, the boosting of low-level signals (and hence the noise reduction effect) is able to be far more constant. Effectively, action substitution is able to take advantage of the beneficial characteristics of both types, without suffering the disadvantages of either.

Like Dolby SR, a key factor in the subjective performance of Dolby S processing is not just that the noise floor is very low, but that it remains apparently unchanged whether there is signal present or not. Noise reduction in the absence of any signal is easy, of course — a noise gate would suffice — but ensuring there is no perceptible change in the noise level when signal is present is an infinitely harder task, requiring the most sophisticated processing. Dolby S, like SR, has the apparent ability to discriminate between valuable signal and unwanted noise, thereby rendering recording artifacts such as modulation noise effectively inaudible. It also achieves considerable reductions in distortion and improvements in transient response.

SPECTRAL SKEWING

The inevitable non-linearity of analogue tape at the frequency extremes, which progressively worsens with increasing level, is counteracted in the Dolby S system by applying a specific 'anti-saturation' characteristic. This consists of a simple shelving filter network that reduces the recording level of frequencies where the tape cannot handle high levels. A complementary shelving boost is then applied on replay, restoring the response. Furthermore, there are the fixed high and low frequency 'Spectral Skewing' filters first seen on Dolby C, again with complementary fixed replay tailoring, which are active at all levels, effectively modifying the record EQ characteristic to achieve the optimum compromise between frequency response and noise. These filters also reduce any tendency for response errors inherent in the record/replay system to cause mistracking of the encode/decode process. Unlike a wide-band compander system, Dolby S is thus extremely tolerant of any HF response errors in the tape machine, such as HF droop through head wear etc.

Finally, 'Modulation Control' deals with the way the Dolby S system reacts to very high level signals just outside the noise reduction bands. With a sliding-band system, it is feasible for such a signal to cause the adjacent band to slide so far as to create a gap between its noise reduction and the masking action of the signal itself. Similarly, signals just outside a fixed band will inevitably cause some gain reduction within it, for the band-pass filters used to create the band will have a finite slope. Modulation Control (another technique borrowed from SR) tries to reduce this effect, and the subjective total absence of modulation noise would seem to bear witness that it works.

Like all Dolby noise reduction systems, Dolby S employs a parallel signal path configuration, with a passive main circuit and a sidechain circuit in which all the processing takes place. The output of the sidechain circuit is combined with the unprocessed signal during the encoding process and subtracted from it during decoding. As the sidechain signal is subject to compression, while the direct signal is not, its contribution is greatest at low levels and diminishes with rising level until, on the loudest signals, it is effectively insignificant. Thus changes in gain are confined to levels where the tape characteristics can be relied upon to recover an accurate version of the encoded signal — potentially damaging effects, such as transient 'overshoot' (short-term high level signals due to the finite attack time of the compressor stage) which can cause distortion and therefore decoder mistracking, are thereby avoided.

That, in essence, is it — like Dolby SR, the S-type system is elegantly simple in concept yet dauntingly sophisticated in execution, and will undoubtedly extend the useful life of the analogue tape medium.

MSR24S IN USE

So what does it sound like? Well, first of all, the MSR24S is pretty quiet. Although Dolby S may not theoretically provide as much noise reduction as dbx, for all practical purposes it is not actually that far off, particularly in the subjectively more noticeable higher frequency regions. I shall not make this an opportunity to join in the running-down of dbx, for the original MSR24 is still an excellent recorder and I would be quite happy to use either variant, however, the Dolby S version does seem to have an operational advantage in being extremely forgiving.

Over-recording of transients, that would cause audible 'squashing' with dbx, seem to pass without any undesirable effects at all, for the Dolby S noise reduction is doing as little as possible when confronted with a high level signal. Relative under-recording of signals with very little HF content (eg. bass guitar) produces little or no obvious modulation noise. Armed with a battery of vicious snare drum samples, totally unrealistic drum machine hi-hats, and the sort of things that people generally use to try to catch out noise reduction systems, I must say that I was unable to induce anything that could really be considered a 'side-effect' at all. Transient 'blurring' is non-existent, nothing can be made to 'pump', and nothing modulates in any unnatural way whatsoever. What you do get is an output signal that simply sounds as if it extends effortlessly to the limits of the machine's nominal 40Hz to 20kHz response.

A blindfold A/B comparison between the MSR24S and a decent 2" multitrack would certainly be interesting, for the top end is surely every bit as 'open' on the MSR24S. There also seems to be plenty of 'weight' at the bottom end — the quality that I normally find I miss most on narrow-gauge multitracks. Again, without a side-by-side comparison with the dbx version, I cannot trust my memory to say that there is now more, but there is certainly as much as I would expect from a 2" machine at 15ips. Natural acoustic sources, such as piano and steel-string acoustic guitar, are handled with a delicacy and integrity that I am simply not used to hearing from a narrow-gauge analogue format — due to severe restrictions on the amount of time that the review machine was available for evaluation, the acoustic test sources were second-hand, via my own PCM F1 digital recordings, but I don't think this devalued the test a great deal.

The final blindfold listening test was performed with a CD running in parallel with its recording on a couple of MSR24S tracks. Being only a two-head machine, it is not possible to perform simple input/off-tape switching for comparison (although in such a test you can work out which of the two signals is ahead of the other and conclude that it must be the source.) By pre-recording the CD and then running the two side by side, you have no way of guessing the positional relationship and are left to compare nothing but the sound itself. I have heard some remarkable demonstrations of Dolby SR, where listeners not only preferred the analogue SR to a digital source but also consistently believed the SR version to be the digital source! Dolby S seems to have inherited that quality of imparting a highly desirable characteristic to the sound, and whilst I will not say that the CD/off-tape Dolby S sound was identical, I have to admit that both I and my colleagues were consistently wrong, and were reduced to simply guessing!

The future of the MSR24 was probably assured anyway, but it will be even more so now with the release of the MSR24S. I daresay the market may well polarise, with the dbx version remaining dominant in the USA, where dbx is the pre-eminent professional noise reduction format, whilst the Dolby S version takes the machine to new heights in the European market, where they seemingly have no qualms about adopting a 'domestic' system if it does the job well enough, as proved by the success of Dolby C.

The future of Dolby S, if anything, looks even brighter, for its ability to operate without side-effects, whilst still providing a significant degree of noise reduction, will surely see it adopted on other machines. Given that the MSR24S still sounds more than respectable when running at its lower 7.5ips speed — more so than the dbx version, I have to admit — there are perhaps implications as to the feasibility of an even narrower track width. There is surely no reason why we should not at some time in the forseeable future have a semi-pro, 1/2" 24-track with Dolby S. I am speculating wildly, I realise, but I bet it would stand comparison with any of the current generation of 1/2" 16-tracks.

The implications are also considerable for the cassette-based multitrack market, where dbx is, in my view, overstretched by the basic infidelity of the medium, and where Dolby C simply doesn't get rid of enough noise. Once Dolby S can be produced more economically, as a single chip per channel (rather than the three chips that are currently required), this will surely be the breakthrough that the 8-track cassette market has been waiting for.

One of the primary characteristics of Dolby S is that it is downwardly compatible with both Dolby B and C. It is also quite respectable when replayed un-decoded, with some obvious spectral shifting (but still reasonably well balanced because of the LF band), but nothing in the way of excessive modulation that would make it unlistenable. In the domestic market, stereo tapes encoded with Dolby S will be able to be played quite happily on machines with either B or C (but not vice-versa, of course), whilst still enjoying the benefits of a measure of noise reduction. Thinking ahead a little, it follows that the forthcoming Dolby S version of Tascam's MSR16 1/2" 16-track (the MSR16S) should be, to some extent, 'compatible' with the large number of Dolby C-based Fostex B16s, E16s, and G16s out there. Exciting times indeed!

THE LAST WORD

The MSR24 would probably still have done well if it hadn't offered any noise reduction at all — indeed, it sounds quite respectable if you switch it all off, so long as you give the machine plenty of level. With only two tracks routed, noise is pretty insignificant. At the risk of sounding like the official apologist for dbx, the MSR24 seemed to me to be the perfect candidate for the dbx system, with good wide-band performance to start with and a wider track width than the 1/2" 16-track machines. However, the unavoidable fact is that some people have distinct reservations, prejudiced or practical, about the dbx system, and they more than anybody else will surely welcome this Dolby S-equipped alternative. The rest of us will simply have to use our ears, but in this instance that may not be too difficult! Dolby S is seriously impressive, and when combined with such an inherently good machine as the MSR24, makes for an extremely powerful package.

So what reasons are there left for buying a 2" multitrack? Well, compatibility with the thousands of other 2" machines in use around the world, for one, but what else is there? Tape costs are lower with a 1" machine like this, it is physically more compact, and the capital outlay is not in the same league. And the sound? I suggest you judge that for yourself at the very first opportunity. Recording technology has advanced so fast in recent years that the old minimum track width standards for the defined 'professional' formats have surely now been rendered utterly meaningless.

Accepting that the advent of the MSR24S will undoubtedly further add to the plethora of private 24-track studios, and given that the machine is so economically priced, in comparative terms, I am sure we will see (as happened with the E16) many commercial studios acquiring one, if only to be compatible with clients preparing work in their own private studios for completion in a commercial establishment. Given the news, as I write, that the much-rumoured 1" 24-track from Fostex (G24S) that we have been anticipating is also to feature Dolby S, this argument becomes even more compelling.

Evidence would seem to suggest that any piece of equipment that can span the gap between the commercial and personal audio worlds, with a foot comfortably placed in each camp, will do very well indeed. Could we be witnessing the birth of a new standard?

FURTHER INFORMATION

Tascam MSR24S £8,395 inc VAT.

TEAC UK Ltd, (Contact Details).