This year's AES convention in Los Angeles didn't just give equipment manufacturers a chance to show off their latest innovations. It gave further evidence that the distinction between the music and recording industries is becoming almost impossible to draw.

COMPARED TO A MUSICAL instrument exhibition like the NAMM Expo, the Frankfurt Musikmesse or the British Music Fair a recording convention like AES has a number of obvious differences.

To begin with, it's a lot smaller. November's AES show occupied a relatively small area of space on two floors of the Los Angeles Convention Centre, plus a few rooms at the Hilton Hotel a few blocks away.

Second, it's likely to be a great deal quieter. Strict rules governing sound levels on stands were enforced throughout the show, so most demonstrations relied on banks of headphones for interested attendees. The exception to this rule was the upper level and Hilton demo rooms, where a decent level of soundproofing enabled exhibitors to make a bigger audible show of their products' capabilities.

But a recording show like AES also exudes a kind of quiet affluence. Wandering around the exhibits, you get the feeling that a lot of serious business is being done here, an impression that's reinforced by a quick glance at some of the exhibitors' price lists. Despite the increasing sophistication of hi-tech music products, recording equipment still rates as the costliest sector of the music business. And visitors to AES who owned recording studios (or who were making purchasing decisions for owners) seemed to have quite a bit of money at their disposal.

But at this latest AES convention, there were plenty of signs that the above differences are becoming little more than superficial. Because in terms of the kind of innovations that were on show and the kind of people they were aimed at, it was clear that the barriers between the recording and music production sides of this industry are crumbling - and crumbling fast.

As examples of this, one could point to the number (larger than ever) of musical instrument manufacturers attending the AES, not to mention the proliferation of music software companies occupying booths.

Then there's the fact that a surprising number of demonstrations were actually presented by musicians - Frank Zappa (for NED) and Jan Hammer (for Fairlight) were two of the biggest draws. And when you consider that both Zappa and Hammer make their living working from their own home studios in any case, you inevitably have to reconsider whether the distinction between musician and recording engineer is still a valid one to draw. Especially when, as in the case of many of the new machines on display at AES, new technology is being made as accessible to one group as it is to another.

Contemporary, direct-to-disk recording systems are a case in point. On the one hand, a system like NED's "Tapeless Studio" represents just that - a tape recorder without the tape. But on the other, it's an obvious extension of the kind of digital sequencing systems musicians have been using since before the advent of MIDI.

Technology is taking what were once two separate worlds, and drawing them inexorably together.

In fact, direct-to-disk recording was one of the major talking points of this particular AES. Its advantages over tape are that it's quicker and often more convenient to use (rewinding is instant, for example), that it potentially offers editing facilities down to a very fine resolution (per sample on the best systems), and that it makes onboard signal processing in the digital domain a realistic possibility.

Lexicon's Opus was the most sophisticated disk-based recording system on show, and at its most complete (and expensive) level, uses four Winchester hard disk drives (together with two CD ROM units for system information) to store up to eight hours of mono signal, four hours of stereo. Systems like the Opus, however, are likely to be of more interest in the film and video post-production arena than in the music business, where the ability to, say, move a piece of soundtrack so that it's in line with dialogue from another source could prove invaluable.

At the other end of the scale - and of more immediate interest to musicians - are systems like the PPG HDU and Hybrid Arts ADAP.

The PPG offers 12 minutes of recording time shared between four tracks, and a second drive memory expansion in the New Year will double the recording time. Time/pitch correction is one of the HDU's big plus points, as is its price - £12,800 next to something like £200,000 for a basic version of the Opus.

If that is still too much, Hybrid Arts' system cuts costs further by using the Atari ST computer for much of its memory and processing power. This gives a total recording time of 20 seconds, but with a new hard disk buffer/interface, you'll be able to access the sort of recording times offered by more expensive systems via a suitable ST hard disk unit.

Incidentally, Hybrid were also showing the SMPTE version of their MIDItrack Professional sequencer for the ST, which besides being able to read and write SMPTE, now has the ability to request and store sound data (of both the synthesised and sampled variety) and send this out to connected machines at the beginning of sequence replay. The great advantage of this is that all your sound and sequence data can be stored on one disk, and you don't need to worry which sounds are loaded into your keyboards and expanders.

Yet despite the flurry of activity surrounding disk-based recording systems, there is every indication that this particular branch of technology will continue to be of minority interest for as long as good ol' magnetic tape remains a great deal cheaper, much easier to edit on, and generally friendlier to use.

In the world of tape, controversy still rages over the relative merits of analogue and digital recording systems. Some tape machine manufacturers - like Otari and Studer - hedged their bets by introducing both analogue and digital recorders at AES. Otari's MX8O is now available in 32-track format for the first time, while the company's new DTR900 (also a 32-track, but using one-inch tape instead of two-inch) uses the PD format for digital recording also used by Mitsubishi and AEG.

Studer have a digital two-track mastering machine in the D820X, but remain committed to analogue recording at the multitrack stage, with a brand-new machine, the A820, making its debut at AES. With the help of Dolby's latest SR (for Spectral Recording) noise-reduction system, the A820 rivals the best digital systems for sound quality, and could lead to more than a few studio managers changing their minds about which direction to take their recording equipment in. Or, as Studer themselves would put it, "the future is not what it used to be".

Sony, one of the pioneers of digital audio, are not surprisingly sticking by their DASH system (which rivals PD; isn't competition wonderful?), and unveiled a number of DASH-compatible products at AES. The most talked-about part of Sony's exhibit, though, was a range of prototype R-DAT units "designed for professional recording applications". R-DAT, if you didn't already know, is the now almost universally accepted standard for domestic digital audio tape recording. The format uses 16-bit resolution and potentially boasts higher sound quality than Compact Disc, while the tape itself comes in the form of small cassettes not much bigger than credit cards.

But R-DAT is controversial, and for a number of reasons, the prime two being that it could (a) lead to wholesale pirating of CDs by domestic users, and (b) ultimately replace CD altogether. This is unlikely to please the record companies, who are doing very well out of CD, thank you, and who are already paranoid about the amount of damage done to their industry by music lovers taping records.

This controversy prompted some very tight-lipped spokesmen on the Sony stand to say not a great deal about when R-DAT recorders will start to be made generally available, and even Sony's press pack was positively brief on the subject.

At the root of all this, of course, is the fact that the hardware and software sides of the music industry (ie. the people who make the equipment and the people who buy and use it) are as far apart as they've ever been, and steadfastly refuse to acknowledge that they are interdependent.

If the record companies, for example, aren't convinced that digital recording makes people buy more records (and there's certainly little evidence to suggest that it does), then they can justify sticking with analogue studios whose rates aren't anything like as expensive as those of the digital ones. The relative technical merits of the two systems take second place to considerations of the bank balance.

Conversely, if the record-buying public wants something as badly as it's likely to want R-DAT (just think, digital recordings of baby's first words), then there's not much the record companies will be able to do about it - especially with half the electronics factories in Japan all set up and ready to go into production of R-DAT machines. And you thought the musical instrument industry was political...

Speaking of musical instruments, quite a number of the major Japanese manufacturers were represented at AES, in the (quite realistic) hope of selling their machinery to demo studios and programming suites, the like of which are springing up all over the US, and beyond.

Korg were showing their SG1 electronic piano, DSS1 sampling synth (complete with a new range of sample disks), and DDD1 drum machine, which will soon benefit from the first shipment of sampling cards. Korg's outboard processors - including the new budget DRV1000 digital reverb - also created interest.

Roland were showing their new S10 and S50 sampling keyboards, both of which we reviewed in last month's MT. It's interesting to note that the S50 now has individual output assignment and a more complete manual - two things we missed on our preliminary review model.

Akai had their X7000 sampling keyboard in its finished form, complete with the ASK7O expansion unit, but none of the clever digital patchbay machinery which graced the company's stand at the summer NAMM show in Chicago.

The Yamaha DX7 continues to inspire more third-party companies to introduce products than any other modern musical instrument. Two such American companies, Symphony and Key Clique, were showing together at AES. The latest sounds from Key Clique feature keyboard splits (up to 12 are possible), which are achieved by some clever jiggery-pokery that tricks the DX's operating system into de-assigning some operators from certain areas of the keyboard.

The results of the co-operation between Key Clique and Symphony have just come to fruition, with the release of the VX7, a programmer and memory expander for FM synths. It stores 512 FM voices (expandable to 1024), with cataloguing and utilities functions. As a programmer, it allows a more direct approach that includes the drawing of envelope shapes, the introduction of 'FM EQ', and the global editing of DX parameters.

Over at the Hilton, Yamaha themselves had several different demonstrations running. One of these was the Yamaha MIDI seminar, which concentrated on the use of MIDI processors (MCS2, MEP4 and MJC8) and controller keyboards (KX76 and KX88). After that came an excellent demo featuring the new QX5 sequencer, which highlighted the machine's ability to run on several different time signatures - 4/4, 5/4, 6/4, 6/8, you name it.

In the audio department, there was an innocent-looking eight-into-two mixing desk which would have been easy to walk straight by if there hadn't been a large crowd of people around it at all times. Innocent-looking it may have been, but this was a programmable, all-digital machine with fader automation and three onboard signal processors.

Turns out it's called the DMP7, and the way it works is this. Eight 16-bit A-to-D converters (with a 44.1kHz sample rate) lie at the audio inputs, while two similar D-to-A converters convert the signal back to analogue at the output end for the stereo outputs. The internal signal processors are the equivalent of three SPX90s, but with their bandwidth expanded by the 44.1 kHz sample rate. Servo-controlled faders are used to show not only the level of each channel, but also the aux send levels to the signal processors.

Each complete setup of the DMP7 can be stored in a memory, exactly like the programs on a synthesiser. There are 32 internal memories, but these can be expanded with 67 on a RAM cartridge to make 99 on-line.

The DMP7 should retail in the UK for less than £4000, and the idea of a 16-bit, all-digital automated mixer for that kind of money is going to appeal to many. Because of its eight-into-two format, the DMP7 is unlikely to be used in multitrack recording environments (though it is possible to chain two DMP7s together using MIDI System Exclusive data). However, seeing as program changes on the mixer can be controlled in real time from a MIDI sequencer such as Yamaha's QX1 (the demo used this feature impressively), there should be plenty of keyboard players for whom the ability to control their own sound on stage via MIDI will prove irresistible.

You could, of course, point out that in the interests of bringing the DMP7's technology down to a decent price level, Yamaha have had to implement the kind of digital parameter access found on modern synths - and there aren't going to be too many sound engineers who will relish the prospect of selecting parametric EQ and digital delay settings via that method, no matter how informative the LCD is.

But there's no getting round the fact that the DMP7 points the way ahead for digital signal control in general, and budget mixing consoles in particular. Like many of the innovations on show at AES, it's a taste of fascinating developments to come, as well as being an incredibly useful tool in its own right.