Razor blade editing has, until now, always been an extremely difficult feature to have on a digital tape machine.

It is usually necessary to use special equipment, such as Sony's 1610/1100 digital editor, to edit digital material. Although this provides very useful options such as preview editing and variable cross-fade lengths, it is also expensive and has its own limitations.

The importance of editing varies from person to person. Many engineers are quite content with the options offered by automated consoles but others want to keep their options open by using editing of the analog master tape as an integral part of the mixing process. This makes it possible to swap verses, choruses, insert a sequence of reversed or odd bits of sound — providing effects which can only be done by editing the master two-track.

The X-80 is Mitsubishi's first two-track PCM digital mixdown machine and it offers both the quality of digital recording and the benefits of razor-blade editing.

It is available either in its own console, or as a two unit portable system — the tape transport unit and the PCM processing section.

The tape transport is of a fairly conventional design. On its own it is of comparable size to a portable Studer B67. Tape transport controls are quite standard — record, stop, play, rewind and fast forward controls are positioned to the right of the red LED tape counter. On the left are controls for power on/off, slow wind, code-record, edit and cue. A single, green, push button for record ready/safe is located above the headblock.

On the meter bridge are control adjustments for record level, playback level and headphone socket and output level control. In addition to the two VU meters are two peak-level LED indicators which display the peak level of the digital data, irrespective of the setting of the playback-attenuator control (unlike the VUs which are post playback-attenuator).

The monitor switching is also located on the meter panel and monitoring can be selected from four sources — input, ee (electronics to electronics), tape or cue.

The input signal can either be taped as soon as it enters the X-80 (input), or after it has passed through all the input circuitry prior to reaching the tape transport.

The third switch selects off-tape monitoring (giving a delay of a little over one second for playback to occur), while the fourth switch is for monitoring the playback from the analog cue-track. This is essential for locating edit points. It should only ever be used for this purpose, however, as its reproduction quality is very poor, especially in comparison to the outstanding sound quality provided by the PCM playback.

The PCM section of the X-80 is the same size as the tape transport section. It has no front panel control, although all the X-80's power supply, remote interfaces, audio, digital and code connections are located on the rear of the PCM unit.

The unit uses 16-bit linear circuitry with a sampling frequency of 50.4kHz. With any fixed head (as opposed to video) format the error correction circuitry must be of a very high standard to cope with the reduction in data stream. This is especially true for editing the tape.

Mitsubishi have overcome the problems of razor-blade editing digital material by cleverly manipulating the PCM's powerful error detection and correction circuitry.

Splicing a stream of PCM data causes an enormous amount of corruption which creates numerous problems for the PCM processor. This usually results in anything from a few seconds of pure glitching to several seconds of muting.

Mitsubishi has diffused this problem by using a cross-fading effect between the two ends of the spliced tape.

When editing the ¼" tape on the X-80, it is necessary to leave a ½mm gap. (The cuts must be vertical, not diagonal as on analog tape.)

This gap is detected by the auxiliary track and the error correction circuitry then cross-fades the two points into one another. This results in extremely smooth edits — and very forgiving ones as well. In particular, level mis-matches are less noticeable on the X-80's edits than on an analog format due to this cross-fading effect.

There are 10 recorded tracks on the X-80. Eight of these are PCM data tracks. The remaining two are used for the recording of the analog cue track and the address data to be used in electronic editing of tapes.

Six of the eight PCM tracks are used to record data for the left and right channels. The remaining two contain a powerful Reed-Solomon (R-S) error correcting code.

The R-S code takes various combinations of four of the six PCM tracks, calculates their logical sum (parity) and records the logical sums of these combinations in one of the error correcting tracks. The other track is then used to record the logical sum of the logical product of the data tracks and a fixed set of logical constants.

The R-S codes have 400 times the correcting power of simple parity codes and can successfully correct up to two error bits occurring simultaneously.

Digital tape has to be of a much finer gauge than analog tape because it is essential that tape-to-head contact remains constant over the designated curved area of the heads.

This is one of the few areas where I felt uncomfortable with the ¼" digital format as the tape must be handled with more care than I would normally take with an analog ¼" or ½" two-track. The instruction manual suggests that a pair of thin gloves should be worn when editing. I did not find it necessary to take such precautions and neither do any of the other engineers I have spoken to who work extensively on the X-80.

To test the robustness of the X-80's format and error correction I recorded some material from compact disk onto the X-80 and then damaged different sections of the tape to varying degrees.

First of all I wiped a finger along the recorded surface of the tape. This had next to no detrimental effect on the quality of the playback.

Then I tried stretching the tape slightly. This sometimes resulted in muting of the signal for perhaps half a second depending on the extent of the damage. A slight amount of stretching, however, can be handled quite well. The result is an occasional reduction of definition, but more often an undetectable, glitch-free correction.

When the tape is 'scrunched-up', the error-correction performs miracles. A reasonable amount of scrunched tape proved to be no problem, (though the damage that might be done by an over-inquisitive child finding out what flavour the tape is, did produce the occasional glitch).

Sony's PCM-F1 or PCM-701 used in conjunction with a SL-F1 or C9 Betamax recorder is the digital format familiar to most engineers and producers. The F1 format is used extensively for many different purposes — from the storing of sampled sounds to providing safety back-ups of mixes.

Although the X-80 is more expensive than the F1, it is still substantially cheaper than Sony's 1610 or 1630 with a U-matic recorder. I therefore compared the X-80 with the PCM-F1/SLF1 as the reference point most people would be familiar with.

I used all-digital compact disc recordings of classical and rock music, as well as a mix from multi track and some test tones with which to A/B test the X-80 against the F1.

On its own, the playback quality of recordings made on the X-80 were indistinguishable from the input material. This was found to be true for both loud, high energy rock music as well as for the quieter, more discerning passages found in Mahler's seventh symphony. The stereo imaging remained faithful to the original and the recording level hardly seemed to make any difference at all on background noise.

When A/B tested against the F1, there was a slight difference. The X-80 seemed to have more bandwidth than the F1 with a rounder and superior bass end.

Workhouse Studios has had an X-80 for eight months. Pete Hammond, the producer/engineer, had similar praise for the quality of the X-80. His main criticism was that the digital tape needed to be so thin and fragile.

Another negative point, as I mentioned before, is the quality of the auxiliary cue track. Many artists and producers would be given a fright if they heard their mix monitored from here. It is also so compressed that it is sometimes difficult to hear an edit point — such as a bass drum in amongst a thick wall-of-sound production.

I would also like to have seen some provision for autolocate — at least a 'return-to-zero' such as on the Studer A80.

The layout of the tape transport controls is curious. Both the stop, rewind and fast-forward controls are to the right of the play button, which was confusing to start with. There is also no tape dump (motor off) facility which is a sad omission for a machine whose primary selling point is its editing ability.

The negative points are outweighed, however, by the overall excellence of the machine. It has most of the requirements for the 'ideal' mixdown machine.

It is compact and easily transportable. The recording quality and error-correction are outstanding and its edits are kinder than those of an analog tape format. A single 10" reel of tape (although a bit expensive) will hold one hour of recording (enough for most albums). It never requires any lining-up and it is almost a quarter of the price of the editing system for the 1610 or 1630 with the 1100 digital editor and two U-matic recorders.

The X-80 has already found favour with the highly critical engineers for classical music. They have been particularly enthusiastic about easily editing different takes of the movements of a symphony on a digital format.

Mitsubishi aims to launch the X-86 digital two track in June. It will have many similarities with the X-80 but it will be Pro-Digi format and therefore compatible with the Otari and Telefunken machines. It will also have two extra tracks — a second auxiliary cue track as well as an auxiliary data track (for CD sub-codes) — built-in varispeed and return-to-zero autolocate, RS323 and 422 ports, variable cross-fade times and individual control over recording of the left and right channels.

Mitsubishi Distributors. Contact: Hilton Sound, (Contact Details).