A new software package for the Commodore 64 that puts step-time sequencing, recording and scorewriting in an elegant perspex box. Ian Waugh finds it transparently appealing.
Instant music transcription takes another step forward with the introduction of the JMS Scorewriter package for Commodore 64. If you like the company's sequencing software, you'll love this.
Every musician, composer and arranger has a dream. Musicians traditionally dream about fame and fortune, birds and booze. Composers dream about conducting the LPO (the PLO if you're suicidal, 'Phil and his Harmonica' if you're a Falkland Islander) at a performance of their latest work. Arrangers dream about all sorts of things, but not very often - they're too busy chasing deadlines. The same goes for journalists, too, but that's another story.
But what they all have in common is a desire to communicate using musical ideas. 'Wouldn't it be great', you hear them say, 'if everything we play could be turned into printed music automatically as we play/compose/arrange it?' A generalisation, I know, but you know the sort of thing I mean.
Computer software is theoretically an excellent medium in which to create such a scorewriting system, but thus far, music transcription packages for home computers haven't been exactly thick on the ground. The main reason for this, I suspect, is that it's incredibly difficult to write such a program, far trickier than most people would ever think. These computers might be awfully clever, but they can only do what you tell them to do; yes, you've heard it before and you're hearing it again. As far as music transcription goes, the hard bit is lining up, spacing the notes and joining the stems together so that the score looks like a proper piece of music, not the hybrid offspring of a Stockhausen score and a Rorschach test. (Well, those and a few dozen other niceties.)
But here we are at the beginning of 1986 (or very nearly), and a few scorewriting programs are starting to appear for computers that regular musicians, composers and arrangers (though probably not journalists) can afford. From Worcestershire comes the Joreth music system, Commodore-based but with a music notation section that enjoys only afterthought status, while two months back I looked at a more comprehensive (but step-time only) Stateside package for the same computer, currently being marketed by Passport Designs. Now the Germans at Jellinghaus Music Systems have come up with something similarly thorough, this time based around the real-time recording principle. It's called the Scorewriter, and it comes as part of a hardware interface card that also houses two other JMS programs, the 12-track Recording Studio and the Sequence Chain.
Basically, the Scorewriter takes files recorded with the Recording Studio and turns them into printed music. You can't use the Writer without the Studio, so it's just as well that JMS have decided to incorporate all three MIDI programs into one package, all of them burned into EPROMs. The main advantage of the EPROM (as opposed to disk, cassette, cartridge and the rest) is instant access to all the programs at switch-on - for Commodore owners, this means no more waiting for the wretched 1541 diskdrive to wake up. The disadvantage lies only in a higher price, as in this case, you pay for the plug-in board and its additional chips as well.
The EPROMs are neatly mounted on a board and covered with perspex. This plugs into the cartridge port and the JMS MIDI interface, in turn, plugs into the back of this. It reminds me a little of ancient US sci-fi horror shows like The Outer Limits which, in their more daring episodes, contained close-up shots of brains floating around in glass tanks. More practically, the exposed ends of the casing aren't going to do anything to protect the circuitry inside, so you could find yourself with a load of rusty chips after someone's spilt coffee over your gear. But at least this way you can see the JMS module has no moving parts.
Before looking at the Scorewriter in detail, let's take a quick look at the other two programs. The Recording Studio was reviewed in E&MM December '84; see the Back Issues page if you missed it. Only two things to add to that review: a photocopied insert testifies that a non-MIDI drum machine (using the 24ppqn sync code format) can be plugged into the system through the Control In socket, thus alleviating many a syncing frustration. And strange though it may seem, only 6813 events are free when you enter the program — the disk-based software gave 7677 events. Odd, but I don't suppose it'll make any difference at all to the majority of users.
The Sequence Chain program is used to link sequences (well, what did you expect?) that have been recorded on the Recording Studio program. You can build up quite complex pieces in this way, using a program that works a little like the Linker module in The Music System (reviewed E&MM October '85). The program allows you to transpose the sequences, alter key velocity and tempo. You can effect voice changes, too, which is rather nice. However, if you already have the Recording Studio, you can purchase Sequence Chain as a separate program; it costs around £45. If you buy the Scorewriter, however, it comes as part of the deal.
But back to the Scorewriter. The software writers have obviously realised that producing a good score is no easy task. It's not simply a question of pushing a few buttons and watching your score roll off the presses — there are far too many variables for that.
What the program does do is to set a list of default values which can successfully print out some of your pieces. We'll look at the options in the order you're likely to encounter them.
The first screen presents the main menu from which you can: Transform a Sequence, Display/Print a Score, Display Directory, Issue Disk Command and View/Change Presets; all options are selected with the C64's function keys. At the top of the menu page you're prompted for the Sequence name and Score name, the Sequence being a file you have already saved to disk with the Recording Studio program. The Score name defaults to the same name as the Sequence, but you can alter it if you wish. The transformation of a sequence takes place without further trouble, the Score file being written to disk. Although the files have the same name, they're saved with different suffixes so you won't overwrite a file.
After a sequence has been transformed into a score, the second option scrolls the score up the screen and, if the printer option is set to on, prints it out. Before the system actually gets around to printing the score, the drive whirs away for a few minutes while the computer does its sums. After the whirring, the screen clears and the score scrolls. If it's printing out as well, a page of music would take approximately (actually very approximately, as it depends on the number of staves per page etc) 10 minutes to print. At least, that's how long it took in the Waugh household, using an Epson printer with a Commodore-emulator interface, which I assume was printing fairly quickly.
There are three secondary menu screens: Control Parameters, Quantization Parameters and Layout Parameters. Once set, all three pages of variables can be saved as a file, which means you can quickly load the settings you use most for your printer and the type of music you are printing out.
The Control Parameters are fairly straightforward. You can select which bars are to be printed and whether output is to the screen, printer, or both. The program supports the MPS801 and Epson FX80 printers, and you can alter the device number and secondary address to tailor the output to suit you and your printer. A little more information about this in the manual would not have gone amiss; all you get with the present booklet is a signpost to your printer's handbook (there aren't any secondary addresses in the Epson manual). More serious is the fact that, in order to view the results of new settings, you have to wait about four minutes. If output is sent to the screen only, the score appears quicker, but headings are not printed.
Plodding on, you can set the key signature of your piece and tell the program whether you want the bar and the page numbering on or off. There's also a mysterious option unhelpfully labelled 'Special Symbol F. Time Sign', which defaults to a '-' value and can be changed to a 'C'. Set to 'C', the program inserts the Common Time Signature instead of 4/4 or 2/4. I wonder what titles JMS rejected before plumping for 'Special Symbol F. Time Sign'...
The Quantization Parameters screen is divided into two sections. The left side sets Tolerances and helps correct any timing inaccuracies you may have made during recording (who, me? never!). First item here is SIM (for Simultaneous Tolerance) which corrects the way chords have been played. You know how it is. Just occasionally you don't manage to hit all the notes in a chord at exactly the same time, so the thing sounds a bit ragged; SIM helps pull the notes together, and its resolution can be varied from 0 to 9 MIDI clock pulses.
TIM (for Timing Tolerance) ranges from 1 to 47 MIDI clock pulses, and is similar to the quantisation settings of the Recording Studio in that it pulls notes backwards or forwards onto the nearest beat.
OVR (for Overlap Tolerance) ranges from 0 to 9 MIDI clock pulses. This determines to how great an extent notes which have been played legato will overlap in the printed score. With a setting of 0 there is no tolerance, and any overlapped notes are written out.
REL is the Release Time, ranging again from 0 to 9. This is used to extend the length of a note whose 'on' time (ie. the time between your pressing a key and releasing it) is shorter than the actual sound produced, something that can easily happen when you're using a synth sound that has a long-ish release time.
Lastly on the left side, the Split option selects the pitch at which a track is divided when assigned to a two-part score. Split can be switched off, of course, but when it's on, it divides a track into two parts. If you try to be clever and set the split too high or low, the program, being clever in return, takes to using octave signs.
The right-hand side of the Quantization Parameters screen determines which stave parts 1 and 2 of the track will be written on. As many as six staves can be used, and any track part can be assigned to any stave. This is done under the SY (for System) column, while ST (for Stems) sets the turnaround point for note stems.
The final screen controls the Layout Parameters. Here you can give your magnum opus a title and a subtitle, and control both note-spacing and the size of the left and right margins. You can also set the number of staves you want printed on each page.
The lower half of this screen determines how those staves will be printed. They can be in two lines (as in a piano part) or in orchestral format with a continuous line down the left-hand side. And quite excellent they look, too.
As an added bonus, there's a transposition function called TRS, which is incredibly useful for scoring transposing instruments, or for just showing friends how clever you are by playing a piece in C and printing it out in D#.
Each stave can be assigned one of three clefs - treble, alto or bass - and the spacing between staves can be adjusted so you can insert lyrics, music instructions or both. You can also adjust the amount of music that appears above and below the middle line of a stave - this affects the printing of ledger lines.
All in all, making proper use of these controls requires a little thought and more than a little care. But these are damn clever, these computer programs. This one was sharp enough to spot one of my feeble attempts to get it to foul up by printing two staves on top of each other; it gently altered my instruction and printed a copy as near to my request as possible — without fouling up.
Now, all this may seem like a lot of bother just to get a printout. To a certain extent that's true: it can take a while to sort out the myriad options that appear on the screen in front of you. But so long as you remember that they're all there to help you organise the printout the way you want it, you should be OK. Unfortunately, loading a file and printing it out using the default parameters isn't a guarantee of decent results: you need to spend a little time setting some values first.
Mind you, working your way round the Scorewriter's parameters isn't made much easier by the manual, which, on the review sample, was 24 photocopied sheets of A4 paper. It didn't say 'Temporary Manual' anywhere, but there were spaces for non-existent diagrams, and I'd like to think JMS could come up with something a bit more substantial in exchange for your £340. A little more information would help the novice no end, and why not work through a few examples to how adjusting each parameter affects a score? Actually, compared with some, the Scorewriter manual isn't too bad (though the Recording Studio 'leaflet' still deserves to have perforations put into it) but manuals are every bit as important as the product they describe. I rest my case.
As for the printout itself, the examples speak for themselves (assuming E&MM's printers haven't messed things up again). The notes are aligned perfectly, and I just love the way the stems are joined together. Clefs and staves look terribly professional and the program prints legato slurs and ties. Dynamics, if you want to see them, have to be added manually, but that's no great hardship. The Scorewriter will handle up to 20 voices and approximately 50 bar lines can be shown on each page printout of the score.
As a package, the Recording Studio, Sequence Chain and Scorewriter programs should arouse serious attention from programmers, arrangers and composers, so long as they don't mind playing their music in real-time; step-time writing involves going into the Recording Studio's editing section and manipulating MIDI note values,-which is No Fun.
I doubt the Scorewriter is everyone's dream come true, but it's a piece of software that makes music transcription less of a nightmare.
More useful than a brain in a glass tank. Much more.
Gear in this article:
Review by Ian Waugh
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!