Notes on my NoteWorthy Composer files

Stylistic decisions

The files on this site were primarily created to sound right, rather than to look good when printed. (All else being equal, though, I've tried to at least make it look readable on the screen.)

Using NWC's grace notes, for example, steals time from the following note. If I want to steal from the previous note instead, or just want more control over exactly how long it lasts, then I need to use a normal note of the correct length, in the correct place. Particularly in files I've entered more recently, I've tended to do just this.

Many of the files were entered with older versions of NWC, and so don't take advantage of useful features in the newer versions. In the very oldest ones, for example, a tempo specification couldn't have arbitrary text associated with it, and so I either put in a separate text item, or else omitted the text altogether.

Similarly, many of the files predate version 1.75, and so cannot combine a (short) rest with a (longer) note on the same staff. As a workaround, I've either replaced the long note with two shorter notes tied together (the first being the same length as the rest, and the second being part of a chord), or else have moved the note onto a different staff. This looks ugly, but sounds right. The newest files, though, are able to do this properly in many cases.

Another concession to older versions is right at the end of some pieces, where I've put a breath mark. This is because the final note was cutting off too early. Newer versions automatically put a delay there, for this reason, and so I've not done this in newer files.

The filenames, where possible, include the number from the score, as well as something more descriptive (usually the first line). Sometimes the split is in a different place to the score. I've tried to make sure the filenames still sort alphabetically into the correct order. In cases where the numbering restarts in the second act, act 2 has "II_" prefixed. Unless specified otherwise, all these files are taken from the vocal score, and include piano acompaniment. "O_" in the filename indicates an orchestral arrangement.


Where I've made a substantial change from the score I was working from (e.g. to correct a mistake, or to add missing parts), I've sometimes marked the changes in the "Highlight 2" colour (green by default). I only started doing this fairly recently, so older files won't have this.

This doesn't apply to less substantial changes. In particular, while I've tried to be guided by the dynamics in the score, I've not always followed them rigidly; sometimes a part just seems to sound overpoweringly loud, or inaudibly quiet, relative to the others, and so needs adjustment. And sometimes the dynamics aren't marked (or aren't marked clearly enough – e.g. a crescendo with no target volume mentioned).

Where a repeated section has exactly the same rhythm each time, I've usually kept it with the repeat barlines. But where the words fit differently each time, I've usually written it out separately for each. Occasionally, where there's only a very small difference, particularly if it's near the end of the repeated section, I've used special endings to provide different notes each time.

Where the score specifies a tempo, I've usually followed the suggestion. Elsewhere, I've used my own judgement. This judgement may be a little flawed, as repeated listening while transcribing often makes it sound too slow, even when it isn't. So, particularly in newer files, you may want to adjust the speed.

If a part splits anywhere, I've usually tried to split the whole lot into two lines (e.g. having separate 1st and 2nd bass parts throughout, even if they only differ in a few places.) Similarly, I've usually assigned a separate staff to each solo part, even when the printed score shares them. When the chorus is singing, I've only duplicated these parts into the principal lines when the score explicitly tells me to do so.


I've assumed the General MIDI instrument assignments. And I've assumed that MIDI channel 10 is for the standard silly noises (aka percussion). If you've not got things configured in this way, then you'll need to do some reassignment of parts.

I've tried to give each voice part a different instrument, to aid in picking it out when all parts are played together, and to add variety. They all sound ok on my soundcard, but there's no guarantee that they'll do so on others. Things may therefore need some tweaking. (For example, I once came across a (cheap on-motherboard) soundcard with woodwind sounding much quieter than other instruments – so much so that it couldn't be heard at all if anything else was playing.)

In a few cases, there are more than 15 parts. Only 15 MIDI channels are supported on a single controller (excluding percussion on channel 10). My soundcard provides two controllers, so I can get up to 30 channels. In these cases, I've therefore made use of this feature. If the file is loaded into a machine with just a single controller, and you're lucky, the parts on corresponding channels on the two controllers may be merged. I've tried to group similar parts in each pair of channels, to minimise the impact of this situation. But not all configurations seem able to do this, so I've provided two versions of these files. To find which version of the files to use, see the midi test page.

When NWC exports to a MIDI file, it loses all timings from fermata and breath marks. While these effects can be simulated (with changes of tempo, extra rests, etc.) in a way that exports correctly, I've not usually done this. Playing the .nwc file directly doesn't suffer from this.

I've been using a patched version of nwc32.exe that I translated into English (with crotchets instead of quarter-notes, bars instead of measures, etc.). I don't think this has affected the saved files in any way, but I can't guarantee it.

Chris Hall <>

Valid XHTML 1.0! Valid CSS!