C-LAB

MUTES

The Mute column to the right of the arrange window has two muting uses that need occasional reminders. The track mutes can be used to program the real time muting that is part of the 'Screen Recording' ability of Creator and Notator: clicking these whilst in Record mode records the muting and demuting commands in the current track. However, the Mute column is also used by the arrange mode for its 'arrange mutes', namely, the ability to give each pattern in the arrangement a set of track mutes that mute the tracks for the duration of that pattern (no need to record — just click the desired mute).

It is possible to get the two to usefully interact, eg. by starting a pattern in the arrange mode with a track muted ('arrange mute') then demuting it a couple of bars later with a demute command (Screen Recording). The confusion arises when you switch off the arrange mode to concentrate on one pattern, but you left the arrange cursor on a pattern which has arrange mutes muting some of the tracks. This will interfere with the current pattern; do not be tempted to click them off, or you will not have them again when you switch the arrange mode back on. Instead, it is better to leave the arrange cursor on a pattern that has no arrange mutes, then scroll the pattern number at the top of the pattern window to the desired pattern. (Pssst! Version 3.1 cures this...)

STEINBERG

SONG TRANSFER AND TEMPO CHANGES

There always seems to be a certain range of questions that never go away, however many times answers are given. Apart from the old chestnut of how many MIDI Thru connections make a noticeable delay, the other common problem area is transferring complete songs from one sequencer to another.

Generally this means that people want to transfer the MIDI data from the computer sequencer in which it was written to a hardware sequencer for live performance. The procedure itself has been covered in recent months in this column, when we talked about making sure the data was actually sent in the most efficient way. To briefly recap:

The MIDI data is sent between the source and destination via one MIDI lead, while the synchronising data is sent in the other by a second lead. In this case the 'receiver' of the MIDI note data was the 'transmitter' of the MIDI clock data, which ensures that your note data is not pushed out of the way by the clock data.

So as soon as that was settled, someone comes up with a new problem; why isn't the Mastertrack data being transferred between the two devices? Due to the very nature of a system locked by MIDI clocks, it cannot be done, unless special provision has been made in the hardware sequencer.

The fundamental problem lies in the nature of the synchronisation between two systems locked together by MIDI clock data. MIDI clock allows a form of synchronisation that I call position sync, as opposed to time sync (which is what MIDI timecode and SMPTE allow) These terms are not universally recognised, mainly because I made them up in explanations to the unsync'ed masses.

Position sync works because one device is telling another to move its actual position in a piece of music by a small section of a beat. After any data associated with that sub-beat has been sent, the master device can send another message to move the slave on by the same unit of a bar.

The reason the tempo changes that occur in the Mastertrack of the master device never appear in the slave's recollection of what happened in the song is that the slave isn't interested in the actual rate of arrival of the MIDI clocks (ie. the tempo) but merely what happens on each step of the way. The slave is oblivious of the fact that the MIDI clock rate is changing in the master device (which happens because the tempo is changing). When the slave plays back the song it steps through each pulse of the song linearly, unless you subsequently use its own tempo control procedures.

In fact it is possible to transfer a sequence from one sequencer to another and have the Mastertrack tempo changes realised in the slave. It is actually done by not trying to synchronise the two systems. Imagine the slave is put into record, and then the master is set to play, but no external synchronisation is selected on either. All the MIDI data is recorded from one to the other, and the rate of transmission is preserved in the final recording, as the slave was simply recording everything that arrived on each pulse of its internal clock.

There are some compromises however: the slave device used like this cannot really be thought of as a sequencer. What you are really doing is making the MIDI equivalent of a tape recording of your music, as the timebase (the bars, beats and clocks) of the slave's memory will have no relation to the perceived timebase of the MIDI data it plays back. This means that editing will be tough, but then again the whole point was to get a sequence that will play back for live performances, and I haven't come across many audiences that have much patience for live MIDI editing.

If you want to try this idea, there is one last thing you should be think about: the tempo of the slave sequencer during this operation. Surely it doesn't matter, you cry. In a perfect world it wouldn't, if the slave sequencer had a infinite record resolution. But as this is less than likely it is best to set the slave sequencer's tempo to (near) maximum before you start the recording.

The reason is that if you take two finite resolution systems which are not synchronised, you will get a degree of event time shifting as the recording device waits for its next time pulse to come along. This effect is minimised by increasing the apparent resolution of the recording device by increasing its tempo. To use an analogy from the world of sampling, you are actually increasing the sampling rate of the time of arrival of the notes to be recorded, hence making a more accurate recording.

DR T's KCS

MUTE NEW TRACKS

When you're working on a solo section or you just want to have several goes at getting that blistering bass line right without disturbing your flow, set KCS's Auto Record on and turn the Mute New Tracks toggle on. You can now continuously record multiple takes and each new track you record will be muted. You can then audition each one at your leisure and edit the best bits to suit.

TIME COMPRESSION

There may be occasions when you need to alter the overall length of a track or sequence in order to fit it to a specific time interval, particularly when the piece has been recorded in real time and has no reference to the bar lines. While you could use the Compress/Expand tool to to do this, you'll need to go poking around with a calculator to work out the percentage decrease or increase required. You could also alter the tempo, but if you're working to specific timecode points already established, or you want to make one phrase fit against another, it may prove easier to try the following.

Let's assume that your sequence is 1m 24s long, and it needs to be 1m 12s long (we're assuming that there are no tempo events buried in the sequence — you can deal with these but it's a little trickier). First copy the sequence to Track Mode (use the Seq to Track option and not Seq to All Tracks) and make sure that the first event is right at the start of the track. Go to Tiger and display this track — if the sequence is multi-channel do not alter the channel number in the Track Info line line, or else everything else will be switched to that channel. Check that the tempo in the conductor track is set to the correct value and switch on the Real-time Mouse Locator. Ensure that the end of the track shows the right time by placing the cursor at the end; you should see "1m 24s" in this example.

Zoom the window to show the entire track, toggle the Range icon on, Quantise Range Borders off, and draw a range encompassing everything — if the data is particularly dense, you might see the highlighted range 'jerk' across the screen as you get towards the end of the track (it depends on the density of the music data). To fix the end of the range accurately, re-zoom into the end of the track and shift-click the end of the range into position — the mouse position indicator at the bottom centre of the screen will report the time at the end of the range. Having set the range up, select the Move tool from the Edit menu and select the Stretch/Shrink With Mouse option. You can now place the mouse at the right point in time — you can re-zoom and scroll the window safely without losing the highlighted range to find the right point. Left-click at the required point and the data will compress accordingly; remember to right-click to de-select the Stretch/Shrink tool when you've finished.

PLEASE NOTE: Product information contained within these pages is supplied directly by the software manufacturers, or their UK distributors or agents. The intention is to provide a 'bulletin board' service for SOS readers who own or use software for any type of computer. Although we may occasionally publish new product information, the idea is to publicise update/upgrade news, bug fixes, and hints and tips about software and computer peripherals. It is therefore up to all software companies to keep us posted.