Merging two sets of MIDI data simultaneously demands a little intelligence...
Last month, we looked at how you could alternate between two MIDI Controllers by using a MIDI Switcher, a relatively cheap device due to its passive circuitry and simple design. In passing, it was mentioned that such a device would not allow you to have two MIDI Controllers active at the same time. This is a job for a MIDI Merger.
While it may appear that a MIDI Merger is a relatively simple device for combining the MIDI information from two MIDI Outs, its design actually calls for the use of a microprocessor. The reason for this is that MIDI messages often consist of two or three parts and it is important to ensure that these are kept together. If they are not, the message becomes garbled. The MIDI Merger, therefore, has to be an intelligent device and as such costs far more than a MIDI Thru box or Switcher - typically around £80 for a two-into-one unit.
There are a number of situations which call for the use of a MIDI Merger. Let's look at a couple of examples. The first is where you need to share a sound module with another musician, each of you playing your own MIDI Controller. This is illustrated in Figure 1. For the sake of simplicity, two keyboards are shown, but really, they could be any type of MIDI controller - drum machine, guitar synth or wind controller - as long as they are capable of transmitting MIDI information from a MIDI Out port. The best results here will be with a multitimbral sound module which can play different instruments on different MIDI channels, and a pair of Controllers which can each be set to transmit on a separate MIDI channel.
Another situation requiring the use of a MIDI Merger is when editing the tones in a sound module using a visual editor on a computer. A two-way link between the computer and sound module needs to be established so that the computer program can request, and receive, information from the module. However, a keyboard also needs to be attached so that the edited voices can be played.
Figure 2 shows how this is set up and how a MIDI Merger is connected into the system: the MIDI Out from the keyboard is merged with the MIDI Out from the sound module and then sent to the MIDI In of the computer. The visual editor is likely to have a 're-channelise' facility so that the notes coming in from the keyboard can be set to the MIDI channel of the sound currently being edited in the module - an important point when working with a multitimbral module.
Next month, MIDI By Example begins its examination of synchronisation with a look at how sequencers and drum machines may be linked together.
Two further additions to our glossary of MIDI terms...
In 1987, various software companies decided that a common file format for the saving of songs on a sequencer was required. The Standard MIDI File, as it emerged, was passed by the MIDI Manufacturers Association (MMA) and implemented by practically every sequencer software writer, irrespective of the computer. MIDI Files have also now been universally accepted by all the major manufacturers and can now be found on the latest synths from Korg, Yamaha and Roland which have built-in sequencers.
This has led to a high degree of compatibility in the transfer of song data, although it should be noted that the inclusion of items such as tempo changes and time signatures are optional.
There are three types of MIDI File. Format 0 types consist of a single track and are used by many MIDI File playback devices such as the Yamaha MDF-2. Format 1 types comprise multiple tracks and are used most often when saving a MIDI File from a sequencer. The final type, Format 2, consist of multiple patterns but to date, have rarely been used.
Generally, a sequencer will not give you the option of saving as a Format 0 or 1 type file. If a song consists of multiple tracks and you need to save it as a Format 0 file, you have to merge all tracks into a single track.
Though a common three-letter extension for MIDI Files is used by Atari, PC and Amiga computers (ie, .MID), problems can occur when transferring MIDI Files between different computers. In general, PC compatibles and Atari STs can read MIDI Files from each other's disks, but both Amiga and Apple Macintosh computers have to pass data from a PC or ST through a conversion program.
MIDI Implementation Chart
One of these should be found at the back of every manual for a MIDI device. It outlines which of the various MIDI functions are transmitted and recognised by the device and how they are treated. Take the example of a keyboard: does it transmit Aftertouch? Can it transmit on all MIDI channels? Does it recognise MIDI Volume (controller #7)? What happens if you send it note numbers outside of the range of its sound generator? These questions, and many others, can be answered by knowing where to look in the chart and how to understand the comments.
A chart usually consists of four columns: Function, Transmitted, Recognised and Remarks. Function lists all of the various MIDI events, Transmitted and Recognised give listings of which events the device can send and receive, and Remarks explains any ambiguities with a '0' being used for yes and a 'X' for no.