Q: “Why do you think a serial interface is used instead of a parallel interface for the MIDI protocol?”
A: While this may be up for argument, the general belief is that serial was chosen by MIDI manufacturers because it was less expensive and more reliable than a parallel interface at that time (early 1980s). Other technologies that were once parallel have now gone to a serial protocol as well, in large part due to cost and reliability issues. For example, SCSI, a parallel data transfer protocol has largely been replaced by Firewire, a serial protocol, though there are still some high end applications where SCSI is preferred. For internal drives SCSI has largely been replaced by technologies such as ATA, which are much less expensive and can be in the ballpark in terms of speed. But, even ATA, which is parallel is now beginning to be replaced by SATA (Serial ATA), and the net result is actually increased speed, among other things. So, while parallel may seem better/faster on paper it isn’t necessarily so in practice. It depends on many factors, most of which are well beyond the scope of inSync.
The speed of a MIDI serial interface is 31,250 bits per second. There are 10 bits needed for every MIDI digital word or 3125 messages per second (snap your finger and think about how many events could be transmitted during that time). In other words, that was a lot of room for MIDI data in 1983, and by in large it still is today, though there are some applications where the speed of MIDI transmissions is a liability. Thus we’ve begun to see some adaptations to MIDI that help provide tighter timing resolution if not more overall bandwidth. These include innovations such as MTS and LTB. We’ll probably see more and more enhancements to the basic capabilities of MIDI in the next few years.