Re: [AH] DIN Sync Splitter and DIN Sync timing gotchas

From Robin Whittle
Sent Fri, Nov 28th 2003, 15:49

Colin f wrote:

> But if you buffer clock bytes, and interpolate with a moving average of
> their period, you are going to introduce timing errors even in a rock stable
> midi clock where there is a sudden change in tempo. This may even be
> audible.

My use of the term "buffering" may have been misleading.  I wasn't
suggesting any interpolation etc.  My suggestion is not to send out
Clock cycles any faster than a particular limit, and if MIDI Clock bytes
arrive faster than this, then to keep a count of them and send the Sync
Clock cycles as fast as possible until the right number has been sent.

While it may be sloppy programming in a sequencer sending out clock
bytes faster than is musically needed, it could happen due to the
program not getting all the CPU time it needs when it needs it.  In any
event, that is beyond our control, and a MIDI sequencer will stay in
time with the erroneously timed MIDI Clock codes, so a Sync converter
should try to do the same.

  - Robin

  Devil Fish mods for the TB-303