Question: Is there always going to be a sample reference clock? Can't a
GMPI graph be composed of entirely event I/O objects, such as Ron's
arpeggiator, without any notion of sample time?
Seems to me in this case we have to rely on a non-sample clock reference
for time, such as the UST proposed earlier.
For plugs that generate/process sample streams, such as effects and synths,
the sample clock reference will be preferred. Again, this would suggest
that the host provide both sample clock and non-sample-clock reference
time, as suggested earlier. Plus I agree we need both the absolute and the
transport times.
Another question: Aren't we already agreed that sample rate can only be
changed with a graph reset, i.e. never on the fly? That is, a GMPI graph
that has synchronous sample data has a SINGLE sampling rate, and it can
only be changed when the graph is stopped.
One thing is ESSENTIAL. Synchronization events must have at least sample
period accuracy, so that GMPI will be able to phase lock two mono sample
streams to make a proper stereo stream, for example. So, if these events
are encoded using "musical" time, which seems likely, the conversion must
unambiguously resolve to the proper sample period, no exceptions. I don't
think this can possibly screw up, if two plugs are sent the same musical
time information, they have to calculate the same sample index for
synchronization, right? Yes. And this phase locking must be maintained even
if the host is syncing to an external source. This means that the host must
be consistent in sending the same time information to all plugs in the graph.
I think we're also agreed that the host must supply conversion to/from
musical and sample time, when it makes sense that is, which makes a lot of
the discussion moot.
I find myself leaning toward the synthesis of the Ron, Todor, Angus et al.
discussion earlier, but I need to reread everything to be sure.
Bill
---------------------------------------------------------------------- Generalized Music Plugin Interface (GMPI) public discussion list Participation in this list is contingent upon your abiding by the following rules: Please stay on topic. You are responsible for your own words. Please respect your fellow subscribers. Please do not redistribute anyone else's words without their permission.
Archive: //www.freelists.org/archives/gmpi Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe
