Tim Hockin wrote:
First attempt at wrapping up UST, which will fold up into the bigger wrap up attempt #2.
The main point of UST is to relate the audio clock to a neutral clock. This allows GMPI plugs to sync to other clock sources (such as video). For a good read on UST: http://www.lurkertech.com/lg/time/intro.html
Because UST is not defined on some platforms, if we want to use it, GMPI will need to define the UST source and resolution on every platform GMPI supports. This may be best as a separate sub-spec.
It was suggested that plugins which need UST get it themselves. The problem here is jitter. Two plugins will get different UST times for the same audio frame, with a non-deterministic jitter to the actual UST time for that frame. Worse, a plugin which is procesed later than another might get a later UST time for an earlier sample frame. If the process buffer size is small enough, maybe this jitter does not matter? Can we guarantee that the process buffer size is small enough? I think no.
So my position is that we should support UST in the following manner:
1) The host or audio driver snapshots the UST when the buffer becomes available for a process() run.
2) Plugins are notified of this UST time along with the sample counter when they are called for process(). This might be an argument to process() or an event, or a host callback - it doesn't matter for now.
3) There must be some way to calculate the UST for any sample. Either the host exposes a UST_per_sample scalar or there is a host callback or something. The host is responsible for recalibrating any scalar sufficiently often as to avoid drift.
4) UST support, as described above, is required of all GMPI hosts.
-- Mike Berry Adobe Systems
---------------------------------------------------------------------- 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
