hi richard, paul, stefan and fellow LADders,
the Debian folks currently are a bit at a loss of the proper
interpretation of the LADSPA_PROPERTY_REALTIME property.
ladspa.h says:
/* Property LADSPA_PROPERTY_REALTIME indicates that
the plugin has a
real-time dependency (e.g. listens to a MIDI device) and so its
output must not be cached or subject to significant latency. */
the discussion started about a year ago (see [760758]), and it seems
that there are two opposing interpretations of that property:
#1 setting the property indicates to the host, that the plugin must not
be used in a non-realtime environment.
this means that a plugin that has this property set, must not be used
for batch processing, e.g. because it uses wall-clocked input (like
MIDI-devices) that simply won't deliver proper input when running at a
speed that is decoupled from wall clock.
OR
#2 setting the property indicates to the host, that the plugin can be
run at a lower (timing) priority; so the host can deliberately add
latency without compromising the usefulness of the full processing
chain. that would be mostly for plugins that do not do any input ->
output processing, but only take input (e.g. metering, recording).
that's just a quick summary. there's more arguments in the bugreport
[760758].
personally i lean towards #1 (as i don't see much use cases for the
second interpretation), but given that someone as deeply involved into
linux audio as fons favours #2, i would like to ask the creators of
LADSPA for their canonical view.
thanks for your insights,
gfmadsr
IOhannes
[760758]
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=760758