[linux-audio-dev] Problems and Solutions: events v. signals

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: [linux-audio-dev] Problems and Solutions: events v. signals
From: Roger Larsson (roger.larsson_AT_norran.net)
Date: la tammi  08 2000 - 21:39:05 EST


(David, you have received another one - delete that one)

Problem: events and signals are treated in different ways. Events may
 update signal buffer locations (as suggested). But a plugin have to
 handle events and signals in different ways.

Observation:
 What is the difference between a signal and an event?
 Not much!
 Events happens rarely.
 There might be a lot more event destinations than signal/connectors.

Solution:
 All events can be treated as an input signal with per fragment varying
 sample rate, if everything that an event might change are assigned to
 a signal/connector of its own.
 This way all parameters used by a plugin may be a signals too :-)

 normal signal event chain:
  {timeN, signal#A, *buffA1} {timeN+time(buffA1), signal#A,*buffA2}

 event, parameter change, signal chain:
  {timeM, param#P, value1} {timeN, param#P, value2}

 If 'time' is treated as described earlier in "P&S: time" there is no
 longer any need to distinguish between signals and parameters. They get
 with different time formats.

Note:
* a signal that is event driven. Can normally be driven with a constant.
  and only during fragments where something happens it changes to a
sampled
  signal format. (even if sampled all the time)

  {durationA1, variable#A, *buffA1} {durationA2, variable#A,*buffA2}
 
* event/signal queue gains by sorting it in variable# as first key.
  Times are always deltas - not sorted explicitly. Arrive sorted!

* The plugin _may_ chose different implementation depending on input
  'times'.

* Several outputs can not be connected to one input can not be connected
  directly.

* One output can be connected to several inputs.

Remaining problems:
 Will be expensive to check every fragment if there are
 lots of channels/signals...

/RogerL

--
Home page:
  http://www.norran.net/nra02596/


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : pe maalis 10 2000 - 07:23:26 EST