Re: [sc-users] Re: will there ever be real multiprocessor support?

[James Harkins <jamshark70@xxxxxxxxx>]

On Mar 6, 2009, at 5:34 AM, Tim Blechmann wrote:

one could think of a supercollider-style pseudocode like:

~rec = Synth.new('record', recargs); // record buffer

~fx  = Synth.new('fx', fxargs);      // effect engine

~players = Array.fill(20, {

    Synth.new('player', playerargs,

              [~rec, \addAfter, ~fx, \addBefore])

    });                              // 20 parallel players

~rec = Synth(\record, ...);

~parallelGroup = ParGroup(target: ~rec, addAction: \addAfter);

~fx = Synth(\fx, [...], target: ~parallelGroup, addAction: \addAfter);

~players = Array.fill(20, {

    Synth('player', playerargs,

              target: ~parallelGroup)

    });                              // 20 parallel players

To me, [~rec, \addAfter, ~fx, \addBefore] is confusing. The alternative I'm suggesting is pretty well in keeping with what we already have -- easy to understand -- and would allow the same level of control.

It wouldn't even be necessary to \addAfter a specific node. If ~rec, ~parallelGroup and ~fx are added successively, in that order, to the tail of the same *serial* group, it would evaluate exactly the same. ~rec precedes the parallel block, and the whole parallel group has to finish before ~fx can do anything.

hjh

: jamshark70@xxxxxxxxxxxxxxxxx

: http://www.dewdrop-world.net

.::!:.:.......:.::........:..!.::.::...:..:...:.:.:.:..:

"Come said the Muse,

Sing me a song no poet has yet chanted,

Sing me the universal."  -- Whitman