Subject: Re: [linux-audio-dev] News about sequencers (not my own though!)
From: Paul Barton-Davis (pbd_AT_Op.Net)
Date: ma tammi 24 2000 - 22:22:23 EST
Not to blow my own (or in some senses, Barry Vercoe's) trumpet too
loudly, but Quasimodo (and similarly Csound) tries to plot a middle course
through the interpreted/native jungle.
True, right now, the language (Csound) is horrible, which is why
Quasimodo has various hooks to plug in one or more different
languages. But in every case, we end up running the code through a
fairly specialized compiler and get down to thread code (linked lists
of function pointers) along with an inlined arithmetic/boolean logic
implementation. The inner loop is obviously written in a native
language as well.
So I think that it is entirely possible to use an interpreted language
and still get excellent real-time performance (i've run Quasimodo with
2ms latencies, without dropouts, though not often). However, we still
need a good language, and I'm with jfm3 (a.k.a. "mr quasimodo-rpm") in
recognizing the merits of keykit as a model of integrating musical
concepts into an existing programming language.
Bill Gribble is doing some work for Quasimodo to bring in Guile as the
configuration language, and we're going to evaluate the possibility of
a more Guile-ish music language as well. However, there are years of
research papers out there on this stuff and its one of those things
where it would be stupid to ignore that body of work.
--p
This archive was generated by hypermail 2b28 : pe maalis 10 2000 - 07:23:27 EST