I think the distinction between the high-level editor and the low-level
editor is somewhat artificial. It also is not necessary if the program gives
the user the ability to take a patch created in the editor (like your
example) and turn it into a box that appears in the palette or pick list for
building other patches in the editor.
Reaktor is a good example of this kind of thing. It does have two levels of
editing, but it also gives the user the ability to create high-level boxes
using the low-level editor, and it comes with a fairly extensive toolkit of
pre-built patches.
In either case, the ability to make user-defined opcodes using the GUI
editor is vital.
-----Original Message-----
From: Bert Schiettecatte [mailto:bschiett@vub.ac.be]
Sent: Sunday, November 07, 1999 4:03 PM
To: 'saol-dev@media.mit.edu'
Subject: SAOL Authoring tool: the lowest level ...
Hi,
As you all known, i'm currently designing my authoring tool for SAOL.
I've decided there will be 4 levels in the application: the workspace to
arrange instruments & effects with buses, the workspace to arrange synthesis
blocks into an instrument (synthesis blocks being filters, oscillators, ...
-> opcodes?), and the workspace to arrange atomic operators into
user-defined synthesis blocks (user opcodes?).
I'm currently doubting if this lowest level is a good idea. What would be
the
best idea: to create a language-independent (if that is possible) graphical
connect-the-operators interface, or only offer to edit SAOL code directly?
or both?
To test if an imported piece of SAOL code would render to an understandable
graph,
i've converted the expression
a[j] = 2*r[j]*cos(2*pi*(freq[j]/s_rate));
from Mr. Lazzaro's marimba instrument into a diagram which would
be the result of importing SAOL code (or constructing it by hand) into my
authoring tool at the lowest level (graphical details not concerned).
The result is attached (JPG file).
I would appreciate any comments on this approach, so that I won't design
this part of the application the wrong way.
Thanks for your time,
Bert Schiettecatte.
This archive was generated by hypermail 2b29 : Wed May 10 2000 - 12:15:44 EDT