[PD] arrays

[sme]

> is it possible to have more than one graphic representation for the
> same array in the same patch?

no, if you have more arrays with the same name, only one shows data

hi,

----- Original Message -----
From: "Orm Finnendahl" <finnendahl at folkwang-hochschule.de>
To: <pd-list at iem.kug.ac.at>
Sent: Thursday, May 16, 2002 10:35 AM
Subject: [PD] arrays


> I would like to have a couple of miniature "iconlike" representations
> of soundfiles on the screen (as arrays) and one current main buffer,
> which is represented in a large graphic array. Then I'd like to be
> able to switch the reference of the main buffer between the miniature
> buffers and get that reflected on screen.

the idea of mini-arrays is good, it saves space and cpu(?).

> Of course this would be possible by copying the values from the small
> buffers to the main buffer, but I would prefer a less expensive
> context switch. Or isn't it less expensive as the visual data has to
> be changed anyway?
> If I have to copy values from a miniature buffer to the main buffer,
> is there a predefined message for accomplishing it or has the data to
> get read out and written by some self-defined loop mechanism? How do I
> access the array size for that?

as far as i know the visual data is always renewed/calculated from the
sample/tabledata, and it is in fact maybe not the best graphical way of
representating it. i always try not to display more data than available
pixels. and i am working on a method that will show (with two arrays in
one graph) the minimum and maximum value of all the samples represented
by one pixel. (which will be an abstraction-like calculation in a
subpatch).
by now i live with "downsampled versions" of soundfiles for the
graphical display, which has only the length (in samples) that the
display can show (200 pixels - 200 samplevalues) and use a simple
tabread4~ to calculate that. the "real" soundfile is in a table-object
somewhere. the display is faster and does´nt grab to much from the cpu.
before the tabread~ i dont use a phasor-loop, but a ramp-table from 0 to
1 which has the same length as the size of the "display" and that is
read via a playtable~ everytime i load a new file into a table (and then
is switched~ to 0 again).
marius.