Forums > MaxMSP

morkov row

May 23, 2006 | 5:42 am

quicky. working on OSX,

Anyone have a method of taking a list of probability weights and
outputting a state? similar to filling a table and banging it.
Iterating through a list is proving to be pretty expensive. algo1
doesn’t seem to be working.

-matt

this is what I want to do but menus and such slow it down. I have no
solid speed test for it.

#P inlet 315 79 15 0;
#P outlet 312 188 15 0;
#P window setfont "Sans Serif" 9.;
#P window linecount 1;
#P newex 311 103 54 196617 t b l clear;
#P newex 320 128 52 196617 listfunnel;
#N vtable 128 20 74 230 241 4 128;
#P newobj 311 158 32 196617 table;
#P connect 2 1 1 0;
#P connect 0 0 3 0;
#P connect 2 2 0 0;
#P connect 2 0 0 0;
#P connect 1 0 0 0;
#P connect 4 0 2 0;
#P window clipboard copycount 5;


May 23, 2006 | 1:37 pm

Hi Matthew,

A table is in a sence a markov row. If you make the table small: 10-10 and you increase one of the values, a bang will send out this value. If you have a value on a another location which is higher, table will weight this and will mostlikely send out the new value.

I build a first-order markov processor in this way. If you have trouble building one, i can send it to you if you like.

Grt,

Danny de Graan


May 23, 2006 | 2:57 pm

hey, thanks for the reply…
the problem is, Im using these for 32nd notes, and (as in my example)
in order to fill a table you have to iterate through your whole list
in order to fill that table (like once every 80 ms). I was wondering
If I could do it from a list.
I know someone has…. just gotta right?
-matt


May 23, 2006 | 8:42 pm

Hi Matthew,

I found this external on the maxobjectlist on the net:

f0.list_arpeggio

Maybe it will do the trick.

Goodluck,

Danny de Graan


May 24, 2006 | 1:44 pm

> Anyone have a method of taking a list of probability weights and
> outputting a state? similar to filling a table and banging it.

in my probability sequencers i do not use lists but only
single values (of course), but i think with a messagebox
filled with variables you might have a better performance
than with a table.

if you want to be able to change 1 out of 100 numbers
in a list, you should use 100 single [i]s and a pack …
so that you update the list – stored somewhere behind it -
only when you need the list data.


May 28, 2006 | 3:58 pm

On 23-May-2006, at 7:42, matthew aidekman wrote:
> Anyone have a method of taking a list of probability weights and
> outputting a state?

prob?

————– http://www.bek.no/~pcastine/Litter/ ————-
Peter Castine +–> Litter Power & Litter Bundle for Jitter

iCE: Sequencing, Recording & |home | chez nous|
Interface Building for |bei uns | i nostri|
Max/MSP Extremely cool http://www.castine.de

http://www.dspaudio.com/


Viewing 6 posts - 1 through 6 (of 6 total)