the zip attached here contains a file granulator patch i am working on, and would like to ask for a little help with. the idea is based roughly on the "Travellizer" patch from Reaktor, and uses various randomizers to read from a tabread4~, and a line~ object to create each individual grain. the version here has 32 voices maximum, and an "oscillator" which can be used to scan back and forth in the sample.

"filegranulator1111.pd" is the parent patch, with multiple instances of "gran111reader.pd" creating each individual voice.

the problems i am having are these:

1. converting between milliseconds and samples in my patch for some reason

turns into some very cryptic math [at least for me]. i can't quite get my little tabread4~ to behave predictably. see "gran111reader.pd" for details.

2. the voices are set by a >= object, with a seperate value for each given as

creation argument. increasing the voices causes no trouble, but decreasing causes a sharp spike in CPU activity and an interruption in the audio stream. can anybody tell me why?

3. i cannot create multiple instances of "filegranulator1111.pd" because the

tables involved don´t seem to want to take their name from the patch creation argument. how is this possible?

thanks in advance for any suggestions. the completed patch will be available here and on a website when i know it is at least beta quality :-)

best, derek

Quoting "vespa@muse" dowlasz@gmx.net:

---

*****1) i can only load about a minute of audio into this table. i always get "soundfiler error : truncated to 40000000 elements" with larger files. is there another type of table that can hold more samples?

******hi, when you send "read" message to soundfiler send also "maxsize" . for example. -

[read -resize -maxsize 99999999999999 - "your_buffer" ]

maxsize...... it tells soundfiler the max array size. so take whatever you want -for me it works.

---

2. converting between milliseconds and samples in my patch for some reason

turns into some very cryptic math [at least for me]. i can't quite get my little tabread4~ to behave predictably. is there another type of table/soundfiler that i can give variable read locations and durations in milliseconds instead of samples? [more along the lines of the way delay tables work in PD...]

what do you mean with cryptic math?

blazej

----- Original Message ----- From: derek@x-i.net To: pd-list@iem.kug.ac.at Sent: Sunday, August 17, 2003 6:29 PM Subject: [PD] tabread4~ granulator: soundfile error

...

hi gang,

i am working on a soundfile granulator, based on tabread4~, and i have a

couple

...

of questions.

1. i can only load about a minute of audio into this table. i always

get "soundfiler error : truncated to 40000000 elements" with larger files.

is

...

there another type of table that can hold more samples?

2. converting between milliseconds and samples in my patch for some reason

turns into some very cryptic math [at least for me]. i can't quite get my little tabread4~ to behave predictably. is there another type of table/soundfiler that i can give variable read locations and durations in milliseconds instead of samples? [more along the lines of the way delay

tables

...

work in PD...]

3. how about stereo??? is there a way to split a stereo soundfile into two

tables? otherwise, i am stuck with mono, which i assume is only the left channel of the stereo sound file...

thanks for any suggestions. i'll be happy to post the completed gadget to

the

...

list when it works right.

best, derek

---

This mail sent through IMP: http://horde.org/imp/

---

PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list

---

PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list

This mail sent through IMP: http://horde.org/imp/

Where can I find random_range?

sh*t!!!!!!!! i knew i was missing something in that zip! ;-) i will post a new one later.

thx, d.

This mail sent through IMP: http://horde.org/imp/