7 Mar
2005
7 Mar
'05
8:10 a.m.
Hello there,
As you might guess, I'm new to the list. Last year I started to develop interest
in using the computer to create music - partly because I'm interested in music
AND computers: so why not join these two things? - and partly because my band
broke up due to the drummer who left town: so why rely on humans, when you can
use a machine? ;-)
Well, I started reading about recording- and mainly synthesis-techniques. Just
recently I started reading the "Computer Music Tutorial", which some of you
probably know. It is highly interesting and I think I learned a lot about this
topic, so far.
Unfortunately, I'm not quite sure about how to use these synthesis techniques in
"real life". I mean, I know the basic principles of a few techniques and the
basics on waveforms etc. but how can I get the sound I'm looking for out of my
softsynth?
In terms of software synthesizers, so far, I only looked into Alsa Modular Synth
and Zynaddsubfx. I like both of them, but I think, to apply techniques like
multiple wavetable synthesis or additive analysis/resynthesis (which sounded
particularly interesting to me) and to get a wide range of different sounds, I
need to get my hands on more "sophisticated" tools like Pd, CSound or whatever
else there is. But actually, I have no idea, which one to choose. The point is,
that I really don't want to delve into one of these languages, only to find
out, that I should have learned another one...
So... what is the tool to use? I want to be able to do the synthesis techniques
described in the before mentioned book, it should be controllable in realtime
(at least for less resource-intensive patches). With "controllable" I mean,
that one should be able to easily change important parameters, e.g. by
assigning MIDI-controllers to them and changing these values over time either
manually or letting a sequencing tool change them. It would be nice, if output
via jack would be supported, so it can be integrated with other Software tools,
as well as my hardware synth and drumcomputer.
I'd be glad, if someone could help me out about my problem...
btw: I hope not to start some kind of religious war... ^_^
thanks in advance,
-Michael Wagner
7 Mar
7 Mar
10:10 a.m.
Hallo,
Michael Wagner hat gesagt: // Michael Wagner wrote:
In terms of software synthesizers, so far, I only
looked into Alsa
Modular Synth and Zynaddsubfx. I like both of them, but I think, to
apply techniques like multiple wavetable synthesis or additive
analysis/resynthesis (which sounded particularly interesting to me)
and to get a wide range of different sounds, I need to get my hands
on more "sophisticated" tools like Pd, CSound or whatever else there
is. But actually, I have no idea, which one to choose. The point is,
that I really don't want to delve into one of these languages, only
to find out, that I should have learned another one...
Well, you will need to do just exactly that: Dive in and try things
out. We all did, we all chose some tools first which we don't use
anymore, but still got something valuable out of them. For example I
started with Csound. I even bought the Csound Book, which is now
collecting dust on my shelf, as I don't use Csound at all anymore. I
tried some more software and then settled on Pd, which still is
exactly what suits my working habits and aesthetics best with its
realtime nature and visual approach. (I'm also capable of "real
programming" with text editors however I like to do music in other
ways.)
So... what is the tool to use? I want to be able to do
the synthesis
techniques described in the before mentioned book, it should be
controllable in realtime (at least for less resource-intensive
patches). With "controllable" I mean, that one should be able to
easily change important parameters, e.g. by assigning
MIDI-controllers to them and changing these values over time either
manually or letting a sequencing tool change them. It would be nice,
if output via jack would be supported, so it can be integrated with
other Software tools, as well as my hardware synth and drumcomputer.
As I wrote above nobody except you can tell what you will like but
from your descrption I guess: You are a typical Pd user! Give it a
try. You also might want to try SuperCollider or Csound. Pd is
powerful, but not really hard to get into, especially not while you
are reading a book like the CMT or the Dodge/Jerse book.
You should subscribe to pd-list: http://iem.at/mailinglists/pd-list/
You should then start with reading Pd's html manual, which many
beginners unfortunatly like to skip and then they ask all those
questions which are explained in the html manual already. This is a
bit annoying of course. ;) Better read it first and afterwards ask
about the points you didn't fully understand. The html docs are
online here: http://www-crca.ucsd.edu/~msp/Pd_documentation/
After you read the html, go through the documentation patches once.
They are numbered for a reason: follow that order, and don't skip the
"2.control.examples" part just because you want to do audio stuff
immediatly. These examples are very important, too, to get the bigger
picture. All Pd patches can be edited, changed, copied etc. Doing
this is *the* way to learn Pd.
Ciao
--
Frank Barknecht _ ______footils.org__
_ __latest track: fqdn _ http://footils.org/cms/show/38
11:09 a.m.
Frank Barknecht wrote:
[re: learning Pd]
After you read the html, go through the documentation
patches once.
They are numbered for a reason: follow that order, and don't skip the
"2.control.examples" part just because you want to do audio stuff
immediatly. These examples are very important, too, to get the bigger
picture. All Pd patches can be edited, changed, copied etc. Doing
this is *the* way to learn Pd.
In all things Pd, Frank is my guru, but I'm a devotee of the "plunge &
sink" school of using computers for making music: Dive into what looks
interesting, work there for a while, let things branch naturally, follow
them to other interesting places in the system. I think Pd lends itself
equally to rational and irrational approaches to learning and using its
system.
Best,
dp
11:38 a.m.
New subject: [linux-audio-user] Synthesis Tools: What to use and how to use it?
On Mon, 2005-07-03 at 10:26 -0500, Dave Phillips wrote:
Frank Barknecht wrote:
[re: learning Pd]
The nice thing about learning Pd is that even if you dislike it as a
synth, it's useful to know for a ton of other things (routing MIDI or
sending OSC, building control panels, prototyping, etc. etc).
So, learning it isn't a waste of time, even if you don't end up using it
as your synth of choice.
++pd;
-DR-
After you read the html, go through the
documentation patches once.
They are numbered for a reason: follow that order, and don't skip the
"2.control.examples" part just because you want to do audio stuff
immediatly. These examples are very important, too, to get the bigger
picture. All Pd patches can be edited, changed, copied etc. Doing
this is *the* way to learn Pd.
In all things Pd, Frank is my guru, but I'm a devotee of the "plunge
&
sink" school of using computers for making music: Dive into what looks
interesting, work there for a while, let things branch naturally, follow
them to other interesting places in the system. I think Pd lends itself
equally to rational and irrational approaches to learning and using its
system.
Best,
dp
1:17 p.m.
Hallo,
Dave Phillips hat gesagt: // Dave Phillips wrote:
Frank Barknecht wrote:
[re: learning Pd]
It probably was a bit unclear what I meant with my last sentence. I
meant to express: "Editing, changing, and copying other patches is
*the* way to learn Pd." Like driving a car is *the* way to learn how
to drive a car. Nevertheless if you only know how to drive a car, but
don't know how to handle a red traffic light, your driving career
might come to a sudden and premature stop. ;)
I think, with Pd it's very easy to forget about the fact, that one is
actually learning a language. With Csound or SuperCollider you are
forced to first learn at least the basics of the language, but Pd is
so easy to get started with, that the finer nuances of using it are
often missed.
Ciao
--
Frank Barknecht _ ______footils.org__
_ __latest track: fqdn _ http://footils.org/cms/show/38
These examples are very important, too, to get the
bigger picture.
All Pd patches can be edited, changed, copied etc. Doing this is
*the* way to learn Pd.
In all things Pd, Frank is my guru, but I'm a devotee of the "plunge
& sink" school of using computers for making music: Dive into what
looks interesting, work there for a while, let things branch
naturally, follow them to other interesting places in the system. I
think Pd lends itself equally to rational and irrational approaches
to learning and using its system. 
9 Mar
9 Mar
2:10 p.m.
New subject: [linux-audio-user] Synthesis Tools: What to use and how to use it? (Wiki summary???)
I think that many interesting things had been said in this thread.
I've set up a Wiki to post some of this info there. If somebody want to
help me... you'll be welcome!
http://twiki.org/cgi-bin/view/Wikilearn/LA_SynthesisTools
(need to register with name+email, sorry for the inconvenience)
En/na Dave Phillips ha escrit:
Frank Barknecht wrote:
[re: learning Pd]
After you read the html, go through the
documentation patches once.
They are numbered for a reason: follow that order, and don't skip the
"2.control.examples" part just because you want to do audio stuff
immediatly. These examples are very important, too, to get the bigger
picture. All Pd patches can be edited, changed, copied etc. Doing
this is *the* way to learn Pd.
In all things Pd, Frank is my guru, but I'm a devotee of the "plunge
&
sink" school of using computers for making music: Dive into what looks
interesting, work there for a while, let things branch naturally,
follow them to other interesting places in the system. I think Pd
lends itself equally to rational and irrational approaches to learning
and using its system.
Best,
dp
6:43 p.m.
New subject: [linux-audio-user] Synthesis Tools: What to use and how to use it? (Wiki summary???)
I've merged all this stuff into
http://twiki.org/cgi-bin/view/Wikilearn/LA_SynthesisTools
I have copied many of your comments literally.
Please, let me know if you don't like them or want to do some changes.
Otherwise, feel free to edit it.
I did this because I wanted to take notes of your comments and give them
to the others who may be interested.
I hope it can be useful.
En/na MarC ha escrit:
I think that many interesting things had been said in
this thread.
I've set up a Wiki to post some of this info there. If somebody want
to help me... you'll be welcome!
http://twiki.org/cgi-bin/view/Wikilearn/LA_SynthesisTools
(need to register with name+email, sorry for the inconvenience)
En/na Dave Phillips ha escrit:
Frank Barknecht wrote:
[re: learning Pd]
After you read the html, go through the
documentation patches once.
They are numbered for a reason: follow that order, and don't skip the
"2.control.examples" part just because you want to do audio stuff
immediatly. These examples are very important, too, to get the bigger
picture. All Pd patches can be edited, changed, copied etc. Doing
this is *the* way to learn Pd.
In all things Pd, Frank is my guru, but I'm a devotee of the "plunge
& sink" school of using computers for making music: Dive into what
looks interesting, work there for a while, let things branch
naturally, follow them to other interesting places in the system. I
think Pd lends itself equally to rational and irrational approaches
to learning and using its system.
Best,
dp
7 Mar
7 Mar
11:36 a.m.
and after you've finished The CMT you can start with this one:
http://www.crca.ucsd.edu/~msp/techniques/latest/book-html/
comes with examples for PD.
m
Frank Barknecht schreef:
Hallo,
Michael Wagner hat gesagt: // Michael Wagner wrote:
--
--------------------------------------------
----------- http://www.mprims.tk
--------------------------------------------
In terms of software synthesizers, so far, I only
looked into Alsa
Modular Synth and Zynaddsubfx. I like both of them, but I think, to
apply techniques like multiple wavetable synthesis or additive
analysis/resynthesis (which sounded particularly interesting to me)
and to get a wide range of different sounds, I need to get my hands
on more "sophisticated" tools like Pd, CSound or whatever else there
is. But actually, I have no idea, which one to choose. The point is,
that I really don't want to delve into one of these languages, only
to find out, that I should have learned another one...
Well, you will need to do just exactly that: Dive in and try things
out. We all did, we all chose some tools first which we don't use
anymore, but still got something valuable out of them. For example I
started with Csound. I even bought the Csound Book, which is now
collecting dust on my shelf, as I don't use Csound at all anymore. I
tried some more software and then settled on Pd, which still is
exactly what suits my working habits and aesthetics best with its
realtime nature and visual approach. (I'm also capable of "real
programming" with text editors however I like to do music in other
ways.)
So... what is the tool to use? I want to be able
to do the synthesis
techniques described in the before mentioned book, it should be
controllable in realtime (at least for less resource-intensive
patches). With "controllable" I mean, that one should be able to
easily change important parameters, e.g. by assigning
MIDI-controllers to them and changing these values over time either
manually or letting a sequencing tool change them. It would be nice,
if output via jack would be supported, so it can be integrated with
other Software tools, as well as my hardware synth and drumcomputer.
As I wrote above nobody except you can tell what you will like but
from your descrption I guess: You are a typical Pd user! Give it a
try. You also might want to try SuperCollider or Csound. Pd is
powerful, but not really hard to get into, especially not while you
are reading a book like the CMT or the Dodge/Jerse book.
You should subscribe to pd-list: http://iem.at/mailinglists/pd-list/
You should then start with reading Pd's html manual, which many
beginners unfortunatly like to skip and then they ask all those
questions which are explained in the html manual already. This is a
bit annoying of course. ;) Better read it first and afterwards ask
about the points you didn't fully understand. The html docs are
online here: http://www-crca.ucsd.edu/~msp/Pd_documentation/
After you read the html, go through the documentation patches once.
They are numbered for a reason: follow that order, and don't skip the
"2.control.examples" part just because you want to do audio stuff
immediatly. These examples are very important, too, to get the bigger
picture. All Pd patches can be edited, changed, copied etc. Doing
this is *the* way to learn Pd.
Ciao 
12:26 p.m.
Michael Wagner wrote:
The point is,
that I really don't want to delve into one of these languages, only to find
out, that I should have learned another one...
So... what is the tool to use?
I second the other posters that said you may have to accept starting
somewhere and ending somewhere else.
I would however raise a flag for csound. It is well documented, stable,
highly flexible and has a large user base. Don't let it's text-based
nature scare you, it's language is quite simple and very powerful.
I tried out pd and found that I spend too much time rearranging the
graphics. Also I wasn't able to get dynamic voice allocation which is a
must-have for me and the default behaviour of csound.
If there were better docs (and I had more time) I'd see if I could get
supercollider going. It seems it has a cleaner and more modern language.
Just my .02
--
peace, love & harmony
Atte
http://www.atte.dk
1:08 p.m.
Hi,
Michael Wagner wrote:
So... what is the tool to use? I want to be able to do
the synthesis techniques
described in the before mentioned book, it should be controllable in realtime
(at least for less resource-intensive patches). With "controllable" I mean,
that one should be able to easily change important parameters, e.g. by
assigning MIDI-controllers to them and changing these values over time either
manually or letting a sequencing tool change them. It would be nice, if output
via jack would be supported, so it can be integrated with other Software tools,
as well as my hardware synth and drumcomputer.
I'd be glad, if someone could help me out about my problem...
what's the right tool, really depends on what you want to do and what
your preferred way of working is.
Pd is graphically oriented in its programming, which may be handy if
you're visually oriented and don't mind drawing lines between boxes a lot.
If you prefer typing over moving the mouse, then it is better to use a
text-based synthesis program.
SuperCollider is quite flexible, and not as horribly documented as a lot
of people may tell. There are some really good tutorials around
(especially the one by David Cottle (free "postcard"-ware, which means
you have to send the author an email to ask for the tutorial as he likes
to know where his tutorial is used and for what purposes), but also the
one by Mark Polishook (included in the help files)) to learn
SuperCollider and it has a kind of online-help system, as well as its
own emaillist and forum.
With the SCUM-addons, you can also make your own GUI's for SuperCollider
in a nice and easy way.
SC also runs with JACK, you can use MIDI and OSC to control it
externally (or control something else externally).
I like SC as it gives possibilities to dynamically change a lot of
things over time, and you can build really complex algorithms with it.
I am sure that in Pd a lot of the same things would be possible, but it
needs a quite different kind of programming strategy to do so.
On the other hand, I think that for synthesis techniques, either program
will do, and probably also some of the other ones available (CSound,
AlsaModularSynth, Galan, ... ? ).
Maybe the best is to take a look at a couple of tutorials of some
programs, then pick the one you like best, and delve further into that
one to really learn it (probably best if you try to make a piece with
it, as that is the best way to learn, I think).
and then, in the future, you can always go back and learn to use another
program, if you feel that you are limited in one way or another by the
one you had chosen before. After all, your musical way of working and
thinking may change too over time :)
sincerely,
Marije
8 Mar
8 Mar
6:30 a.m.
Hallo,
Marije Baalman hat gesagt: // Marije Baalman wrote:
I like SC as it gives possibilities to dynamically
change a lot of
things over time, and you can build really complex algorithms with it.
I am sure that in Pd a lot of the same things would be possible, but it
needs a quite different kind of programming strategy to do so.
Actually I think, those are the things that are a bit harder to do in
Pd than in a textual language as SC is. As Atte mentioned, for example
Pd doesn't do dynamic voice allocation. The design of Pd just works
differently in this regard: You have to code every voice you are going
to need. That's not hard, as there are language features to help with
doing that, but it's definitely not the automatic, batteries included
way of Csound or SC in this regard. Also if you intend to change the
structure of your synthesis engine while it's running, this probably
is better done in SC.
Pd however really shines when it comes to glueing stuff together. I
know of no other free multimedia software that supports such a wide range of
possibilities to connect to "the outside world".
For example: There's no need to use the Pd synthesis part at all, you
can just write your algorithms in SuperCollider and use Pd as a GUI
for SC. The advantage is, that Pd also is a RAD-GUI-Designer tool,
whereas in SC you need to first write a textual description of your
SCUM GUI, if I understand this correctly.
Then afterwards you can hook up a webcam to Pd, analyze its frames and
control SC in turn by using the webcam's input. Similarily maybe Pd's
"data structures" could be used instead of WONDER to design Wavefield
sound paths, which then can be visualized directly in Pd as OpenGL
graphics. The cool thing would be, that basically all of this uses the
same paradigm for programming.
So even if one ends up with and prefers a textual language instead of
a graphical patcher, that knowledge learn when using Pd is not in
vain and can be used to solve problems hard to solve with a text
editor.
Because of all this Pd with all its shortcomings is in such a wide use
and just pops up literally everywhere, even where you don't expect it
to be at all. For example at the "Girbaud fashion show" under the
beautiful feet of too skinny french women as proven here:
http://www.la-kitchen.fr/download/mfg
Ciao
--
Frank Barknecht _ ______footils.org__
_ __latest track: fqdn _ http://footils.org/cms/show/38
7 Mar
7 Mar
4 p.m.
Quoting Michael Wagner <mw(a)wasted.ch>ch>:
Thanks a lot for your responses and the input you all provided!
So, it seems, that the choice of the language to use is more a matter of taste
and there seem to be no real drawbacks between them.
Anyway, from what I have seen so far, I think, for starters, I'll give Pd a
try... In synthesis, graphical representation in my opinion really makes
sense...
Just one more question: How does it look with Jack support in Pd? IIRC I found
some old posts on mailinglist-archives about patches, but I didn't actually
find one...
-Michael
5:47 p.m.
Hallo,
Michael Wagner hat gesagt: // Michael Wagner wrote:
Just one more question: How does it look with Jack
support in Pd? IIRC I found
some old posts on mailinglist-archives about patches, but I didn't actually
find one...
Pd supports OSS, ALSA and Jack out of the box for audio. For midi, Pd
mainstream currently only uses rawmidi devices, but support for ALSA
sequencer is available as a patch in the Sourceforge bugtracker, it
already is in use e.g. in the Debian packages.
Ciao
--
Frank Barknecht _ ______footils.org__
_ __latest track: fqdn _ http://footils.org/cms/show/38
7201
days inactive
7203
days old
linux-audio-user@lists.linuxaudio.org
12 comments
8 participants
participants (8)
-
Atte André Jensen -
Dave Phillips -
Dave Robillard -
Frank Barknecht -
MarC -
Marije Baalman -
Michael Wagner -
mik