I am working on a sound installation in an old train tunnel and I am using PD to control everything. But since I am a PD neophyte, I want to ask the list for direction. Basically, my plan is to use about 8 computers spaced down the length of the tunnel, each with two speakers. They will be networked together and running either OS X or linux.
Here is how I envision it:
(L) (R) (L) (R) (L) (R) (L) (R) (L) (R) [speakers] \ / \ / \ / \ / \ / master slave slave slave slave |---------| |-------| |-------| |-------| |---... [ethernet]
All of the computers will share a pool of samples via the network. I would like to control all of the PDs from the master computer.
For the first aspect of the installation, I would like to be able to trigger any sample on any/all computers and even move samples down the tunnel. The second aspect, I would like to be able to control granular synthesis on these computers.
Does anybody know anyone who has done something like this before? With PD? Any advice you want to offer to get me going in the right direction would be grreatly appreciated. I am just starting with PD, but am familiar with most of the concepts since I have a background in Csound and programming.
.hc
(Someone please tell me if this is post all wrong and terribly uninformed...I'm new to this list.)
Hans-Christoph,
In terms of networking the many computers together and getting all your PD patches to communicate, I think you could use Open Sound Control, or perhaps the netsend and netreceive objects (native PD objects).
I found an OSC (open sound control) object and reference at http://barely.a.live.fm/pd/OSC/
(Don the speed of your computer and the intensity of your installation, you could consider running all the audio outputs from a single computer -- with a multi-output soundcard. For example I've seen Digidesign hardware or Sonorus hardware with 8 analog outputs. HOWEVER, I don't know whether PD supports more than 2 audio outs yet ??(can somebody else answer this?)-- but I know SuperCollider and MAX for the Macintosh do.)
Dave Sabine dave@davesabine.com http://www.davesabine.com/
----- Original Message ----- From: "Hans-Christoph Steiner" hans@eds.org To: pd-list@iem.kug.ac.at Sent: Tuesday, March 05, 2002 9:16 AM Subject: [PD] networked multi-channel PD cluster
I am working on a sound installation in an old train tunnel and I am using PD to control everything. But since I am a PD neophyte, I want to ask the list for direction. Basically, my plan is to use about 8 computers spaced down the length of the tunnel, each with two speakers. They will be networked together and running either OS X or linux.
Here is how I envision it:
(L) (R) (L) (R) (L) (R) (L) (R) (L) (R) [speakers] \ / \ / \ / \ / \ / master slave slave slave slave |---------| |-------| |-------| |-------| |---... [ethernet]
All of the computers will share a pool of samples via the network. I would like to control all of the PDs from the master computer.
For the first aspect of the installation, I would like to be able to trigger any sample on any/all computers and even move samples down the tunnel. The second aspect, I would like to be able to control granular synthesis on these computers.
Does anybody know anyone who has done something like this before? With PD? Any advice you want to offer to get me going in the right direction would be grreatly appreciated. I am just starting with PD, but am familiar with most of the concepts since I have a background in Csound and programming.
.hc
--
"Man has survived hitherto because he was to ignorant to know how to
realize
his wishes. Now that he can realize them, he must either change them or perish."
-William Carlos Williams
Hans-Christoph Steiner http://works.steinertours.com
I am planning on using netsend/netreceive since they seem to do what I need them to and they are already part of PD. OSC would be best in the long run, but it seems like its a bit beta still. I plan on using multiple computers because I want to space the speakers every 75 meters or so along the 500 m tunnel. Running audio cables 500 m isn't really feasible, but ethernet can do 100m cable lengths. I'll have a hub or router every 100m. (FYI: PD does support multiple audio outs).
Also, currently I am not too concerned with latency since its probably going to be a prerecorded bit played back, rather than a live performace. That can come later. I might choose OS X because I am getting a bunch of Macs for this project that I might want to use for other things in the future. Linux is an option too. I am not concerned with the OS really since PD works for me on both platforms.
Has anyone seen anything like this implemented before? I am mainly looking for examples of spacialization in such a setup, plus how to control it all once its setup.
.hc
David Sabine wrote:
(Someone please tell me if this is post all wrong and terribly uninformed...I'm new to this list.)
Hans-Christoph,
In terms of networking the many computers together and getting all your PD patches to communicate, I think you could use Open Sound Control, or perhaps the netsend and netreceive objects (native PD objects).
I found an OSC (open sound control) object and reference at http://barely.a.live.fm/pd/OSC/
(Don the speed of your computer and the intensity of your installation, you could consider running all the audio outputs from a single computer -- with a multi-output soundcard. For example I've seen Digidesign hardware or Sonorus hardware with 8 analog outputs. HOWEVER, I don't know whether PD supports more than 2 audio outs yet ??(can somebody else answer this?)-- but I know SuperCollider and MAX for the Macintosh do.)
Dave Sabine dave@davesabine.com http://www.davesabine.com/
----- Original Message ----- From: "Hans-Christoph Steiner" hans@eds.org To: pd-list@iem.kug.ac.at Sent: Tuesday, March 05, 2002 9:16 AM Subject: [PD] networked multi-channel PD cluster
I am working on a sound installation in an old train tunnel and I am using PD to control everything. But since I am a PD neophyte, I want to ask the list for direction. Basically, my plan is to use about 8 computers spaced down the length of the tunnel, each with two speakers. They will be networked together and running either OS X or linux.
Here is how I envision it:
(L) (R) (L) (R) (L) (R) (L) (R) (L) (R) [speakers] \ / \ / \ / \ / \ / master slave slave slave slave |---------| |-------| |-------| |-------| |---... [ethernet]
All of the computers will share a pool of samples via the network. I would like to control all of the PDs from the master computer.
For the first aspect of the installation, I would like to be able to trigger any sample on any/all computers and even move samples down the tunnel. The second aspect, I would like to be able to control granular synthesis on these computers.
Does anybody know anyone who has done something like this before? With PD? Any advice you want to offer to get me going in the right direction would be grreatly appreciated. I am just starting with PD, but am familiar with most of the concepts since I have a background in Csound and programming.
.hc
--
"Man has survived hitherto because he was to ignorant to know how to
realize
his wishes. Now that he can realize them, he must either change them or perish."
-William Carlos Williams
Hans-Christoph Steiner http://works.steinertours.com
Have you actually experimented with PD on OS-X yet? I'm not sure how stable it is right now, I think its only a test build. Can anyone share their exeperiences?
-Josh
(and yes, Hans is my brother)
On Wed, 6 Mar 2002, Hans-Christoph Steiner wrote:
I am planning on using netsend/netreceive since they seem to do what I need them to and they are already part of PD. OSC would be best in the long run, but it seems like its a bit beta still. I plan on using multiple computers because I want to space the speakers every 75 meters or so along the 500 m tunnel. Running audio cables 500 m isn't really feasible, but ethernet can do 100m cable lengths. I'll have a hub or router every 100m. (FYI: PD does support multiple audio outs).
Also, currently I am not too concerned with latency since its probably going to be a prerecorded bit played back, rather than a live performace. That can come later. I might choose OS X because I am getting a bunch of Macs for this project that I might want to use for other things in the future. Linux is an option too. I am not concerned with the OS really since PD works for me on both platforms.
Has anyone seen anything like this implemented before? I am mainly looking for examples of spacialization in such a setup, plus how to control it all once its setup.
.hc
David Sabine wrote:
(Someone please tell me if this is post all wrong and terribly uninformed...I'm new to this list.)
Hans-Christoph,
In terms of networking the many computers together and getting all your PD patches to communicate, I think you could use Open Sound Control, or perhaps the netsend and netreceive objects (native PD objects).
I found an OSC (open sound control) object and reference at http://barely.a.live.fm/pd/OSC/
(Don the speed of your computer and the intensity of your installation, you could consider running all the audio outputs from a single computer -- with a multi-output soundcard. For example I've seen Digidesign hardware or Sonorus hardware with 8 analog outputs. HOWEVER, I don't know whether PD supports more than 2 audio outs yet ??(can somebody else answer this?)-- but I know SuperCollider and MAX for the Macintosh do.)
Dave Sabine dave@davesabine.com http://www.davesabine.com/
----- Original Message ----- From: "Hans-Christoph Steiner" hans@eds.org To: pd-list@iem.kug.ac.at Sent: Tuesday, March 05, 2002 9:16 AM Subject: [PD] networked multi-channel PD cluster
I am working on a sound installation in an old train tunnel and I am using PD to control everything. But since I am a PD neophyte, I want to ask the list for direction. Basically, my plan is to use about 8 computers spaced down the length of the tunnel, each with two speakers. They will be networked together and running either OS X or linux.
Here is how I envision it:
(L) (R) (L) (R) (L) (R) (L) (R) (L) (R) [speakers] \ / \ / \ / \ / \ / master slave slave slave slave |---------| |-------| |-------| |-------| |---... [ethernet]
All of the computers will share a pool of samples via the network. I would like to control all of the PDs from the master computer.
For the first aspect of the installation, I would like to be able to trigger any sample on any/all computers and even move samples down the tunnel. The second aspect, I would like to be able to control granular synthesis on these computers.
Does anybody know anyone who has done something like this before? With PD? Any advice you want to offer to get me going in the right direction would be grreatly appreciated. I am just starting with PD, but am familiar with most of the concepts since I have a background in Csound and programming.
.hc
--
"Man has survived hitherto because he was to ignorant to know how to
realize
his wishes. Now that he can realize them, he must either change them or perish."
-William Carlos Williams
Hans-Christoph Steiner http://works.steinertours.com
--
"Man has survived hitherto because he was to ignorant to know how to realize his wishes. Now that he can realize them, he must either change them or perish."
-William Carlos Williams
Hans-Christoph Steiner http://works.steinertours.com
Josh .. Yoshi .. Joschi .. xiphoidprocess.com .. eds.org/~joschi
I've played with it on OS X. I was getting sound, controlling it with netsend/netreceive. And it worked fine. But it was hardly a definitive test.
.hc
Josh Steiner wrote:
Have you actually experimented with PD on OS-X yet? I'm not sure how stable it is right now, I think its only a test build. Can anyone share their exeperiences?
-Josh
(and yes, Hans is my brother)
On Wed, 6 Mar 2002, Hans-Christoph Steiner wrote:
I am planning on using netsend/netreceive since they seem to do what I need them to and they are already part of PD. OSC would be best in the long run, but it seems like its a bit beta still. I plan on using multiple computers because I want to space the speakers every 75 meters or so along the 500 m tunnel. Running audio cables 500 m isn't really feasible, but ethernet can do 100m cable lengths. I'll have a hub or router every 100m. (FYI: PD does support multiple audio outs).
Also, currently I am not too concerned with latency since its probably going to be a prerecorded bit played back, rather than a live performace. That can come later. I might choose OS X because I am getting a bunch of Macs for this project that I might want to use for other things in the future. Linux is an option too. I am not concerned with the OS really since PD works for me on both platforms.
Has anyone seen anything like this implemented before? I am mainly looking for examples of spacialization in such a setup, plus how to control it all once its setup.
.hc
David Sabine wrote:
(Someone please tell me if this is post all wrong and terribly uninformed...I'm new to this list.)
Hans-Christoph,
In terms of networking the many computers together and getting all your PD patches to communicate, I think you could use Open Sound Control, or perhaps the netsend and netreceive objects (native PD objects).
I found an OSC (open sound control) object and reference at http://barely.a.live.fm/pd/OSC/
(Don the speed of your computer and the intensity of your installation, you could consider running all the audio outputs from a single computer -- with a multi-output soundcard. For example I've seen Digidesign hardware or Sonorus hardware with 8 analog outputs. HOWEVER, I don't know whether PD supports more than 2 audio outs yet ??(can somebody else answer this?)-- but I know SuperCollider and MAX for the Macintosh do.)
Dave Sabine dave@davesabine.com http://www.davesabine.com/
----- Original Message ----- From: "Hans-Christoph Steiner" hans@eds.org To: pd-list@iem.kug.ac.at Sent: Tuesday, March 05, 2002 9:16 AM Subject: [PD] networked multi-channel PD cluster
I am working on a sound installation in an old train tunnel and I am using PD to control everything. But since I am a PD neophyte, I want to ask the list for direction. Basically, my plan is to use about 8 computers spaced down the length of the tunnel, each with two speakers. They will be networked together and running either OS X or linux.
Here is how I envision it:
(L) (R) (L) (R) (L) (R) (L) (R) (L) (R) [speakers] \ / \ / \ / \ / \ / master slave slave slave slave |---------| |-------| |-------| |-------| |---... [ethernet]
All of the computers will share a pool of samples via the network. I would like to control all of the PDs from the master computer.
For the first aspect of the installation, I would like to be able to trigger any sample on any/all computers and even move samples down the tunnel. The second aspect, I would like to be able to control granular synthesis on these computers.
Does anybody know anyone who has done something like this before? With PD? Any advice you want to offer to get me going in the right direction would be grreatly appreciated. I am just starting with PD, but am familiar with most of the concepts since I have a background in Csound and programming.
.hc
--
"Man has survived hitherto because he was to ignorant to know how to
realize
his wishes. Now that he can realize them, he must either change them or perish."
-William Carlos Williams
Hans-Christoph Steiner http://works.steinertours.com
--
"Man has survived hitherto because he was to ignorant to know how to realize his wishes. Now that he can realize them, he must either change them or perish."
-William Carlos Williams
Hans-Christoph Steiner http://works.steinertours.com
Josh .. Yoshi .. Joschi .. xiphoidprocess.com .. eds.org/~joschi
-
David Sabine -
Hans-Christoph Steiner -
Josh Steiner