hi alls,
i'm investigating possibilities for a general purpose audio hardware device based on a DSP chip.
i know about Pd-anywhere and Hans-Christoph Steiner's efforts to get Pd to run on older devices. this is not my goal here as i'd like to avoid having to load an entire operating system, if possible; i really just need the DSP code and a way of running it on a DSP chip.
is there any way of getting Pd to compile patches internally to some kind of DSP machine code, for example for something like a TI C5x?
if not, does anyone know of anything vaguely similar that might get me partway along to this goal? it doesn't have to load Pd patches necessarily, but there are obvious benefits to being able to do that :-)
cheers damian -- damian stewart . @damian0815 . damian@frey.co.nz frey . contemporary art . http://www.frey.co.nz
Hey Damian,
The gluiph was just that, it ran Pd directly on a DSP: http://www.nime.org/proceedings/2003/nime2003_180.pdf
Depending on your skills, it could be easier to run a über-stripped OS like I did on the Palm Pilots, which ran Pd tho they had 32megs of RAM. There wasn't much more than the linux kernel, a super basic X11 server, and Pd.
.hc
On Apr 2, 2012, at 8:04 AM, Damian Stewart wrote:
hi alls,
i'm investigating possibilities for a general purpose audio hardware device based on a DSP chip.
i know about Pd-anywhere and Hans-Christoph Steiner's efforts to get Pd to run on older devices. this is not my goal here as i'd like to avoid having to load an entire operating system, if possible; i really just need the DSP code and a way of running it on a DSP chip.
is there any way of getting Pd to compile patches internally to some kind of DSP machine code, for example for something like a TI C5x?
if not, does anyone know of anything vaguely similar that might get me partway along to this goal? it doesn't have to load Pd patches necessarily, but there are obvious benefits to being able to do that :-)
cheers damian -- damian stewart . @damian0815 . damian@frey.co.nz frey . contemporary art . http://www.frey.co.nz
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
----------------------------------------------------------------------------
"We have nothing to fear from love and commitment." - New York Senator Diane Savino, trying to convince the NY Senate to pass a gay marriage bill
It should be possible but Pd needs stuff like open(), read(), and write() for files, so it's necessary to make a small library to either carry out or somehow fake those operations.
The great advantage of running Pd on a DSP is that you can probably get audio latencies down much further than on a PC-like CPU.
cheers Miller On Mon, Apr 02, 2012 at 10:58:49AM -0400, Hans-Christoph Steiner wrote:
Hey Damian,
The gluiph was just that, it ran Pd directly on a DSP: http://www.nime.org/proceedings/2003/nime2003_180.pdf
Depending on your skills, it could be easier to run a über-stripped OS like I did on the Palm Pilots, which ran Pd tho they had 32megs of RAM. There wasn't much more than the linux kernel, a super basic X11 server, and Pd.
.hc
On Apr 2, 2012, at 8:04 AM, Damian Stewart wrote:
hi alls,
i'm investigating possibilities for a general purpose audio hardware device based on a DSP chip.
i know about Pd-anywhere and Hans-Christoph Steiner's efforts to get Pd to run on older devices. this is not my goal here as i'd like to avoid having to load an entire operating system, if possible; i really just need the DSP code and a way of running it on a DSP chip.
is there any way of getting Pd to compile patches internally to some kind of DSP machine code, for example for something like a TI C5x?
if not, does anyone know of anything vaguely similar that might get me partway along to this goal? it doesn't have to load Pd patches necessarily, but there are obvious benefits to being able to do that :-)
cheers damian -- damian stewart . @damian0815 . damian@frey.co.nz frey . contemporary art . http://www.frey.co.nz
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
"We have nothing to fear from love and commitment." - New York Senator Diane Savino, trying to convince the NY Senate to pass a gay marriage bill
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
Yeah, I imagine with no kernel doing timeslicing, Pd would literally have all of the CPU time, and the audio output to the hardware happen more often than every 64 samples.
.hc
On Apr 2, 2012, at 11:31 AM, Miller Puckette wrote:
It should be possible but Pd needs stuff like open(), read(), and write() for files, so it's necessary to make a small library to either carry out or somehow fake those operations.
The great advantage of running Pd on a DSP is that you can probably get audio latencies down much further than on a PC-like CPU.
cheers Miller On Mon, Apr 02, 2012 at 10:58:49AM -0400, Hans-Christoph Steiner wrote:
Hey Damian,
The gluiph was just that, it ran Pd directly on a DSP: http://www.nime.org/proceedings/2003/nime2003_180.pdf
Depending on your skills, it could be easier to run a über-stripped OS like I did on the Palm Pilots, which ran Pd tho they had 32megs of RAM. There wasn't much more than the linux kernel, a super basic X11 server, and Pd.
.hc
On Apr 2, 2012, at 8:04 AM, Damian Stewart wrote:
hi alls,
i'm investigating possibilities for a general purpose audio hardware device based on a DSP chip.
i know about Pd-anywhere and Hans-Christoph Steiner's efforts to get Pd to run on older devices. this is not my goal here as i'd like to avoid having to load an entire operating system, if possible; i really just need the DSP code and a way of running it on a DSP chip.
is there any way of getting Pd to compile patches internally to some kind of DSP machine code, for example for something like a TI C5x?
if not, does anyone know of anything vaguely similar that might get me partway along to this goal? it doesn't have to load Pd patches necessarily, but there are obvious benefits to being able to do that :-)
cheers damian -- damian stewart . @damian0815 . damian@frey.co.nz frey . contemporary art . http://www.frey.co.nz
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
"We have nothing to fear from love and commitment." - New York Senator Diane Savino, trying to convince the NY Senate to pass a gay marriage bill
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
----------------------------------------------------------------------------
The arc of history bends towards justice. - Dr. Martin Luther King, Jr.
hey Hans + Miller,
thanks for pointing me towards Sukandar's work.
so to be clear: you think i'd be able to take the Pd source code and just push it through the DSP's compiler and something vaguely usable would come out? (I have never done anything with a DSP chip before and the examples on TI's website are fairly heavy on the assembler, they mostly look like microprocessor bit-bangy code..)
cheers damian
On 02.04.2012, at 19:37, Hans-Christoph Steiner wrote:
Yeah, I imagine with no kernel doing timeslicing, Pd would literally have all of the CPU time, and the audio output to the hardware happen more often than every 64 samples.
.hc
On Apr 2, 2012, at 11:31 AM, Miller Puckette wrote:
It should be possible but Pd needs stuff like open(), read(), and write() for files, so it's necessary to make a small library to either carry out or somehow fake those operations.
The great advantage of running Pd on a DSP is that you can probably get audio latencies down much further than on a PC-like CPU.
cheers Miller On Mon, Apr 02, 2012 at 10:58:49AM -0400, Hans-Christoph Steiner wrote:
Hey Damian,
The gluiph was just that, it ran Pd directly on a DSP: http://www.nime.org/proceedings/2003/nime2003_180.pdf
Depending on your skills, it could be easier to run a über-stripped OS like I did on the Palm Pilots, which ran Pd tho they had 32megs of RAM. There wasn't much more than the linux kernel, a super basic X11 server, and Pd.
.hc
On Apr 2, 2012, at 8:04 AM, Damian Stewart wrote:
hi alls,
i'm investigating possibilities for a general purpose audio hardware device based on a DSP chip.
i know about Pd-anywhere and Hans-Christoph Steiner's efforts to get Pd to run on older devices. this is not my goal here as i'd like to avoid having to load an entire operating system, if possible; i really just need the DSP code and a way of running it on a DSP chip.
is there any way of getting Pd to compile patches internally to some kind of DSP machine code, for example for something like a TI C5x?
if not, does anyone know of anything vaguely similar that might get me partway along to this goal? it doesn't have to load Pd patches necessarily, but there are obvious benefits to being able to do that :-)
cheers damian -- damian stewart . @damian0815 . damian@frey.co.nz frey . contemporary art . http://www.frey.co.nz
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
"We have nothing to fear from love and commitment." - New York Senator Diane Savino, trying to convince the NY Senate to pass a gay marriage bill
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
The arc of history bends towards justice. - Dr. Martin Luther King, Jr.
I think you'll want to start with libpd, since it strips out a lot of the GUI stuff, which definitely won't be needed ;). Then you'll need to get it compiling, then you'll need to write the audio I/O method.
You'll need to put in some kind of hack to replace loading a patch, like having the patch as a compiled-in string.
.hc
On Apr 3, 2012, at 5:10 PM, Damian Stewart (ML) wrote:
hey Hans + Miller,
thanks for pointing me towards Sukandar's work.
so to be clear: you think i'd be able to take the Pd source code and just push it through the DSP's compiler and something vaguely usable would come out? (I have never done anything with a DSP chip before and the examples on TI's website are fairly heavy on the assembler, they mostly look like microprocessor bit-bangy code..)
cheers damian
On 02.04.2012, at 19:37, Hans-Christoph Steiner wrote:
Yeah, I imagine with no kernel doing timeslicing, Pd would literally have all of the CPU time, and the audio output to the hardware happen more often than every 64 samples.
.hc
On Apr 2, 2012, at 11:31 AM, Miller Puckette wrote:
It should be possible but Pd needs stuff like open(), read(), and write() for files, so it's necessary to make a small library to either carry out or somehow fake those operations.
The great advantage of running Pd on a DSP is that you can probably get audio latencies down much further than on a PC-like CPU.
cheers Miller On Mon, Apr 02, 2012 at 10:58:49AM -0400, Hans-Christoph Steiner wrote:
Hey Damian,
The gluiph was just that, it ran Pd directly on a DSP: http://www.nime.org/proceedings/2003/nime2003_180.pdf
Depending on your skills, it could be easier to run a über-stripped OS like I did on the Palm Pilots, which ran Pd tho they had 32megs of RAM. There wasn't much more than the linux kernel, a super basic X11 server, and Pd.
.hc
On Apr 2, 2012, at 8:04 AM, Damian Stewart wrote:
hi alls,
i'm investigating possibilities for a general purpose audio hardware device based on a DSP chip.
i know about Pd-anywhere and Hans-Christoph Steiner's efforts to get Pd to run on older devices. this is not my goal here as i'd like to avoid having to load an entire operating system, if possible; i really just need the DSP code and a way of running it on a DSP chip.
is there any way of getting Pd to compile patches internally to some kind of DSP machine code, for example for something like a TI C5x?
if not, does anyone know of anything vaguely similar that might get me partway along to this goal? it doesn't have to load Pd patches necessarily, but there are obvious benefits to being able to do that :-)
cheers damian -- damian stewart . @damian0815 . damian@frey.co.nz frey . contemporary art . http://www.frey.co.nz
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
"We have nothing to fear from love and commitment." - New York Senator Diane Savino, trying to convince the NY Senate to pass a gay marriage bill
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
The arc of history bends towards justice. - Dr. Martin Luther King, Jr.
Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev
----------------------------------------------------------------------------
As we enjoy great advantages from inventions of others, we should be glad of an opportunity to serve others by any invention of ours; and this we should do freely and generously. - Benjamin Franklin
-
Damian Stewart -
Damian Stewart (ML)
-
Hans-Christoph Steiner -
Miller Puckette