Hi all,

I saw this behavior on a machine where Pd couldn't open audio -- it just hangs. I hope to fix this. In the meantime, if you've got the same problem, just run "pd -nosound" and it should start up...

cheers Miller

On Mon, Apr 29, 2002 at 06:28:55PM +0100, Nick Fells wrote:

Hi,

I'm just trying to setup pd under osx. I've downloaded/unpacked Tk8.4 from the sourcefourge site 'snapshot', put the Frameworks under /Library, and put the Wish Shell.app under /Applications. But I can't get Pd to start the gui - there's no error message, just nothing happens.

Sorry if this is more a tk question than a pd one, but any pointers out there?

thanks,

nick

-- Dr Nick Fells, Music Department, University of Glasgow, Tel: +44 (0) 141-330 4096 Glasgow G12 8QQ Fax: +44 (0) 141-330 3518 UK. http://www.music.gla.ac.uk/~nick

Hello all,

I was having the same problem as Nick but running pd -nosound did not solve the problem. I get:

MIDI: not yet implimented in 2 out 2 rate 44100 using default device number -1 nchan 2, flags 3, bufs 8, framesperbuf 256

and then nothing, no GUI. I tested wish by itself and it works fine. (installed according to the install guide that tcl/tk for OSX came with)

running ./pd -nosound and ./pd both do the same thing. I suppose I should just try compiling?

Thanks Ben

On Tue, 30 Apr 2002, Nick Fells wrote:

Hi,

Yep, sure enough that works (but no sound, obviously...)... any fixes would be gladly welcomed!!.... (also obvious, I guess).

Another thing I noticed - in the menus under osx there's a 'tear off' option. If you happen to select it, all the menus appear to stop working...

thanks,

Nick

Miller Puckette wrote:

...

Hi all,

I saw this behavior on a machine where Pd couldn't open audio -- it just hangs. I hope to fix this. In the meantime, if you've got the same problem, just run "pd -nosound" and it should start up...

cheers Miller

On Mon, Apr 29, 2002 at 06:28:55PM +0100, Nick Fells wrote:

...

Hi,

I'm just trying to setup pd under osx. I've downloaded/unpacked Tk8.4 from the sourcefourge site 'snapshot', put the Frameworks under /Library, and put the Wish Shell.app under /Applications. But I can't get Pd to start the gui - there's no error message, just nothing happens.

-- Dr Nick Fells, Music Department, University of Glasgow, Tel: +44 (0) 141-330 4096 Glasgow G12 8QQ Fax: +44 (0) 141-330 3518 UK. http://www.music.gla.ac.uk/~nick

B. Bogart

I had this problem of the gui not starting a while back and tracked it down to the make process not modifying the first line of the pd.tk file correctly. Open pd.tk in a real text editor and make sure the first line reads: set pd_nt 2 once I changed this it worked fine. But the latest 0.35-test21 precompiled version works fine on my machine out of the tarball. I also trust everybody is using HFS+ filesystems? the unix filesystem doesn't work with tcl/tk for OSX yet.

On MacOS X.1.4, pd runs pretty well. I can save patchs, load externals, and I have compiled and run all of the third party externals (sonogram~ and the like) without problems and they work!!! The only thing is that I don't have any midi stuff to test the midi side with. Also, I have hacked the pd.tk file to make the GUI more Mac like so the menu bar is in the correct place and that command key modifiers work in a macintosh way. Doesn't seem to have messed up the X11 version either.

Anyway, great job miller!!!

m

On Tuesday, April 30, 2002, at 10:34 AM, Ben Bogart - FMPM/F1999 wrote:

Hello all,

I was having the same problem as Nick but running pd -nosound did not solve the problem. I get:

MIDI: not yet implimented in 2 out 2 rate 44100 using default device number -1 nchan 2, flags 3, bufs 8, framesperbuf 256

and then nothing, no GUI. I tested wish by itself and it works fine. (installed according to the install guide that tcl/tk for OSX came with)

running ./pd -nosound and ./pd both do the same thing. I suppose I should just try compiling?

Thanks Ben

On Tue, 30 Apr 2002, Nick Fells wrote:

...

Hi,

Yep, sure enough that works (but no sound, obviously...)... any fixes would be gladly welcomed!!.... (also obvious, I guess).

Another thing I noticed - in the menus under osx there's a 'tear off' option. If you happen to select it, all the menus appear to stop working...

thanks,

Nick

Miller Puckette wrote:

...

Hi all,

I saw this behavior on a machine where Pd couldn't open audio -- it just hangs. I hope to fix this. In the meantime, if you've got the same problem, just run "pd -nosound" and it should start up...

cheers Miller

On Mon, Apr 29, 2002 at 06:28:55PM +0100, Nick Fells wrote:

...

Hi,

I'm just trying to setup pd under osx. I've downloaded/unpacked Tk8.4 from the sourcefourge site 'snapshot', put the Frameworks under /Library, and put the Wish Shell.app under /Applications. But I can't get Pd to start the gui - there's no error message, just nothing happens.

-- Dr Nick Fells, Music Department, University of Glasgow, Tel: +44 (0) 141-330 4096 Glasgow G12 8QQ Fax: +44 (0) 141-330 3518 UK. http://www.music.gla.ac.uk/~nick

B. Bogart

I put them on my site with a link to your site... http://homepage.mac.com/mthomp/html/software.html

I think you will need to create aliases from the each external directory to the root level of the /extra directory for pd to find and load them correctly.

For the most part they complied straight up. I had to make minor changes to the source for others. The only change was that on BSD systems: #include <malloc.h> lives in #include <sys/malloc.h> and that was it... Oh, and the c2tk.bash scripts changed to use sh instead of bash as I dont have bash installed on my box. (I should like bash, its better than tcsh but I have never been able to learn to like it on the linux machines I used to use before MacOS X.) :-) Sometimes the scripts don't catch all of the excape " sequences. i.e. " should be replaced with " etc... but this is minor...

Michael

On Tuesday, April 30, 2002, at 01:38 PM, Yves Degoyon wrote:

hi,

...

and I have compiled and run all of the third party externals (sonogram~ and the like) without problems and they work!!!

great work too !!!

just tell me where they are available so that i can refer to them from my site.

i've got friends who'd like to use PD on OSX and it would be nice if i could provide them a link to these objects.

if you can't host them, contact me off-list in order to release them to the public.

cheers,

yves

Hi Thomas,

Is this with the test tone patch open? It's now new and "improved"... but it takes 10% of my 866 machine all by itself... without that, I see 4%.

cheers Miller

On Sat, May 04, 2002 at 02:13:46AM +0200, Thomas Musil wrote:

hi list, hi miller

on my old pentium/400MHz pd-0.34-4-linux consumed about 3 % cpu-power without any objects, pd-0.35test11 or test17 swallow 22 % of cpu-power. did anybody else watch this glutton-phenomenon?

thomas

Yes.. test17 eats about 7% cpu when idle on my althon xp 1800+.

Ben

On Sat, May 04, 2002 at 02:13:46AM +0200, Thomas Musil wrote:

hi list, hi miller

on my old pentium/400MHz pd-0.34-4-linux consumed about 3 % cpu-power without any objects, pd-0.35test11 or test17 swallow 22 % of cpu-power. did anybody else watch this glutton-phenomenon?

thomas

Hi all,

In RedHat 7.2, I tried Pd 0.35 test 22 with my delta66 (same chip as delta44), Alsa 0.9B10... I have an 866 MhZ P3 with Intel I815E chipset...

I get:

pd 4% pd -channels 4 5% pd -alsa < 1%!

This is both from "top" and from the "load meter" patch which seem to agree.

Typing just "pd" also gives me 4% CPU load in version 0.34-4..

just grasping at a straw, but perhaps for some reason the newest version isn't compiling with -O6 flag on your machines? Perhaps the configure script is acting funny...

cheers Miller

On Fri, May 03, 2002 at 10:26:54PM -0400, Bill Sack wrote:

yes, i notice something like this as well. a patch that runs robustly on my pII-350 in 0.34 is swamped in 0.35-17 (i wrote a mail about this on 4/24). i use delta44 card w/ alsa0.9b10 - what soundcard/driver are you using?

-bill

Thomas Musil wrote:

...

hi list, hi miller

on my old pentium/400MHz pd-0.34-4-linux consumed about 3 % cpu-power without any objects, pd-0.35test11 or test17 swallow 22 % of cpu-power. did anybody else watch this glutton-phenomenon?

thomas

-- _________ Bill Sack wsack@acsu.buffalo.edu

---

günter geiger hat gesagt: // günter geiger wrote:

On Sat, 4 May 2002, Miller Puckette wrote:

...

In RedHat 7.2, I tried Pd 0.35 test 22 with my delta66 (same chip as delta44), Alsa 0.9B10... I have an 866 MhZ P3 with Intel I815E chipset...

I get:

pd 4% pd -channels 4 5% pd -alsa < 1%!

This means pd works multichannel with alsa ? is there a special alsa version required, I couldnt get it running with alsa0.9B10 and the rme card.

This probably is the old RME problem of PD with ALSA, that I assume lies in either PD's or ALSA's handling of the plughw layer in the combination ALSA-RME. (0.9.0beta10, here as well, I should upgrade)

The ICE chipset on the Delta and on my Audiophile has a similar problem: It can only do a certain combination of parameters in hardware, and the usual CD file format doesn't work in hardware without translation to the hardware format:

$ aplay -D hw:0,0 ~/loops/hmmbaby.wav
Playing WAVE '/home/freak/loops/hmmbaby.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo aplay: set_params:739: Sample format non available

The plugin layer of ALSA does this translation: $ aplay -D plughw:0,0 ~/loops/hmmbaby.wav Playing WAVE '/home/freak/loops/hmmbaby.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo

In PD the plugin layer doesn't work at all here, only the hw:layer $ pd -alsadev hw:0,0 [Midiman: OK] $ pd -alsadev hw:1,0 [SBLive: OK] $ pd -alsadev plughw:0,0 [Midiman: audio stuck] $ pd -alsadev plughw:1,0 [SBLive: audio stuck]

A strange thing is, that even using the hw:0,0 device, I have a [samplerate~] of 44100. I don't know, what the samplerate at the chipset is, though...

Unfortunatly I have no idea who to blame for these symptoms: ALSA, PD, both???

ciao,

Frank Barknecht _ _______footils__

hello again,

the mystery continues ... when i start pd-0.35 with no midi (i.e.: not specifying my usual midi device at -midiin[out]dev 2) then cpu-load is comparable to pd-0.34 (0-3%) ... starting with midi it goes to ~30% even before starting dac

midi interface is an old MPU-401 card using alsa mpu401 driver ... no problems in pd-0.34

hmm

-bill Miller Puckette wrote:

Hi all,

In RedHat 7.2, I tried Pd 0.35 test 22 with my delta66 (same chip as delta44), Alsa 0.9B10... I have an 866 MhZ P3 with Intel I815E chipset...

I get:

pd 4% pd -channels 4 5% pd -alsa < 1%!

This is both from "top" and from the "load meter" patch which seem to agree.

Typing just "pd" also gives me 4% CPU load in version 0.34-4..

just grasping at a straw, but perhaps for some reason the newest version isn't compiling with -O6 flag on your machines? Perhaps the configure script is acting funny...

cheers Miller

On Fri, May 03, 2002 at 10:26:54PM -0400, Bill Sack wrote:

...

yes, i notice something like this as well. a patch that runs robustly on my pII-350 in 0.34 is swamped in 0.35-17 (i wrote a mail about this on 4/24). i use delta44 card w/ alsa0.9b10 - what soundcard/driver are you using?

-bill

Thomas Musil wrote:

...

hi list, hi miller

on my old pentium/400MHz pd-0.34-4-linux consumed about 3 % cpu-power without any objects, pd-0.35test11 or test17 swallow 22 % of cpu-power. did anybody else watch this glutton-phenomenon?

thomas

-- _________ Bill Sack wsack@acsu.buffalo.edu

---

Thomas Musil schrieb:

hi list

did anybody watch this phenomenon:

Yes, with the cverb~ external.... After turning off the input signal CPU usage jumped up to 80 % or more (for one object!) after a second or so. This is due to the architecture of the CPU that switches to (slow) floating point mode. To prevent this just do the following

if(((*(unsigned int*)&sample)&0x7f800000)==0) sample=0.0

whereever a signal gets smaller without realy getting zeroe (like in the reverb tail of cverb~, freeverb~ and the like).

Olaf

PS: I think there is a document at htt://musicdsp.org describing more / other methods of undenormalisation

the CPU is a PII 800 MHz; I tried under w2k and linux; a patch with some sample-loops and LFOs consumes 50% cpu-power. adding some (10 .. 20) filters of iemlib (like hp1~) it eats up to 100 % (after 5 minutes). in my test-patches, one filter takes about 0.1 to 0.2 %. I figured out, if there is no signal in, the IIR-filters produce NaNs. I think, Miller wrote in his sources of "d_filter.c" a NAN-protection. but it did not work very well for me. I added to each filter a white-noise-signal multiplied by "*~ 1e-10" (-200 dB), now the filters are cpu-power-stable. (I generate only once, and then i sended it with s~ and r~)

Thomas Musil