Re: [PD] Maximum playback/capture with jack

2010-08-30 Thread IOhannes m zmoelnig
On 2010-08-29 11:58, Christoph Kuhr wrote:
 Hi everyone,
 
 I have a problem using jack audio...
 
 I want puredata to register 48channels of capture and 8 of playback.
 But i get only 18channels running.
 Else pd says:
 error: JACK: unable to connect to JACK server
 JACK: jack returned status 17
 error: JACK error: cannot deliver port registration request
 error:JACK: can only register 18 input ports (instead of requested 48)
 
 Is pd limited to 26 channels IO?
 

depends on which version of Pd you are using (which you happen to have
properly concealed from us).
older versions of Pd had an insanely low number of maximum jackports
(though i'm pretty sure it was a power-of-two, and not a number like
18 or 26)

anyhow, with recent Pd versions i have been able to use Pd fully
connected to a MADI-card, so at least 64 channels IO.

fgmasdr
IOhannes



smime.p7s
Description: S/MIME Cryptographic Signature
___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] OT: Ir leds

2010-08-30 Thread Derek Holzer
You probably want the brightest ones (usually measured in in mcd) you 
can get! Besides that, if you know anything about the camera you are 
using that would help a lot, for example which part of the spectrum in 
nanometers it is most sensitive to. Then you can match it to the 
spectrum emitted by the LEDs.


Another characteristic is the angle of the LED, in your case you 
probably want them to be pretty wide.


Some LEDs are low current or ultrabright, which mean they are better at 
making light from current. Ultrabrights, however use a ton of current at 
their brightest. This is usually only a concern if you are running from 
a battery or very weak power supply (or in something like a voltage 
controlled synthesizer when a sudden big current draw can have audible 
consequences!).


So once you understand the characteristics of the LEDs, you can choose 
the ones for your budget. You'll still have to do a lot of testing to 
get the right amount of light for the space and your camera. It's always 
of a question of whether you have more money than time, or time than money.


Somehow I seem to have neither these days, I wonder what I'm doing 
wrong! ;-)


Best,
Derek

On 8/30/10 4:44 AM, ronni montoya wrote:

Hello everybody, i dont know if here is the correct place to ask, if
anybody know another forum related to this , i will be very happy if
somebody can show me:

Im making an interactive floor and i need to iluminate a space of 4 x
4 meters with ir light.
I been seen there are some ir leds matrix on survillance stores but
they are expensive to my budget, so i was thinking that i can build my
own ir led matrix .
I been seen there are tons of different types of ir leds with
different characteristics.

Do anybody know which ir leds do i need for this?

which model, label or serial number?


Any idea would be appreaciated

thank you very much

Ronny

___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -  
http://lists.puredata.info/listinfo/pd-list



--
::: derek holzer ::: http://macumbista.net :::
---Oblique Strategy # 183:
When is it for?

___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] abstraction precedence

2010-08-30 Thread Frank Barknecht
On Fri, Aug 27, 2010 at 11:51:52AM +0200, cyrille henry wrote:
 create a abs folder in the patch folder
 put your abstraction in it
 create [abs/abstraction_name] object

Technically it's the same as renaming the files, so Matteo would still need to
find/replace how objects.

Ciao
-- 
Frank

___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] abstraction precedence

2010-08-30 Thread Frank Barknecht
Hi,

On Fri, Aug 27, 2010 at 07:36:50PM +0200, Matteo Sisti Sette wrote:
 On 08/27/2010 06:39 PM, cyrille henry wrote:
 for now, extern have priority over abstractions.

 Ok thanks

 but replacing [abstraction_name] to [abs/abstraction_name] can be made
 in few sec with every text editor.

 Yes of course, but I have to do it for every abstraction_name.

 However I now realise that I can get the list of abstraction names from  
 the directory listing.

 Also I didn't know sed, thank you for the suggestion.

Check this:
http://puredata.info/docs/tutorials/TipsAndTricks#replacing-a-lot-of-objects-in-a-lot-of-pd-patches

Ciao
-- 
 Frank BarknechtDo You RjDj.me?  _ __footils.org__

___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] OT: Ir leds

2010-08-30 Thread Gerard Paresys

I illuminated an installation with 18 IR diodes OPE5685.
http://radiospares-fr.rs-online.com/web/4550802.html

For an Unibrain camera ref 2061:
 http://www.unibrain.com/Products/VisionImg/Fire_i_BC_remoteCCD.htm
with 2.1mm Wide Lens (no IR coating)

Gerard

___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] OT: Ir leds

2010-08-30 Thread Gerard Paresys

only 18 leds? wow , can you tell me please which large was the  space
you iluminated? how many meters does it have?


A circle of approximately diameter 2 meter.

The 18 LEDs are placed at a distance of approximately 1 to 2 meters.

http://gerard.paresys.free.fr/Installation/index.html

Gerard



thanks!!!

2010/8/30 Gerard Paresys gerard.pare...@ens.fr:

 I illuminated an installation with 18 IR diodes OPE5685.
 http://radiospares-fr.rs-online.com/web/4550802.html

 For an Unibrain camera ref 2061:
  http://www.unibrain.com/Products/VisionImg/Fire_i_BC_remoteCCD.htm
 with 2.1mm Wide Lens (no IR coating)

 Gerard




___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] live pd patch mixed with Gem or pdp?

2010-08-30 Thread IOhannes m zmoelnig
On 2010-08-30 17:04, John Harrison wrote:
 I'd like to take a patch from a pd window and mix it live with a Gem or pdp
 window. pdp_capture would have been the ticket, but I can't get it to
 compile and looking at the archives of this list, it doesn't seem like
 others have had luck with it either. Any other ideas for me?


a bit complicated, but you can do that by doing screenshots with
gstreamer, send it to a videoloopback device and read that (with pdp,
Gem, GridFlow or whatelse)

- get v4l2loopback (http://code.google.com/p/v4l2loopback/)
 compile, install, load the module
- get gst-v4l2loopback (http://github.com/umlaeute/gst-v4l2loopback)
 compile (need gst-plugin-base-dev), install

run the gstreamer-pipeline, e.g.
$ gst-launch ximagesrc use-damage=false startx=0 starty=0 endx=640
endy=480 ! queue ! ffmpegcolorspace ! v4l2loopback

access the video using [pix_video], [pdp_v4l2] [#camera] or cat


running here.
this of course only grabs a portion of the screen, rather than a window
and track that...

fgmadfs
IOhannes



smime.p7s
Description: S/MIME Cryptographic Signature
___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] live pd patch mixed with Gem or pdp?

2010-08-30 Thread Pagano, Patrick
Depends on if it needs to be one machine as well.
Can easily be done with two machines and a scan converter

pp

-Original Message-
From: pd-list-boun...@iem.at [mailto:pd-list-boun...@iem.at] On Behalf Of 
IOhannes m zmoelnig
Sent: Monday, August 30, 2010 11:44 AM
To: John Harrison
Cc: pd-list@iem.at
Subject: Re: [PD] live pd patch mixed with Gem or pdp?

On 2010-08-30 17:04, John Harrison wrote:
 I'd like to take a patch from a pd window and mix it live with a Gem 
 or pdp window. pdp_capture would have been the ticket, but I can't get 
 it to compile and looking at the archives of this list, it doesn't 
 seem like others have had luck with it either. Any other ideas for me?


a bit complicated, but you can do that by doing screenshots with gstreamer, 
send it to a videoloopback device and read that (with pdp, Gem, GridFlow or 
whatelse)

- get v4l2loopback (http://code.google.com/p/v4l2loopback/)
 compile, install, load the module
- get gst-v4l2loopback (http://github.com/umlaeute/gst-v4l2loopback)
 compile (need gst-plugin-base-dev), install

run the gstreamer-pipeline, e.g.
$ gst-launch ximagesrc use-damage=false startx=0 starty=0 endx=640 endy=480 ! 
queue ! ffmpegcolorspace ! v4l2loopback

access the video using [pix_video], [pdp_v4l2] [#camera] or cat


running here.
this of course only grabs a portion of the screen, rather than a window and 
track that...

fgmadfs
IOhannes


___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] OT: Ir leds

2010-08-30 Thread Damian Stewart
 Date: Sun, 29 Aug 2010 19:44:16 -0700
 From: ronni montoya ronni.mont...@gmail.com

 Im making an interactive floor and i need to iluminate a space of 4 x
 4 meters with ir light.


Hi Ronni,

this question would be best asked on the openFrameworks forum 
http://openframeworks.cc/forum since many people there work professionally with 
computer vision and related interactive technology on a day-to-day basis...

hth
d

--
damian stewart . @damiannz .  dam...@frey.co.nz
frey . contemporary art . http://www.frey.co.nz


___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] live pd patch mixed with Gem or pdp?

2010-08-30 Thread John Harrison
is there a way to do these screenshots such that they don't actually show up
on a screen visible to the user i.e I don't have to show the raw Pd patch to
the user, only the mix? This is a 2-headed machine (one machine --- I'll use
2 if I absolutely have to) and the plan is to show a Gem or pdp window
fullscreen on one head and a pdp window fullscreen on the other. This leaves
me not sure where to put the Pd patch for the screenshots.

On Mon, Aug 30, 2010 at 10:44 AM, IOhannes m zmoelnig zmoel...@iem.atwrote:

 On 2010-08-30 17:04, John Harrison wrote:
  I'd like to take a patch from a pd window and mix it live with a Gem or
 pdp
  window. pdp_capture would have been the ticket, but I can't get it to
  compile and looking at the archives of this list, it doesn't seem like
  others have had luck with it either. Any other ideas for me?
 

 a bit complicated, but you can do that by doing screenshots with
 gstreamer, send it to a videoloopback device and read that (with pdp,
 Gem, GridFlow or whatelse)

 - get v4l2loopback (http://code.google.com/p/v4l2loopback/)
  compile, install, load the module
 - get gst-v4l2loopback (http://github.com/umlaeute/gst-v4l2loopback)
  compile (need gst-plugin-base-dev), install

 run the gstreamer-pipeline, e.g.
 $ gst-launch ximagesrc use-damage=false startx=0 starty=0 endx=640
 endy=480 ! queue ! ffmpegcolorspace ! v4l2loopback

 access the video using [pix_video], [pdp_v4l2] [#camera] or cat


 running here.
 this of course only grabs a portion of the screen, rather than a window
 and track that...

 fgmadfs
 IOhannes


___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] live pd patch mixed with Gem or pdp?

2010-08-30 Thread IOhannes m zmoelnig
On 2010-08-30 18:26, John Harrison wrote:
 is there a way to do these screenshots such that they don't actually show up
 on a screen visible to the user i.e I don't have to show the raw Pd patch to
 the user, only the mix? This is a 2-headed machine (one machine --- I'll use
 2 if I absolutely have to) and the plan is to show a Gem or pdp window
 fullscreen on one head and a pdp window fullscreen on the other. This leaves
 me not sure where to put the Pd patch for the screenshots.
 

if you want to use screenshots, then you have to do screenshots.
unfortunately screenshots have to be visible somehow (even if the
audience cannot see them).

the simplest solution would be to go 3-headed:
- head #1 for the patch (good for debugging, btw), not visible to the
audience
- head #2 for pdp screen #1
- head #3 for pdp screen #2


get yourself a matrox triplehead2go to multiply your 2nd head by 3 (so
you can actually have 4 different screens)

fgmasdr
IOhannes



smime.p7s
Description: S/MIME Cryptographic Signature
___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] live pd patch mixed with Gem or pdp?

2010-08-30 Thread ydego...@gmail.com

ola,

pdp_capture was working once with an old version of ImageMagick
but anyway it was too slow,
so now its compilation is made optional,
it is a bit abandonware..

use webcamstudio instead ( http://www.ws4gl.org/ ),
it does the same as gstreamer that Iohannes recommends,
but with an interface...
it will output the screenshots to a vloopback device ..

about the disposition on the screen, no eye deer ..

ahoj!
sevy

John Harrison wrote:
is there a way to do these screenshots such that they don't actually 
show up on a screen visible to the user i.e I don't have to show the 
raw Pd patch to the user, only the mix? This is a 2-headed machine 
(one machine --- I'll use 2 if I absolutely have to) and the plan is 
to show a Gem or pdp window fullscreen on one head and a pdp window 
fullscreen on the other. This leaves me not sure where to put the Pd 
patch for the screenshots.


On Mon, Aug 30, 2010 at 10:44 AM, IOhannes m zmoelnig zmoel...@iem.at 
mailto:zmoel...@iem.at wrote:


On 2010-08-30 17:04, John Harrison wrote:
 I'd like to take a patch from a pd window and mix it live with a
Gem or pdp
 window. pdp_capture would have been the ticket, but I can't get
it to
 compile and looking at the archives of this list, it doesn't
seem like
 others have had luck with it either. Any other ideas for me?


a bit complicated, but you can do that by doing screenshots with
gstreamer, send it to a videoloopback device and read that (with pdp,
Gem, GridFlow or whatelse)

- get v4l2loopback (http://code.google.com/p/v4l2loopback/)
 compile, install, load the module
- get gst-v4l2loopback (http://github.com/umlaeute/gst-v4l2loopback)
 compile (need gst-plugin-base-dev), install

run the gstreamer-pipeline, e.g.
$ gst-launch ximagesrc use-damage=false startx=0 starty=0 endx=640
endy=480 ! queue ! ffmpegcolorspace ! v4l2loopback

access the video using [pix_video], [pdp_v4l2] [#camera] or cat


running here.
this of course only grabs a portion of the screen, rather than a
window
and track that...

fgmadfs
IOhannes



___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -  
http://lists.puredata.info/listinfo/pd-list
   



___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


[PD] plug-in gnomevfssrc working with gstreamer [pdgst] is not recognised

2010-08-30 Thread Jack
Hello,

I'm trying to get stream from an icecast server and want to use
[gnomevfssrc].
I have installed 'gstreamer0.10-plugins-base' via Synaptic.
[gnomevfssrc] from gstreamer0.10-plugins-base can't be created. However,
I can create [adder] and others from gstreamer0.10-plugins-base.
Any clue about this problem ? I have missed something ?

pdgst $Revision: 0.0 $
(copyleft) IOhannes m zmoelnig @ IEM / KUG
compiled on Aug 30 2010 at 21:11:24 
compiled against Pd version 0.42.5.

Ubuntu 10.04, Pd 0.42.5, GEM 0.92.3
Thanx.
++

Jack




___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] Font weirdness with 0.43test

2010-08-30 Thread Hans-Christoph Steiner


On Sun, 2010-08-29 at 21:17 -0700, Miller Puckette wrote:
 Or the Mathieu idea - let the text be small, the box be 'standard' size,
 and let the text not fill the box -- but Pd could still know the true
 size of the font so editing could work.  Clearly we'll never be able to
 reliably set the font metrics and will have to play with what we're dealt,
 one way or another.

IMHO, I think it would be a fair amount of work to implement Matju's
idea, and that time would be better spent on other things. But I
wouldn't object to someone doing that.

 I'm giving up on getting this out by Sept. 1 -- there's another show-stopper
 bug that will take me more than 28 hrs. to fix (more confusion as to
 who's 'visible' versus graph on parent)
 
 d'oh...

Is this the bug I filed in the tracker?  I took a look, but I still
haven't made sense of all the GOP code so I didn't get far.

donecanvasdialog stops arrays from redrawing until reload
http://sourceforge.net/tracker/?func=detailaid=3052388group_id=55736atid=478070

.hc

 
 M
 
 
 On Sun, Aug 29, 2010 at 01:45:13PM -0400, Hans-Christoph Steiner wrote:
  
  The problem is that sending the measured font sizes via pd init puts us
  back at square one: the boxes will be different sizes depending on
  platform and now depending on font.  Then patch/GUI layout breaks across
  different platforms, and now when people use different fonts.
  
  The least of all evils with the current implementation is having the box
  sizes fixed, and the fonts being a bit too small on the large box/font
  sizes.  That is achieved by undoing the part you just added where it
  sends the measured sizes rather than the fixed sizes via pd init.
  
  The only way I can see properly doing this is having the pd -- pd-gui
  communication be logical not pixel-based, and have pd-gui handle all of
  the editing operations.
  
  .hc
  
  On Fri, 2010-08-27 at 22:01 -0700, Miller Puckette wrote:
   OK... applied that one, then went in and made the tcl script report the
   real font metrics to Pd so it can draw the boxes correctly (they were 
   still
   off on 2 of the font sizes in Fedora 13 and anyway I despair of ever being
   certain that the font sizes are exactly what Pd wished for) ... is the 
   current
   git upload working on others' machines?
   
   thanks
   Miller
   
   On Thu, Aug 26, 2010 at 04:43:16PM -0400, Hans-Christoph Steiner wrote:

Looks like here are the options:

- have boxes fit around fixed font sizes and have different box
sizes on each platform (vanilla up til 0.42)

- have font width fit into fixed box sizes and have long texts
overrun the bottom of boxes (Pd-extended since 0.40)

- hopefully the best for now, have the width, then height fit into
the fixed box sizes, and use fonts that are known to work the best.
I attached a patch which implements this and it seems to work pretty
well.  I discovered that the fixed box sizes were different than 
Pd-extended
so I synced them up so layout should be the same everywhere.

   
   
 (apply patch with git am to accept it as it).

.hc


On Aug 26, 2010, at 2:43 PM, Miller Puckette wrote:

OK.. but for 0.43... is there a stopgap?  I can't think of any way to
get text editing working (i.e., getting the character you click on to
be the one Pd sees you click on) except by going back to status quo.

cheers
MIller

On Thu, Aug 26, 2010 at 01:54:15PM -0400, Hans-Christoph Steiner
wrote:

On Aug 26, 2010, at 12:52 AM, Miller Puckette wrote:

My machine is running 8.5... but I don't think it's good to have
Pd's default font choice come out all wrong... ideally, it should do
something reasonable no matter what the font (well, at least if it's
fixed-pitch).  At the very least, Pd has to know the correct font
metrics... and it seems now not to.

I should add, given my experience, I don't think this is a realistic
goal without major surgery.  Its something I've put a lot of work
into trying to get working.  My experience tells me that the best
bet is using a default font.

Even better would be making the pd -- pd-gui communication not
deal with pixel values and mouse motion, but instead logical
operations. For example, instead of pd knowing anything about the
text editing operation, that should happen entirely in the gui.
Then once the editing operation is complete, the GUI would send the
result to pd.

.hc

Perhaps a solution for cross-platform patching (such as for making
help
files) could be to say this will look woerd on other machines
unless
you use font blaz to develope the patch so that you see the
canonical
sizes.

cheers
M

On Wed, Aug 25, 2010 at 04:18:49PM -0400, Hans-Christoph Steiner
wrote:

Here is the answer to this font problem:

- use Tcl/Tk 8.5 

Re: [PD] Font weirdness with 0.43test

2010-08-30 Thread Miller Puckette
On Mon, Aug 30, 2010 at 06:04:55PM -0400, Hans-Christoph Steiner wrote:
 
 
 On Sun, 2010-08-29 at 21:17 -0700, Miller Puckette wrote:
  Or the Mathieu idea - let the text be small, the box be 'standard' size,
  and let the text not fill the box -- but Pd could still know the true
  size of the font so editing could work.  Clearly we'll never be able to
  reliably set the font metrics and will have to play with what we're dealt,
  one way or another.
 
 IMHO, I think it would be a fair amount of work to implement Matju's
 idea, and that time would be better spent on other things. But I
 wouldn't object to someone doing that.
 
Maybe I'm not seeing something but I think I can code it in 1/2 hour... just
not right now while I'm getting ready to travel.

  I'm giving up on getting this out by Sept. 1 -- there's another show-stopper
  bug that will take me more than 28 hrs. to fix (more confusion as to
  who's 'visible' versus graph on parent)
  
  d'oh...
 
 Is this the bug I filed in the tracker?  I took a look, but I still
 haven't made sense of all the GOP code so I didn't get far.
 
 donecanvasdialog stops arrays from redrawing until reload
 http://sourceforge.net/tracker/?func=detailaid=3052388group_id=55736atid=478070

Not sure.  It's a whole complex of bugginess that has to do with the order in
which things are drawn and erased, that I had sort of swept under the rug
for 0.42 but which came bounding back when i 'fixed' SF patch 2913280 (at
least I think that's where it's coming from).  fun fun...

 .hc
 
  
  M
  
  
  On Sun, Aug 29, 2010 at 01:45:13PM -0400, Hans-Christoph Steiner wrote:
   
   The problem is that sending the measured font sizes via pd init puts us
   back at square one: the boxes will be different sizes depending on
   platform and now depending on font.  Then patch/GUI layout breaks across
   different platforms, and now when people use different fonts.
   
   The least of all evils with the current implementation is having the box
   sizes fixed, and the fonts being a bit too small on the large box/font
   sizes.  That is achieved by undoing the part you just added where it
   sends the measured sizes rather than the fixed sizes via pd init.
   
   The only way I can see properly doing this is having the pd -- pd-gui
   communication be logical not pixel-based, and have pd-gui handle all of
   the editing operations.
   
   .hc
   
   On Fri, 2010-08-27 at 22:01 -0700, Miller Puckette wrote:
OK... applied that one, then went in and made the tcl script report the
real font metrics to Pd so it can draw the boxes correctly (they were 
still
off on 2 of the font sizes in Fedora 13 and anyway I despair of ever 
being
certain that the font sizes are exactly what Pd wished for) ... is the 
current
git upload working on others' machines?

thanks
Miller

On Thu, Aug 26, 2010 at 04:43:16PM -0400, Hans-Christoph Steiner wrote:
 
 Looks like here are the options:
 
 - have boxes fit around fixed font sizes and have different box
 sizes on each platform (vanilla up til 0.42)
 
 - have font width fit into fixed box sizes and have long texts
 overrun the bottom of boxes (Pd-extended since 0.40)
 
 - hopefully the best for now, have the width, then height fit into
 the fixed box sizes, and use fonts that are known to work the best.
 I attached a patch which implements this and it seems to work pretty
 well.  I discovered that the fixed box sizes were different than 
 Pd-extended
 so I synced them up so layout should be the same everywhere.
 


  (apply patch with git am to accept it as it).
 
 .hc
 
 
 On Aug 26, 2010, at 2:43 PM, Miller Puckette wrote:
 
 OK.. but for 0.43... is there a stopgap?  I can't think of any way to
 get text editing working (i.e., getting the character you click on to
 be the one Pd sees you click on) except by going back to status quo.
 
 cheers
 MIller
 
 On Thu, Aug 26, 2010 at 01:54:15PM -0400, Hans-Christoph Steiner
 wrote:
 
 On Aug 26, 2010, at 12:52 AM, Miller Puckette wrote:
 
 My machine is running 8.5... but I don't think it's good to have
 Pd's default font choice come out all wrong... ideally, it should 
 do
 something reasonable no matter what the font (well, at least if 
 it's
 fixed-pitch).  At the very least, Pd has to know the correct font
 metrics... and it seems now not to.
 
 I should add, given my experience, I don't think this is a realistic
 goal without major surgery.  Its something I've put a lot of work
 into trying to get working.  My experience tells me that the best
 bet is using a default font.
 
 Even better would be making the pd -- pd-gui communication not
 deal with pixel values and mouse motion, but instead logical
 operations. For example, instead of pd knowing anything 

Re: [PD] live pd patch mixed with Gem or pdp?

2010-08-30 Thread John Harrison
bleh I'm failing to get either v4l2loopback or webcamstudio modules to 
actually load.


I'm running ubuntu Lucid 10.04 with 2.6.32-24-generic kernel

v4l2loopback svn builds and installs ok

however, modprobe v4l2loopback returns in dmesg:

   [   46.328926] v4l2loopback: disagrees about version of symbol
   video_unregister_device
   [   46.328929] v4l2loopback: Unknown symbol video_unregister_device
   [   46.328996] v4l2loopback: disagrees about version of symbol
   video_device_alloc
   [   46.328998] v4l2loopback: Unknown symbol video_device_alloc
   [   46.329091] v4l2loopback: disagrees about version of symbol
   video_register_device
   [   46.329093] v4l2loopback: Unknown symbol video_register_device
   [   46.329178] v4l2loopback: disagrees about version of symbol
   video_device_release
   [   46.329179] v4l2loopback: Unknown symbol video_device_release

similarly, modprobe webcamstudio returns in dmesg:

   [   39.281212] webcamstudio: disagrees about version of symbol
   video_devdata
   [   39.281215] webcamstudio: Unknown symbol video_devdata
   [   39.281309] webcamstudio: disagrees about version of symbol
   video_unregister_device
   [   39.281311] webcamstudio: Unknown symbol video_unregister_device
   [   39.281376] webcamstudio: disagrees about version of symbol
   video_device_alloc
   [   39.281377] webcamstudio: Unknown symbol video_device_alloc
   [   39.281446] webcamstudio: disagrees about version of symbol
   video_register_device
   [   39.281448] webcamstudio: Unknown symbol video_register_device
   [   39.281551] webcamstudio: disagrees about version of symbol
   video_device_release
   [   39.281553] webcamstudio: Unknown symbol video_device_release

a reboot doesn't help.

-John

On 08/30/2010 11:40 AM, ydego...@gmail.com wrote:

ola,

pdp_capture was working once with an old version of ImageMagick
but anyway it was too slow,
so now its compilation is made optional,
it is a bit abandonware..

use webcamstudio instead ( http://www.ws4gl.org/ ),
it does the same as gstreamer that Iohannes recommends,
but with an interface...
it will output the screenshots to a vloopback device ..

about the disposition on the screen, no eye deer ..

ahoj!
sevy

John Harrison wrote:
is there a way to do these screenshots such that they don't actually 
show up on a screen visible to the user i.e I don't have to show the 
raw Pd patch to the user, only the mix? This is a 2-headed machine 
(one machine --- I'll use 2 if I absolutely have to) and the plan is 
to show a Gem or pdp window fullscreen on one head and a pdp window 
fullscreen on the other. This leaves me not sure where to put the Pd 
patch for the screenshots.


On Mon, Aug 30, 2010 at 10:44 AM, IOhannes m zmoelnig 
zmoel...@iem.at mailto:zmoel...@iem.at wrote:


On 2010-08-30 17:04, John Harrison wrote:
 I'd like to take a patch from a pd window and mix it live with a
Gem or pdp
 window. pdp_capture would have been the ticket, but I can't get
it to
 compile and looking at the archives of this list, it doesn't
seem like
 others have had luck with it either. Any other ideas for me?


a bit complicated, but you can do that by doing screenshots with
gstreamer, send it to a videoloopback device and read that (with 
pdp,

Gem, GridFlow or whatelse)

- get v4l2loopback (http://code.google.com/p/v4l2loopback/)
 compile, install, load the module
- get gst-v4l2loopback (http://github.com/umlaeute/gst-v4l2loopback)
 compile (need gst-plugin-base-dev), install

run the gstreamer-pipeline, e.g.
$ gst-launch ximagesrc use-damage=false startx=0 starty=0 endx=640
endy=480 ! queue ! ffmpegcolorspace ! v4l2loopback

access the video using [pix_video], [pdp_v4l2] [#camera] or cat


running here.
this of course only grabs a portion of the screen, rather than a
window
and track that...

fgmadfs
IOhannes



___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] live pd patch mixed with Gem or pdp?

2010-08-30 Thread John Harrison

ok I got the modules to load by reloading the kernel image

   sudo apt-get install --reinstall linux-image-`uname -r`
   sudo depmod -a

now Yves solution works with pdp_v4l.

although modprobe v4l2loopback seems to work, Iohannes's incantation to 
launch the pipeline returns: WARNING: erroneous pipeline: no element 
v4l2loopback


also /dev/video3 becomes available when I modprobe v4l2loopback but 
nothing can seem to access it --- invalid argument errors.


Assuming I can configure webcamstudio to autostart with the 
configuration I need tho, I should be all set at this point.


Thanks for help!

-John

On 08/30/2010 11:59 PM, John Harrison wrote:
bleh I'm failing to get either v4l2loopback or webcamstudio modules to 
actually load.


I'm running ubuntu Lucid 10.04 with 2.6.32-24-generic kernel

v4l2loopback svn builds and installs ok

however, modprobe v4l2loopback returns in dmesg:

[   46.328926] v4l2loopback: disagrees about version of symbol
video_unregister_device
[   46.328929] v4l2loopback: Unknown symbol video_unregister_device
[   46.328996] v4l2loopback: disagrees about version of symbol
video_device_alloc
[   46.328998] v4l2loopback: Unknown symbol video_device_alloc
[   46.329091] v4l2loopback: disagrees about version of symbol
video_register_device
[   46.329093] v4l2loopback: Unknown symbol video_register_device
[   46.329178] v4l2loopback: disagrees about version of symbol
video_device_release
[   46.329179] v4l2loopback: Unknown symbol video_device_release

similarly, modprobe webcamstudio returns in dmesg:

[   39.281212] webcamstudio: disagrees about version of symbol
video_devdata
[   39.281215] webcamstudio: Unknown symbol video_devdata
[   39.281309] webcamstudio: disagrees about version of symbol
video_unregister_device
[   39.281311] webcamstudio: Unknown symbol video_unregister_device
[   39.281376] webcamstudio: disagrees about version of symbol
video_device_alloc
[   39.281377] webcamstudio: Unknown symbol video_device_alloc
[   39.281446] webcamstudio: disagrees about version of symbol
video_register_device
[   39.281448] webcamstudio: Unknown symbol video_register_device
[   39.281551] webcamstudio: disagrees about version of symbol
video_device_release
[   39.281553] webcamstudio: Unknown symbol video_device_release

a reboot doesn't help.

-John

On 08/30/2010 11:40 AM, ydego...@gmail.com wrote:

ola,

pdp_capture was working once with an old version of ImageMagick
but anyway it was too slow,
so now its compilation is made optional,
it is a bit abandonware..

use webcamstudio instead ( http://www.ws4gl.org/ ),
it does the same as gstreamer that Iohannes recommends,
but with an interface...
it will output the screenshots to a vloopback device ..

about the disposition on the screen, no eye deer ..

ahoj!
sevy

John Harrison wrote:
is there a way to do these screenshots such that they don't actually 
show up on a screen visible to the user i.e I don't have to show the 
raw Pd patch to the user, only the mix? This is a 2-headed machine 
(one machine --- I'll use 2 if I absolutely have to) and the plan is 
to show a Gem or pdp window fullscreen on one head and a pdp window 
fullscreen on the other. This leaves me not sure where to put the Pd 
patch for the screenshots.


On Mon, Aug 30, 2010 at 10:44 AM, IOhannes m zmoelnig 
zmoel...@iem.at mailto:zmoel...@iem.at wrote:


On 2010-08-30 17:04, John Harrison wrote:
 I'd like to take a patch from a pd window and mix it live with a
Gem or pdp
 window. pdp_capture would have been the ticket, but I can't get
it to
 compile and looking at the archives of this list, it doesn't
seem like
 others have had luck with it either. Any other ideas for me?


a bit complicated, but you can do that by doing screenshots with
gstreamer, send it to a videoloopback device and read that (with 
pdp,

Gem, GridFlow or whatelse)

- get v4l2loopback (http://code.google.com/p/v4l2loopback/)
 compile, install, load the module
- get gst-v4l2loopback 
(http://github.com/umlaeute/gst-v4l2loopback)

 compile (need gst-plugin-base-dev), install

run the gstreamer-pipeline, e.g.
$ gst-launch ximagesrc use-damage=false startx=0 starty=0 endx=640
endy=480 ! queue ! ffmpegcolorspace ! v4l2loopback

access the video using [pix_video], [pdp_v4l2] [#camera] or cat


running here.
this of course only grabs a portion of the screen, rather than a
window
and track that...

fgmadfs
IOhannes



___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list




--
John Harrison http://alumni.media.mit.edu/~harrison
___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] [PD-announce] pd 0.43 test 2 (first 'real' test version) released

2010-08-30 Thread Hans-Christoph Steiner

This works for me now.

.hc

On Sun, 2010-08-22 at 12:03 +0200, cyrille henry wrote:
 you can use pd~ help file.
 cyrille
 
 
 Le 22/08/2010 00:06, Hans-Christoph Steiner a écrit :
 
  Can you post a test patch?
 
  .hc
 
  On Sat, 2010-08-21 at 21:03 +0100, Marco Donnarumma wrote:
  mm, I was going to write about the same issue.
  It might be OT, but the same happens here on pd-ext 42.5 rc5 - Ubuntu
  Jaunty - compiled
 
  M
 
 
 
 
   hello
 
   i've got probleme with pd~
   when sending a start message, i've got an error :
   pd~: can't stat /usr/local/lib/pd/pd
 
   (pd is install somwhere else.)
 
   cheers
   Cyrille
   
 
  --
  Marco Donnarumma aka TheSAD
  Independent New Media Arts Professional, Performer, Teacher -
  Edinburgh, UK
 
 
  PORTFOLIO: http://marcodonnarumma.com
  LAB: http://www.thesaddj.com | http://cntrl.sourceforge.net |
  http://www.flxer.net
  EVENT: http://www.liveperformersmeeting.net
  ___
  Pd-list@iem.at mailing list
  UNSUBSCRIBE and account-management -  
  http://lists.puredata.info/listinfo/pd-list
 
 
 



___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list


Re: [PD] [PD-announce] pd 0.43 test 2 (first 'real' test version) released

2010-08-30 Thread Hans-Christoph Steiner

Ok, should be fixed in the day after tomorrow's nightly build (its too
late for tonight's builds, they are already happening, and I should be
going to sleep...

.hc

On Sat, 2010-08-21 at 21:03 +0100, Marco Donnarumma wrote:
 mm, I was going to write about the same issue.
 It might be OT, but the same happens here on pd-ext 42.5 rc5 - Ubuntu
 Jaunty - compiled
 
 M
 
 
 
 
 hello
 
 i've got probleme with pd~
 when sending a start message, i've got an error :
 pd~: can't stat /usr/local/lib/pd/pd
 
 (pd is install somwhere else.)
 
 cheers
 Cyrille
 
 
 -- 
 Marco Donnarumma aka TheSAD
 Independent New Media Arts Professional, Performer, Teacher -
 Edinburgh, UK
 
 
 PORTFOLIO: http://marcodonnarumma.com
 LAB: http://www.thesaddj.com | http://cntrl.sourceforge.net |
 http://www.flxer.net
 EVENT: http://www.liveperformersmeeting.net
 ___
 Pd-list@iem.at mailing list
 UNSUBSCRIBE and account-management - 
 http://lists.puredata.info/listinfo/pd-list



___
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management - 
http://lists.puredata.info/listinfo/pd-list