GEM-dev January 2012

gem-dev@lists.iem.at

9 participants
38 discussions

Re: [GEM-dev] A/D video converter DFG/1394-1e with unicap
by kubro bubro 11 Jan '12

11 Jan '12

yes, i dont need it in unicap, if it works just somehow i will be happy. i tried dc1394 and dv4l before. removed unicap, and try dsc1394: [pix_video]: backend #0='v4l2' : v4l2 analog [pix_video]: backend #1='v4l' : v4l analog [pix_video]: backend #2='dv4l' : dv4l dv [pix_video]: backend #3='dc1394' : dc1394 iidc [pix_video]: backend #4='vlc' : vlc videoDC1394: no cameras found doesnt matter if run as root, or set dimen 720 576 with dv4l is more fun: videoctl: dimen 720 576 //doesnt matter about this, allways the same videoctl: driver dv4l videoctl: device /dev/fw1 info: proplist numread 10 info: proplist read 50_fields float 1 info: proplist read PAL float 1 info: proplist read is_color float 1 info: proplist read letterbox float 1 info: proplist read normal float 1 info: proplist read progressive float 1 info: proplist read quality float 0 info: proplist read recording_datetime symbol YYYY-MM-DD hh:mm:ss info: proplist read timestamp float 0 info: proplist read wide float 1 info: proplist numwrite 1 info: proplist write quality float 5 GEM: Direct Rendering enabled! GEM: GLEW version 1.5.4 GEM: Start rendering starting capture thread GEM: Stop rendering no errors, but picture is black, with some still(i mean not mooving) noisy artefacts on the bottom of the picture..

2 1

[ pd-gem-Feature Requests-2520241 ] include new libparticle
by SourceForge.net 11 Jan '12

11 Jan '12

Feature Requests item #2520241, was opened at 2009-01-19 08:35 Message generated for change (Comment added) made by zmoelnig You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=2520241&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None >Status: Closed Priority: 5 Private: No Submitted By: IOhannes m zmölnig (zmoelnig) Assigned to: Nobody/Anonymous (nobody) Summary: include new libparticle Initial Comment: libparticle has been updated. please include the new version with Gem ---------------------------------------------------------------------- >Comment By: IOhannes m zmölnig (zmoelnig) Date: 2012-01-11 05:12 Message: actually i think that the particle system should not be part of Gem at all. the only things that require Gem are [part_draw] and [part_render], and both should be implementable with [gemlist]. maybe it would be good to create a Pd-library "particles" from the library[1] [1] http://www.particlesystems.org/ ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=2520241&group_…

1 0

[ pd-gem-Feature Requests-3028679 ] last reactivision in pix_fiducialtrack
by SourceForge.net 10 Jan '12

10 Jan '12

Feature Requests item #3028679, was opened at 2010-07-12 15:22 Message generated for change (Comment added) made by zmoelnig You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=3028679&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: image processing Group: None >Status: Pending Priority: 5 Private: No Submitted By: jack (jackovic) Assigned to: IOhannes m zmölnig (zmoelnig) Summary: last reactivision in pix_fiducialtrack Initial Comment: In pix_fiducialtrack.cpp, it is written : '2005 IOhannes m zmoelnig.'. The lastest release of reactivision dated May 19 2009. IOhannes, do you know if it is possible to update this object with the latest reactivision (1.4) ? Maybe, it is useless ? ---------------------------------------------------------------------- >Comment By: IOhannes m zmölnig (zmoelnig) Date: 2012-01-10 09:16 Message: i updated the underlying libfidtrack to the last release (reacTIVIsion-1.4) i had to change a single line for a trivial fix in pix_fiducialtrack.cpp, so that file is basically still (c)2005 (though i did update the changelog); a better indicator is to have a look at the libfidtrack sources.... ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=3028679&group_…

1 0

A/D video converter DFG/1394-1e with unicap
by kubro bubro 10 Jan '12

10 Jan '12

hello last year we was trying to run firewire A/D video converter DFG/1394-1e with unicap in gem whitout success. there are three sources Composite 1, Composite 2, and S-video. im giving it next try. i have: libunicap-0.9.12-1 tcl-8.5.11-1 tk-8.5.11-1 gem-git-201201010 - today pd-git-201201010 - today ls -l /usr/local/lib/pd/extra/Gem/*.so -rwxr-xr-x 1 root root 31368 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_filmAVIPLAY.so -rwxr-xr-x 1 root root 31360 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_filmQT4L.so -rwxr-xr-x 1 root root 35520 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_imageJPEG.so -rwxr-xr-x 1 root root 43728 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_imageMAGICK.so -rwxr-xr-x 1 root root 35200 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_imageSGI.so -rwxr-xr-x 1 root root 47840 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_imageTIFF.so -rwxr-xr-x 1 root root 56192 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_recordQT4L.so -rwxr-xr-x 1 root root 18864 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_recordV4L.so -rwxr-xr-x 1 root root 18872 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_recordV4L2.so -rwxr-xr-x 1 root root 64544 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_videoDC1394.so -rwxr-xr-x 1 root root 48064 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_videoDV4L.so -rwxr-xr-x 1 root root 60416 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_videoUNICAP.so -rwxr-xr-x 1 root root 80672 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_videoV4L.so -rwxr-xr-x 1 root root 76704 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_videoV4L2.so -rwxr-xr-x 1 root root 43712 jan 5 12:40 /usr/local/lib/pd/extra/Gem/gem_videoVLC.so connecting device to firewire: [ 1925.300583] firewire_core: created device fw1: GUID 0007480029010454, S400 [ 1925.300595] firewire_core: phy config: card 0, new root=ffc1, gap_count=5 ls -l /dev/fw* crwxr-xr-x 1 root root 252, 0 jan 10 13:32 /dev/fw0 crwxr-xr-x 1 root root 252, 1 jan 10 14:05 /dev/fw1 ///but anyway i have to run pd as root, otherwise i dont even get device properties ///i will work with pix_video-help.pd [pix_video]: backend #0='v4l2' : v4l2 analog [pix_video]: backend #1='v4l' : v4l analog [pix_video]: backend #2='dv4l' : dv4l dv [pix_video]: backend #3='dc1394' : dc1394 iidc [pix_video]: backend #4='unicap' : unicap analog libvid21394 [pix_video]: backend #5='vlc' : vlc ///this error is displayed only if i run pd as root videoctl: enumerate info: devices 3 info: device /dev/fw1 dv4l info: device DFG/1394-1 29010454 unicap info: device vlc vlc videoctl: driver unicap (Tcl) INVALID COMMAND NAME: invalid command name ".x262c5d0.c" while executing ".x262c5d0.c delete 262cfa0BASE0" ("uplevel" body line 6) invoked from within "uplevel #0 $cmd_from_pd" ///this error is related to the [pd properties] subpatch in pix_video-help.pd, as when not present, error dissaperars.. ///when i enter "driver unicap" 2nd time, it gives no error, but property list: videoctl: driver unicap info: proplist numread 16 info: proplist read Auto Exposure Mode symbol Full Auto info: proplist read Focus float 50000 info: proplist read Gain float 255 info: proplist read Iris float 16 info: proplist read Shutter float 255 info: proplist read White Balance float 2 info: proplist read Zoom float 50000 info: proplist read brightness float 1 info: proplist read contrast float 1 info: proplist read firmware version float 99999 info: proplist read force odd/even float 1 info: proplist read link speed float 2 info: proplist read rs232 baud rate float 4 info: proplist read rs232 io bang info: proplist read source symbol Composite 1 info: proplist read video norm symbol NTSC info: proplist numwrite 16 info: proplist write Auto Exposure Mode symbol Full Auto info: proplist write Focus float 50000 info: proplist write Gain float 255 info: proplist write Iris float 16 info: proplist write Shutter float 255 info: proplist write White Balance float 2 info: proplist write Zoom float 50000 info: proplist write brightness float 1 info: proplist write contrast float 1 info: proplist write firmware version float 99999 info: proplist write force odd/even float 1 info: proplist write link speed float 2 info: proplist write rs232 baud rate float 4 info: proplist write rs232 io bang info: proplist write source symbol Composite 1 info: proplist write video norm symbol NTSC /// adjusting properties: videoctl: get video norm info: prop video norm PAL videoctl: set video norm PAL videoctl: get source info: prop source Composite 1 ///start render GEM: Direct Rendering enabled! GEM: GLEW version 1.5.4 GEM: Start rendering sending "dimen 720 576" in any time doesnt change situation, picture remains white am i doing something wrong, or there is some error? kubriel

2 1

announce: assimp model loader
by IOhannes m zmoelnig 06 Jan '12

06 Jan '12

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 i just pushed a changeset to sourceforge, that adds some possibilities to 3d model loading to Gem. first, model-loading is now factored out into plugins as well. the old .obj loader code has been ported into a new "gem_modelOBJ" plugin. then i tried to create another asset importer plugin based on assimp. the plugin is (not surprisingly) called "gem_modelASSIMP". assimp [1] should allow us to read a large number of 3d model formats (including dfx, 3ds, md5, blender, collada,...) what works: - - loading of models - - UV-textures what needs to be done: - - check for other texcoord generations. - - allow feedback on which textures should be loaded - - probably do the texture loading within the model-importer (on user-request) - - implement animations - - improve performance cudos to hans for bringing assimp to my attention. check and have fun (and report back any problems...) fg,asdr IOhannes [1] http://assimp.sourceforge.net there are also debian packages, and ubuntu (starting with precise(!)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk8F7YIACgkQkX2Xpv6ydvTwtwCgyE8VPOReImeMmypUSJru4Qig 7IIAn1ZOYqVRe7SgsNaGfe77Uzm8//ER =NhqH -----END PGP SIGNATURE-----

3 2

[ pd-gem-Feature Requests-3468897 ] having more drawing style for primitive in Base/GemShape.cpp
by SourceForge.net 03 Jan '12

03 Jan '12

Feature Requests item #3468897, was opened at 2012-01-03 02:54 Message generated for change (Comment added) made by zmoelnig You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=3468897&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None >Status: Pending Priority: 5 Private: No Submitted By: Cyrille Henry (nusmuk) Assigned to: Nobody/Anonymous (nobody) Summary: having more drawing style for primitive in Base/GemShape.cpp Initial Comment: Thanks to the new vertexbuffer object, on can create a wild range of geometry. but the current rendering style are limited to : GL_LINE_LOOP, GL_POLYGON, GL_QUADS, GL_POINTS, GL_TRIANGLES, GL_TRIANGLE_STRIP; This is a bit limiting. By example, the audio oscilloscope in the vertexbuffer help file would benefit being rendered as GL_LINE_STRIP better than GL_LINE_LOOP. in fact, the missing rendering style are : GL_LINE, GL_LINE_STRIP, GL_TRIANGLES_STRIP, GL_TRIANGLES_FAN and GL_QUADS_STRIP. The implementation problem comes from the fact that only the 1st letter of the draw message was tested in order to discriminate from Line / points / triangle etc. So adding more message may break backward compatibility. Should we consider that "draw Lollipop" is not a legal message and should return an error, better than drawing Line_Loop? ---------------------------------------------------------------------- >Comment By: IOhannes m zmölnig (zmoelnig) Date: 2012-01-03 05:25 Message: i simplified the drawstyle logic a bit, by using a string->enum map, which can be changed by objectclasses if they are not happy with the defaults (which now includes almost all drawstyles you mention) if the user provides an illegal draw-type, they get an error and a list of all possible draw-styles. the only missing drawstyle from the default set is GL_LINE (for backward compat, "line" out to stay GL_LINE_LOOP), for which i have no good name; if an objectclass really needs it, they could simply override the mapping from "line" to GL_LINE. pushed. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=3468897&group_…

1 0

[ pd-gem-Feature Requests-3468897 ] having more drawing style for primitive in Base/GemShape.cpp
by SourceForge.net 03 Jan '12

03 Jan '12

Feature Requests item #3468897, was opened at 2012-01-03 02:54 Message generated for change (Tracker Item Submitted) made by nusmuk You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=3468897&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Priority: 5 Private: No Submitted By: Cyrille Henry (nusmuk) Assigned to: Nobody/Anonymous (nobody) Summary: having more drawing style for primitive in Base/GemShape.cpp Initial Comment: Thanks to the new vertexbuffer object, on can create a wild range of geometry. but the current rendering style are limited to : GL_LINE_LOOP, GL_POLYGON, GL_QUADS, GL_POINTS, GL_TRIANGLES, GL_TRIANGLE_STRIP; This is a bit limiting. By example, the audio oscilloscope in the vertexbuffer help file would benefit being rendered as GL_LINE_STRIP better than GL_LINE_LOOP. in fact, the missing rendering style are : GL_LINE, GL_LINE_STRIP, GL_TRIANGLES_STRIP, GL_TRIANGLES_FAN and GL_QUADS_STRIP. The implementation problem comes from the fact that only the 1st letter of the draw message was tested in order to discriminate from Line / points / triangle etc. So adding more message may break backward compatibility. Should we consider that "draw Lollipop" is not a legal message and should return an error, better than drawing Line_Loop? ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=507082&aid=3468897&group_…

1 0

vertexbuffer
by cyrille henry 03 Jan '12

03 Jan '12

hello, I commit few change in the vertexbuffer branch. it look like it's ready to be merged with the master. Iohannes, could you check and merge it if you agree? thanks Cyrille

2 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

GEM-dev January 2012