Bugs item #3007370, was opened at 2010-05-26 13:53
Message generated for change (Comment added) made by sf-robot
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507079&aid=3007370&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: Pixes (pix_ objects)
Group: None
>Status: Closed
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Thomas O Fredericks (mrtof)
Assigned to: Nobody/Anonymous (nobody)
Summary: pix_dump does not properly after an image has been resized
Initial Comment:
pix_dump does not dump the pixel data properly of a greyscale video feed that has been resized to 25x25 pixels:
1) The number of pixels returned is 624 instead of 625!
2) If the dumped data is fed back to a pix_set the resulting image is garbage
Note: I want to resize the video to such a low resolution to send the pixel data by serial to a Peggy 2.
----------------------------------------------------------------------
>Comment By: SourceForge Robot (sf-robot)
Date: 2010-06-10 02:20
Message:
This Tracker item was closed automatically by the system. It was
previously set to a Pending status, and the original submitter
did not respond within 14 days (the time period specified by
the administrator of this Tracker).
----------------------------------------------------------------------
Comment By: IOhannes m zmlnig (zmoelnig)
Date: 2010-05-26 16:07
Message:
[pix_dump] in greyscale mode currently needs the number of pixels be
dividible by 4.
so if you can live with a 24*24 or a 26*26 image, you will be on the save
side.
anyhow, the issue only skips the remaining pixels, so adding dummy-pixels
manually could help...
i committed a fix to svn/trunk with revision #3357
i committed a fix to svn/branches/0.92 with revision #3359
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507079&aid=3007370&group_…
Bugs item #3007373, was opened at 2010-05-26 13:59
Message generated for change (Comment added) made by sf-robot
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507079&aid=3007373&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: Pixes (pix_ objects)
Group: None
>Status: Closed
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Thomas O Fredericks (mrtof)
Assigned to: Nobody/Anonymous (nobody)
Summary: pix_resize crashes PD if a dimension is set before rendering
Initial Comment:
pix_resize crashes PD once the gemwin is created if a "dimen 25 25" was sent to the object before the gemwin was created.
This also happens if the dimensions are set as arguments.
----------------------------------------------------------------------
>Comment By: SourceForge Robot (sf-robot)
Date: 2010-06-10 02:20
Message:
This Tracker item was closed automatically by the system. It was
previously set to a Pending status, and the original submitter
did not respond within 14 days (the time period specified by
the administrator of this Tracker).
----------------------------------------------------------------------
Comment By: IOhannes m zmlnig (zmoelnig)
Date: 2010-05-26 16:11
Message:
ah, i finally managed to reproduce.
the problem seems to happen only with greyscale images.
i suspect that the internally used gluScaleImage() needs a bit more memory
allocated than just the target image.
a simple workaround is, to do the first resize to a bigger image (e.g.
only downsize once the rendering has started); the image will not free the
allocated space if the new size fits into the old memory, thus giving you
enough headroom.
i committed a fix to svn/trunk with revision #3358
i committed a fix to svn/branches/0.92 with revision #3360
this "fix" assumes that 4 bytes per pixel are enough.(which is an
assumption on my side)
----------------------------------------------------------------------
Comment By: IOhannes m zmlnig (zmoelnig)
Date: 2010-05-26 15:00
Message:
unable to reproduce
i tried with both Gem-0.92 (svn/branches/0.92) and svn/trunk an there ar
no crashes.
please provide detailed information about your system (versions!) and a
simple example patch that triggers the problem.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507079&aid=3007373&group_…
Bugs item #3013128, was opened at 2010-06-08 11:06
Message generated for change (Tracker Item Submitted) made by paresys
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507079&aid=3013128&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: Pixes (pix_ objects)
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: paresys (paresys)
Assigned to: Nobody/Anonymous (nobody)
Summary: pix_color bad colors
Initial Comment:
The color at the output of pix_color is not correct
by example: Message 0.23 0.23 0.23 display color black
Problem with colorspace yuv, Pd-0.42.5-extended-rc1 (GEM: ver: 0.92.3), MacOSX.4, Intel or PPC.
No problem with colorspace rgba
No problem with Pd-0.41.4-extended
I have not tested under Linux.
I have not tested under Windows
Gerard.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507079&aid=3013128&group_…
Hi list,
Is there any way in current Gem to use asynchronous readback of pixel
buffer objects to speed up recording (or other GPU<->CPU transfer)?
In one of my projects in plain C/OpenGL I used this technique successfully:
1) naive (blocking) glReadPixels direct to CPU ram and piped via
stdout to ppmtoy4m|y4mscaler|mpeg2enc
-> about 5fps
2) async glReadPixels to pixel buffer object mapped into the CPU
with one frame latency and piped via stdout to same encoder
-> about 20fps
3) async glReadPixels to PBO to CPU piped via stdout to /dev/null
-> about 25fps
(laptop with 4GB ram, intel core2 dual core cpu, nvidia g105m gpu,
tested with 32bit gnu/linux/ubuntu/karmic generic-pae kernel)
Thanks for any tips,
Claude
--
http://claudiusmaximus.goto10.org
Feature Requests item #2998083, was opened at 2010-05-07 09:35
Message generated for change (Settings changed) made by nusmuk
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507082&aid=2998083&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: Closed
Priority: 5
Private: No
Submitted By: Cyrille Henry (nusmuk)
Assigned to: IOhannes m zmlnig (zmoelnig)
Summary: raw camera
Initial Comment:
guppy camera and lot's of good quality FW cam output raw data.
(by example : http://www.stemmer-imaging.fr/fr/pages/products/line.php?item=AVT.Guppy.CCD…)
Most sotftware (processing , jitter, ..) can use this cam , but not Gem.
it would be really usefull to be able to use this kind of hardware
----------------------------------------------------------------------
>Comment By: Cyrille Henry (nusmuk)
Date: 2010-06-02 13:03
Message:
yes, it now work with DC1394.
thanks a lot
----------------------------------------------------------------------
Comment By: IOhannes m zmlnig (zmoelnig)
Date: 2010-05-27 07:15
Message:
i guess with "raw" you mean cameras adhering to the "IIDC" standard.
in current trunk there is support for this kind of cameras ("videoDC1394"
plugin)
----------------------------------------------------------------------
Comment By: Cyrille Henry (nusmuk)
Date: 2010-05-07 10:40
Message:
i try coriender, and it segfault...
indeed, direct suport would be nice.
----------------------------------------------------------------------
Comment By: IOhannes m zmlnig (zmoelnig)
Date: 2010-05-07 10:11
Message:
i guess you could use coriander to make a loopback device readabke by Gem.
of course it would be nicer to have direct support (so i leave this FR
open)
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=507082&aid=2998083&group_…