Hi Matthew,
(I work with dischi on Freevo, and have in the past done most of the
work on the GUI core and in particular video player integration.)
On Sun, 2008-04-13 at 11:59 +0100, Matthew Allum wrote:
There is some very rough support for uploading YUV - though I suspect it
could just push
On Tue, 2008-04-15 at 12:50 +0200, Florent wrote:
(with the lazy flag should still work).
WELL DONE !
Very nice indeed! Thanks to Blake and Emmanuele.
Cheers,
Jason.
--
To unsubscribe send a mail to [EMAIL PROTECTED]
I understood that it was possible to position actors with subpixel
precision. I have a texture that I want to move by 0.1f along the x and
y axis every 30ms. I'm sure the best way to do this would be using a
timeline, but as a test implementation, I just did:
gboolean slide(gpointer
[Resubmitting to list; sorry for the dupe, Øyvind]
On Mon, 2008-06-30 at 10:27 +0100, Øyvind Kolås wrote:
GPU. The actual contents of the texture should move with subpixel
precision though (try rendering a grid with 1px wide lines and wee how
that behaves).
You're quite right.
Thanks for the
On Mon, 2008-06-30 at 17:36 +0100, Matthew Allum wrote:
oh doh, and did it work with the original patch ?
Yep. Worked with the original patch by Andy (attached to that bug),
worked with my kludge I described in my first email (adding a damage
region for the window and handling damage events
On Mon, 2008-06-30 at 12:45 -0400, Jason Tackaberry wrote:
Yep. Worked with the original patch by Andy (attached to that bug),
worked with my kludge I described in my first email (adding a damage
region for the window and handling damage events myself), but does not
work with latest trunk
On Mon, 2008-06-30 at 13:03 -0400, Jason Tackaberry wrote:
See the attached patch, which corrects the problem.
A bug to report, however. (Not introduced by today's changes, they
existed yesterday as well.)
It seems that the pixmap is not being automatically updated when it uses
On Mon, 2008-06-30 at 23:01 +0100, Matthew Allum wrote:
Doh, similar patch from Mr Wingo (see
http://bugzilla.openedhand.com/show_bug.cgi?id=1008 ) now applied to
trunk.
Looks good, working now.
Thanks,
Jason.
--
To unsubscribe send a mail to [EMAIL PROTECTED]
On Mon, 2008-06-30 at 18:44 -0400, Jason Tackaberry wrote:
No, after:
tex = clutter_glx_texture_pixmap_new_with_window (win_remote);
clutter_texture_set_filter_quality(CLUTTER_TEXTURE(tex),
CLUTTER_TEXTURE_QUALITY_HIGH);
If I
On Tue, 2008-07-01 at 00:17 +0100, Matthew Allum wrote:
Ok thanks. Could you post on bugzilla with details of driver/card your
using and if possible a small test case ?
Filed bug 1009: http://bugzilla.openedhand.com/show_bug.cgi?id=1009
Thanks,
Jason.
--
To unsubscribe send a mail to [EMAIL
On Tue, 2008-07-01 at 17:56 +0100, Matthew Allum wrote:
We hope to get clutter_texture_set_from_yuv_data() fixed up in some form
for Clutter 0.8. Its the last remaining niggle really. Patches warmly
welcomed.
I do hope to look into it, although it looks like implementing it
properly might
On Tue, 2008-07-01 at 14:31 -0400, Kashyap Ashwin wrote:
with RGBA seems to be impossible. Does any one have any ideas on how
this can be done without memcopy and without multi-texture?
I'm only familiar with using fragment shaders to convert YV12 with
multiple textures. I'm not sure how this
On Tue, 2008-07-01 at 23:27 +0100, Matthew Allum wrote:
I've never noticed high CPU usage with
clutter_texture_set_from_rgb_data(). Have you tried profiling it to
check if it really is set_from_rgb_data and if it is, if anything bad is
happening below in COGL ?
I had profiled it insofar as
[Resending to list again; sigh, clutter is the only list I use which doesn't
rewrite Reply-To :)]
On Wed, 2008-07-02 at 10:13 +0100, Neil Roberts wrote:
Clutter happily supports 3 bytes per pixel if by that you mean 1 byte
for each of the R, G and B components and no alpha channel. You just
Matthew Allum wrote on 02/07/08 05:14 AM:
Could you try adding 'g_object_set (tex, disable-slicing, TRUE, NULL);
to the test and see if that makes any difference for you.
No change here.
I asked someone to test this on his system using Intel GMA for me, and
he says that the actual time spent
James Ketrenos wrote on 02/07/08 01:21 PM:
By setting LIBGL_ALWAYS_INDIRECT=true I am able to run trunk's
tests/pixmap-test TFP code path without a crash, although it doesn't
update the actor with the redirected window contents...
Indeed, that does seem to be the case for my tester as well.
Hi,
Before filing a bug, I thought I'd fly this by others on the list to see
if either I'm doing something wrong, or nobody but me is seeing this
issue.
I'm applying a simple behaviour to a ClutterGLXTexturePixmap and I'm
seeing many skipped frames in the timeline. However, the redirected
On Wed, 2008-07-02 at 21:09 -0400, Jason Tackaberry wrote:
seeing many skipped frames in the timeline. However, the redirected
window is a video, and no frames seem to be dropped on the video itself,
it's just the animation that is choppy. This suggests to me it's not a
bottleneck in the GPU
Matthew Allum wrote on 03/07/08 09:06 AM:
How fast are you running the actual timeline ? My guess is its just
CPU/X overhead from both the decoding and constant rebinding of texture
pixmaps (for each frame) that is sucking cycles from the timeline
scheduling. The texture updates probably
On Thu, 2008-07-10 at 01:15 +0100, Matthew Allum wrote:
Doh, TFP is way too painful.
Yes but it's terribly useful. Please don't give up on it. :)
Some day video drivers won't suck. Well, we can all hope.
Cheers,
Jason.
--
To unsubscribe send a mail to [EMAIL PROTECTED]
On Sat, 2008-07-12 at 00:39 +0100, Matthew Allum wrote:
Hopefully its pretty much working everywhere now nearly as best it can.
I think the code/API could probably do with a little re-factoring at
some point but that can wait for now.
Unless you changed something, I suspect we still have that
On Sun, 2008-07-13 at 00:04 +0100, Matthew Allum wrote:
You should file a bug with a test case showing the problem. Im not sure
Will do.
if its strictly a TFP issue but maybe more a general performance or
scheduling issue and maybe not even in Clutter. Have you tried running
clutter with
Is anyone working on this? Are patches welcome? I don't want to
duplicate any efforts.
Cheers,
Jason.
--
To unsubscribe send a mail to [EMAIL PROTECTED]
On Wed, 2008-07-16 at 13:24 +0100, Emmanuele Bassi wrote:
I'm going to resync trunk with my changes. after that: patches
welcome :-)
Excellent, thanks Emmanuele. I'll hold off until your commit.
Cheers,
Jason.
--
To unsubscribe send a mail to [EMAIL PROTECTED]
On Wed, 2008-07-30 at 18:50 +0200, Florent wrote:
Nope, i deactivated it (if settings Appearance settings to None does the job).
I ran into tearing issues as well on one of my development systems, also
after having disabled compiz. I found that actually restarting the X
server (and not loading
On Sun, 2008-08-03 at 15:38 -0400, Jason Tackaberry wrote:
The attached patch adds ClutterX11TexturePixmap and
ClutterGLXTexturePixmap support to pyclutter.
Hi Emmanuele,
At risk of sounding like I'm nagging, I wanted to make sure this patch
didn't slip through the cracks and that you'd have
On Tue, 2008-08-05 at 16:15 +0100, Emmanuele Bassi wrote:
using python on win32 to call the x11 API - and vice versa. I can
prevent people to compile it on different platforms, by #ifdef'ing
everything out using the defines Clutter provides, or using the
clutter-${platform} pkg-config file.
Hi,
I spent the night trying to track down a very bizarre memory corruption
issue and I've narrowed it down to this bit of code in
clutter/__init__.py:
# fixes weird linker bugs on nvidia
try:
import dl
#sys.setdlopenflags(dl.RTLD_LAZY|dl.RTLD_GLOBAL)
On Thu, 2008-08-14 at 23:07 -0400, Jason Tackaberry wrote:
issue and I've narrowed it down to this bit of code in
clutter/__init__.py:
# fixes weird linker bugs on nvidia
try:
import dl
#sys.setdlopenflags(dl.RTLD_LAZY|dl.RTLD_GLOBAL)
except
On Sun, 2008-09-07 at 15:18 +0200, Mike Massonnet wrote:
I had definitely say no. IMHO, you keep your default stage, and uses
something like DBUS so that you can display specific messages onto the
default stage.
I wonder if it'd be possible to use tfp and redirect the XComposite
overlay
On Thu, 2008-09-25 at 09:16 +0100, Emmanuele Bassi wrote:
threading and python are two very different beasts, with different
interactions at different levels. python is using user-level threads in
its API, as far as I know, and this is already not entirely well
behaving with the concept of
On Fri, 2008-10-03 at 19:58 +0100, Emmanuele Bassi wrote:
using python on embedded devices in my experiene isn't such a good idea,
especially for long running tasks, given the tendency of python 2.6 to
leak like a victorian pipe.
Don't you mean 2.5? The arena-freeing obmalloc landed in 2.5.
On Thu, 2008-10-16 at 15:06 +0800, HASWANI HARISH-VCKR47 wrote:
We have got a png file (attached) and I am also attaching the app I
have written along with gles+egl lib from vincent.
Attaching a 1MB file to an email sent on a mailing list a serious faux
pas. Your previous 100KB attachment
On Thu, 2008-10-16 at 19:21 +0100, Neil Roberts wrote:
cogl_tex = hash(tex.get_property(cogl_texture))
Ah, this is handy, I didn't know you could do this. (That is, get the
pointer to the underlying C object with pyclutter.)
Is it similarly possible to use ctypes to create a ClutterActor
On Sun, 2008-11-02 at 20:39 +, Emmanuele Bassi wrote:
you appear not to have read the documentation, or Jussi's reply: each
Timeline will drop frames to keep the length of the animation constant -
be that animation set using the number of frames or the duration in
milliseconds.
This
On Sun, 2009-01-18 at 10:32 -0500, Jason Tackaberry wrote:
3. If not, is there some alternative approach, equivalent in
functionality, for which a patch would be accepted?
I discovered one possible solution: attach two handlers with
clutter_threads_add_idle_full(), one at priority
On Sun, 2009-01-18 at 21:31 +, Emmanuele Bassi wrote:
let me see if I understand what you're trying to achieve: you want to be
able to make Clutter's redraw cycle depend on another set of conditions
-- basically, slaving Clutter's redraw to another redraw cycle.
What I'm trying to do can
On Tue, 2009-01-20 at 20:39 +, Emmanuele Bassi wrote:
set the :ellipsize property to TRUE; this will ellipsize the text if it
goes beyond the allocated width.
A nice feature (which we implement ourselves directly with cairo) would
be to have the text fade out to fully transparent, as
On Wed, 2009-01-21 at 08:52 +, Emmanuele Bassi wrote:
ugh. I'd argue on the usability of that...
Sure, arguments could be made either way, but depending on the use-case.
Of course what I propose would be horrible in some situations.
The primary use-case for text fading to transparent in
On Wed, 2009-01-21 at 09:48 -0500, Danick Desjardins wrote:
To solve this problem I access the pango layout directly and reduce the
font size until it fits perfectly given a width in pixels before
rendering text.
We're solving this problem by using a ClutterCairo texture, and using
cairo to
On Thu, 2009-01-22 at 16:52 +0800, Rob Kramer wrote:
When playing with the video-player in clutter-gst/examples, I noticed the
texture is not updated during the period that the sample effect ('e' key)
animates. Is this a bug/incomplete feature, or is it expected behaviour?
I'm completely
Hi Julien,
On Thu, 2009-01-22 at 20:03 +0100, Julien Pauty wrote:
One benefit of the idle call is that clutter drawing is done in the
clutter main thread. I've remarked that, on some driver, even putting
clutter_thread_enter/leave does not prevent crashes. Making sure that
everything happens
On Thu, 2009-01-22 at 20:47 +0100, Bastian Winkler wrote:
However, videos might have tearing effects when vblanking is disabled,
Which is even worse than wrong frame timing, IMHO. :)
Nice approach. Maybe I'll this too. Is there already some example code
in kaa/freevo?
Not yet. There's some
On Thu, 2009-01-22 at 22:03 +0100, Bastian Winkler wrote:
OK, there are still some things missing in GStreamer, but for DVD menus
resindvd [1] works quite well for me and the deinterlace2 [2] plugin
also made some progress. My main reasons for using GStreamer is the
stable API (instead others
On Sat, 2009-01-31 at 14:02 +0200, Tero Saarni wrote:
with blank screen. If I change the value that libswscale writes to
alpha channel I finally seem to get the picture. By default it sets
zero as alpha value. Do they disagree on the definition of the value?
Any ideas why?
Yes; I
Hi,
I know that ASM shaders are not directly supported with clutter, but
does anyone have a sense if (and how) it's possible to apply ASM shaders
to clutter textures?
For example, multi-texturing isn't natively supported yet either, but
with a decent knowledge of clutter's architecture, it's
On Tue, 2009-11-10 at 14:23 +0100, Matej Kupljen wrote:
Is there a way to get window ID so I can pass it to MPlayer?
Any other option?
Another option would be to use ClutterGLXTexturePixmap to redirect the
MPlayer window into a clutter texture.
This works nicely on nvidia, at least.
Jason.
[Readding the clutter mailing list in case others might find the code
snippet helpful.]
On Wed, 2009-11-11 at 09:55 -0700, Kevin DeKorte wrote:
Do you have an example for this?
Nothing really self contained. I have a vast glob of horrid prototype
code for a clutter-based video renderer which
Hi,
I'm currently using ClutterGLXTexturePixmap to redirect video. Rather
than having updates from the redirected window be rendered immediately,
I'd like to instead buffer 4 or 5 frames into separate COGL textures,
and then replace the underlying COGL texture of a ClutterTexture using
my own
49 matches
Mail list logo
