Hi Mathieu,
you are probably using the older osgLeap-0.3.0 which had such problems.
Please have a look at osgLeap-0.5.0 which I just tagged.
Thank you!
Cheers,
Johannes
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=58788#58788
Hi,
my company has an exhibition booth at AIX this year and I will be there, too,
showing our VR Framework including osgLeap Integration.
Anybody interested can find me and the Leap next week at the AIX 2014 in
Hamburg, Germany. Booth of P3 Voith Aerospace, Hall 6D, Booth 6D60-7.3.
I will be
Hi Mathieu,
nice to hear you find osgLeap useful.
Please make a zip of all osgLeap files you checked out including all changed. I
will make a diff to check for changes.
What version/revision did you check out and what were the difficulties you had?
Thank you!
Cheers,
Johannes
Hi Robert,
thanks for your answer. I agree with you for most applications.
Now the thing is, we are currently adding a simple text-based Shader Editor in
our application. So while writing the shader the developer will change uniform
names almost for certain. Now the developer will not
Hi,
in our application we just created an user interface for setting up
osg::Uniform data on our nodes for 'artist usage' (setting shader input values
like colors or textures).
Now I encountered that the uniform name must not be changed which is from a
artist user perspective quite
Hi,
osgLeap-0.3.0 is tagged now.
* osgLeap-0.3.0 requires at least OpenSceneGraph?-3.1.4 including build of the
osgdb_png plugin to read the PNG images files for the osgLeap::HandState? class
* Introduced osgLeap::PointerEventDevice? to generate native OpenSceneGraph?
events from your Leap
Hi,
osgLeap v0.1.2 is tagged now. Changes are:
* Build support for g++ on Linux (tested on g++ 4.6 on Xubuntu 12.04)
* Introduced singlehanded and trackball single handed manipulator modes
* Introduced compatiblity mode for LeapSDK 0.8.0 (in CMake options please
enable
Hi,
running out of tracking volume is just a matter of correct positioning and
training. I've made a small video for demonstration.
http://www.youtube.com/watch?v=0wBiiKo_hys
Thank you!
Cheers,
Johannes[/url]
--
Read this topic online here:
Hi,
I find it still way too fiddly for my taste - the rotations are fairly
hard to do consistently. It is easy to perform a rotation, but to
perform one that actually puts the object into the orientation I want
(instead of some semi-random one) is hard. I didn't realize that one
has to
Jan Ciger wrote:
I have tried again with the visualizer tool and I can get some
reasonable hand rotation - if the fingers are visible. They obviously
use some Kalman filter or something to keep the rotation smooth and
going even if the fingers disappear for a moment. However, tracking of
a
Hi,
in the SVN trunk I've now added a Mode setting to the LeapManipulator for
switching between the Two-Handed mode you can see in the video and the new
simpler single-handed mode which allows ROTATE ZOOM, only (just like in the
Cyber-Science Motion App in Airspace).
Cheers,
Johannes
Hi Jan,
today I installed both MSVC++ 2010 and MSVC++ 2012 and I had the include
directory missing as well. The include fix is now patch and committed.
osgLeap now compiles on MSVC++ 2008, 2010 and 2012 (x86). Building x64 is
tested on 2008, only.
I would appreciate anybody confirming x64
Jan Ciger wrote:
x64 works for me on 2012. Well, works - as much as the Leap itself works, the
device is really underwhelming after all that hype.
Actually that's true it gets kind of hyped. From what I learned by now the
tracking is not that stable and reliable as one in an industrial
Hi,
i've just started a SVN repo for the osgLeap Leap Motion Integration.
The current code delivers a CameraManipulator derived from
osg::OrbitManipulator called osgLeap::LeapManipulator.
As there is no doc, yet, here is some basic info how to use.
To ensure stable detection, I've implemented
Hi,
now here is the URL to the repository at Google Code:
https://code.google.com/p/osgleap/
Best regards,
Johannes
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=55473#55473
___
osg-users
Hi Jan,
thanks for your input. I will have a look at this within the next days.
Best regards,
Johannes
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=55475#55475
___
osg-users mailing list
Hi,
my development environment is MSVC++ 2008 Express. Compiler support for 2010 to
2012 and GCC is planned, but that are topics for the next days.
Jan, could you please send you CMakeCache.txt so that I may review your
settings? Thanks in advance!
Cheers,
Johannes
--
Read
Hi,
I've made quite good progress on a OSG Leap Motion Integration for use as
CameraManipulator, I call it osgLeap by now.
In a few days I will release the code publicly.
Thank you!
Cheers,
Johannes
--
Read this topic online here:
Hi,
I just got my leap. any progress on this topic?
Thank you!
Cheers,
Johannes
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=55396#55396
___
osg-users mailing list
Hi,
It might be interesting other models seems to be saved and loaded without
problems into OSGT including texture data. I've tested it with the OSG-Data
clock.osgt, for example. But actually I didn't yet get the difference.
Thank you!
Cheers,
Johannes
--
Read this topic
Hi Sebastian,
I get your points. The situation we come from is:
We have Image data that is produced during application runtime that never
existed on disk separately. So our osg::Image has no filename defined. So we
decided to inline the data rather than pushing them to disk.
When writing our
Hi guys,
some time ago I discoverd that inlining osg::Image data into the native
OpenSceneGraph serializer formats is possible for the binary format (OSGB),
only.
Now that we also need to support inling osg::Image data in the .osgt format, I
am curious to ask why this is currently not
Hi,
Now that I had another look into the osgCompute repository I see that
END_BRACKET stuff was already patched a day before my post, so you better stick
to the osgCompute trunk instead of using the version I posted last week.
Thank you!
Cheers,
Johannes
---
Johannes
additional render to textures):
NVidia 1 ms Draw + 9 ms GPU - ~100 fps
Intel HD 4000 ~ 30ms Draw + ~ 30ms GPU - ~16 fps
Fixed OpenGL pipeline enabled:
NVidia 1 ms Draw + 12 ms GPU - ~100 fps
Intel HD 4000 ~20ms Draw + ~20ms GPU - ~25 fps
Best regards,
Johannes
---
Johannes
::Resources::instance()-getImageResource(:/some-label/my_picture.jpg);
Would that be a nice add-on to OSG or is it just a stupid idea because there's
already a cool way to do this?
Thank you!
Cheers,
Johannes
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH
Hi Sebastian,
great, that's it. I love bugs that are already fixed.
Thank you!
Cheers,
Johannes
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH, Bremen, Germany
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php
just don't get it where we get the dependency on any NVidia driver
library.
Plase, do you have any suggestions or experienced anything like that?
Thank you!
Cheers,
Johannes
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH, Bremen, Germany
- Renders as expected.
So what might be the difference?
Thank you!
Cheers,
Johannes
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH, Bremen, Germany
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=51885#51885
Hi,
Jesus. I'm now compiling any OSG release from 3.0.1 to trunk to see where the
error occurs first.
Hopefully tomorrow I'll find something.
Meanwhile I'd be glad for recommendations...
Thank you!
Cheers,
Johannes
---
Johannes Scholz
Software Engineer
P3 Voith
Chris Hanson wrote:
You're just using the AlphaPixel 3rdparty library binaries then?
Yes, Alphapixel only.
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH, Bremen, Germany
--
Read this topic online here:
http://forum.openscenegraph.org
if the crash bug is gone.
Thank you!
Cheers,
Johannes[/code]
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH, Bremen, Germany
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=51903#51903
]
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH, Bremen, Germany
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=49487#49487
___
osg-users mailing list
osg-users
the 0x.
Best regards
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=49471#49471
___
osg-users mailing list
osg
make the dummy osgt File
convert correctly, but then again the 0x's are missing.
didnt have the Time to go further, i have to continue tomorrow.
Thank you!
Cheers,
Johannes
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH
--
Read this topic
Hi,
jup. If i have the time possibly tomorrow.
Thank you!
Cheers,
Johannes
---
Johannes Scholz
Software Engineer
P3 Voith Aerospace GmbH
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=49476#49476
Hi,
well, even if it's now over a year of your last posts here are some points that
might help others.
We had a similar problem when using the Cull Mask for showing and hiding
objects.
If you set the Cull Mask a camera using
setCullMask(NODEMASK_SHOW);
then you also have to tell the
Hi Jens,
thanks for your reply.
Currently we DO NOT set those flags before realize() because using the
osgCompute functionality is only a sub-case of the application (only needed
when the user whats to deform a geometry).
So if the user wants to start deformation osgCompute is initialized.
Hi,
we got a problem with the osgGeometryDemo example.
This problem occurs on one developer machine only. The first devmachine works
without problems - as far as I can see with the same software setup, but
different hardware.
Problematic machine:
NVidia GeForce 9400GT (G96b, 16 cuda cores,
Hi,
argh. That did the trick. I just didnt check this on that machine.
Installing the osg sample data in our OSG_FILE_PATH solved it.
Thanks a lot!
Cheers,
Johannes
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=41605#41605
Finally I found it working, when manually assigning the context:
Code:
framecnt_++;
if (framecnt_ == 2) {
// osgCompute
setThreadingModel(osgViewer::CompositeViewer::SingleThreaded);
setReleaseContextAtEndOfFrameHint(false);
Hi,
I am adding an osgCompute::Computation node during runtime inside the
updateTraversal of our viewer.
This works fine when added in the first updateTraversal frame, but fails in all
later frames. What am I misunderstanding oder missing? Any hints?
Below the code:
Code:
void
Thanks a lot!
Cheers,
Johannes
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=37598#37598
___
osg-users mailing list
osg-users@lists.openscenegraph.org
Hi,
today I encountered a problem using osgText::Text3D.
When displaying the String Viewports using a osgText::Text3D, the display of
the text gets messed up. Other strings tend to work.
Viewports appears as iewports V.
Problem occurs even in the example code. Just changed
Hi,
you could also use OpenCASCADE (www.opencascade.org). It's open source and able
to read IGES/igs STEP/stp.
Using the functions of OpenCASCADE you may read a IGES/STEP file (and also some
other formats) using eg. STEPControl_Reader class.
The resulting faces can be triangulated using the
Hi,
as currently there are a lot of other things to do i didnt manage to look into
the details yet.
We set up a simple OSG-Application where we only have our Deformation Code and
a simple viewer. There we discovered the multi-threaded deformation working and
gaining performance on more cpus.
Hi,
looking through the implementation of Thread::setProcessorAffinity in
PThread.c++ I realized that on a NON-SGI platform the method always returns -1
even if pthread_setaffinity_np or sched_setaffinity might have been executed
correctly. Is this intended?
Cheers,
Johannes
Hi Robert, hi Ulrich,
we are using Ubuntu 9.04 64-bit on AMD Opterons.
And yes, we built OSG by ourself. We are using OSG release 2.8.3.
We'll check cmake and come up with further details later.
Thanks so far.
Best Regards,
Johanens
--
Read this topic online here:
Hi,
we are currently working on a routine that deforms large vertex arrays on the
fly. To speed this up we implemented multithreading using OpenThreads.
Simple example:
We have a point cloud of e.g. 10 million points. The calculation is done on
e.g. 4 processors.
The first thread is assigned
Ahh well, now I see.
Today I tested the code on our VR environment machine. As we start with stereo
enabled by default we already have the GL context as needed for quad buffering.
Now we can use the code to turn it off and on again.
The same code does not work when the vr scene is startet
Hi Sergey,
thanks for your hint to GImpact, I will look into it next week!
Hi Paul,
did you have time to try the solution from Sergey, yet?
Otherwise I would try it myself next week, when I'm back in the office.
Best regards and thanks for your help so far,
Johannes
--
Well, without diving into details, I also added the ability to switch on/off
stereo during runtime in my application. I didnt test it using quadbuffer so
far(!), but it works very well for cyan/red stereo. I didn't try using
osg::DisplaySettings::instance()
What I do is:
// Check if the
Hi Paul,
well, I just opened an issue at the osgWorks wiki.
Best regards,
Johannes
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=31407#31407
___
osg-users mailing list
To the stereo mode issue: This morning I had a quick look at the colladaread
example in osgBullet. I just set OSG_STEREO to ON and I am getting quite the
same picture in the colladaread example like in my environment. The distance
between the left and right eye images is wrong somehow
Hey guys,
I am currently integrating rigid body physics using osgBullet into our VR
environment, which is mainly used for viewing CAD models generated in CATIA V5.
Well, I got everything running using a separate physics thread and the
osgwTools::AbsoluteModelTransform like shown in the
54 matches
Mail list logo