[osg-users] OSG-3.2.1-rc1 + Qt-5.2.0 building
Hi! I'm try to build OSG-3.2.1-RC1 +Qt-5.2.0. I use mingw-w64 toolchain with GCC-4.8.2. I apply patch from Kristofer Tingdahl before build and now got another errors: http://pastebin.com/Vz3xNnLU Regards, Alexey. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osg::MatrixTransform delays
ok. thanks. as I mentioned in my other post, it is a black box with plugin architecture allowing to plug osg code here and there, so it is not flexible. I tried everything and I think the particles are updated prior to the matrix that moves the entity, so it is always a bit ahead when it moves. I got confirmed by their support people that they have their own schema on updating models, so it is definitely not happening in the update traversal. Merry Christmas to you as well :-) Nick On Tue, Dec 24, 2013 at 3:24 PM, Robert Osfield wrote: > On 24 December 2013 11:54, Trajce Nikolov NICK < > trajce.nikolov.n...@gmail.com> wrote: > >> when the particle system is frozen, then it is moving without delays. >> Otherwise I see still delays. Robert, any comment on this issue. It is sort >> of big for me >> > > It is almost certainly an order of operations issue, but can't say what > exactly you are doing wrong or what to do about it. I would have commented > already if I had any constructive ideas. > > > Merry Christmas :-) > Robert. > > ___ > osg-users mailing list > osg-users@lists.openscenegraph.org > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > > -- trajce nikolov nick ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osg::MatrixTransform delays
On 24 December 2013 11:54, Trajce Nikolov NICK < trajce.nikolov.n...@gmail.com> wrote: > when the particle system is frozen, then it is moving without delays. > Otherwise I see still delays. Robert, any comment on this issue. It is sort > of big for me > It is almost certainly an order of operations issue, but can't say what exactly you are doing wrong or what to do about it. I would have commented already if I had any constructive ideas. Merry Christmas :-) Robert. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osg::MatrixTransform delays
Hi Sebastian,
when the particle system is frozen, then it is moving without delays.
Otherwise I see still delays. Robert, any comment on this issue. It is sort
of big for me
Thanks a bunch!
Nick
On Fri, Dec 20, 2013 at 1:30 PM, Sebastian Messerschmidt <
sebastian.messerschm...@gmx.de> wrote:
> Hi Nick,
>
> just one idea: Can you replace the particle system with something else. So
> maybe it is strictly particle system related. (I'm assuming, that you use
> osgParticle here)
>
> Hi Sebastian,
>
> You are right, that is the structure and the first thing I tried was to
> attach it directly to the body but then I saw this artifact like the
> particles following the model with delay. So I tried the update callback.
> Now I am thinking to have the particles attached to the main camera so they
> move with the view but here is again update callbacks and such.
>
> Thanks anyway !
>
> Nick
>
>
> On Thu, Dec 19, 2013 at 9:35 PM, Sebastian Messerschmidt <
> sebastian.messerschm...@gmx.de> wrote:
>
>> Am 19.12.2013 12:09, schrieb Trajce Nikolov NICK:
>>
>> Sebastian, it was too early to be happy. Changing it to traverse first
>> then update didn't helped actually. On the windshield model I have attached
>> ParticleSystem that is actually delayed with the windshield. Any other
>> ideas?
>>
>> I also tried to attach this directly to parts of the model but same
>> results.
>>
>> It is about rain effect. This particle system is firing rain drops on
>> the windshield geometry and it suppose to move with the car model, however
>> it is delayed as it moves.
>>
>> Sorry, I'm out of ideas here too. If I got you correctly you have some
>> group which represents a "body" and another one representing the windshield
>> with the particlesystem attached.
>> And you are updating the later with the update-callback.
>> I don't understand why the two groups must be disjoint. Can you explain
>> the reason for this? Are those two different render passes?
>> If you explain your structure a bit more in depth I maybe can point to an
>> alternative solution.
>>
>> cheers
>> Sebastian
>>
>>
>> Thanks
>>
>> Nick
>>
>>
>> On Thu, Dec 19, 2013 at 12:02 PM, Sebastian Messerschmidt <
>> sebastian.messerschm...@gmx.de> wrote:
>>
>>> You're welcome
>>>
>>> Thanks Sebastian. That was it. It did helped!
>>>
>>> Nick
>>>
>>>
>>> On Thu, Dec 19, 2013 at 11:53 AM, Trajce Nikolov NICK <
>>> trajce.nikolov.n...@gmail.com> wrote:
>>>
Hi Sebastian,
that is actually a good idea to do the traversal first. Let me try it
first. Thanks !
Nick
On Thu, Dec 19, 2013 at 11:51 AM, Sebastian Messerschmidt <
sebastian.messerschm...@gmx.de> wrote:
> Am 19.12.2013 09:26, schrieb Trajce Nikolov NICK:
>
> Hi Nick,
>
> My first question would be: Why don't you simply attach the windshield
> to the model ;-)
> No seriously, I had those effects too, when retrieving e.g. the
> ModelView matrix in callbacks. The problem usually was gone whe doing the
> traverse first and the get the matrices.
> Maybe you can give it a try, or paste some of your update-callback
> code here, so we can help.
>
> So basically this:
> void CameraUpdateCallback::operator()( osg::Node* node,
> osg::NodeVisitor* nv )
> {
> osgUtil::CullVisitor* cv = dynamic_cast(nv);
> if (!cv)
> {
> return;
> }
> traverse(node,nv);
>
> osg::Camera& camera = *cv->getRenderInfo().getView()->getCamera();
> camera.getViewMatrix()
> }
>
> will yield different results than this:
>
> void CameraUpdateCallback::operator()( osg::Node* node,
> osg::NodeVisitor* nv )
> {
> osgUtil::CullVisitor* cv = dynamic_cast(nv);
> if (!cv)
> {
> return;
> }
>
>
> osg::Camera& camera = *cv->getRenderInfo().getView()->getCamera();
> camera.getViewMatrix()
>
> traverse(node,nv);
> }
>
> This at least solved problems in multipass-rendering, where multiple
> cameras had to be synced relative to each other.
>
> Hope this helps.
>
> cheers
> Sebastian
>
> Hi Community,
>
> I have a simple scene of a car model with a windshield (two separate
> models combined in runtime). There is osg::MatrixTransform node on top of
> the car that moves it around, and the windshield is attached to the scene
> that also has osg::MatrixTransform on top of it with an UpdateCallback
> that
> copies the car's matrix.
>
> ( the real scene is a bit different but this is the concept)
>
> I am seeing the windshild has some delay in following the car model
> while moving and at lower framerates. I think I just tried everything but
> always this is the same case
>
> Any clue and hints?
>
> Thanks a bunch!
>
> Nick
>
> -
Re: [osg-users] osg rain effect on multiple channels
Hi Robert, after examining all the options it turned to be the visual system (which is a commercial black box OSG based) doing this somehow. When I attach the osgParticle::PrecipitationEffecte somewhere else in the scenegraph it show to work. So sorry for the false report again. And thanks! Happy Holidays Nick On Mon, Dec 23, 2013 at 4:46 PM, Trajce Nikolov NICK < trajce.nikolov.n...@gmail.com> wrote: > Hi again Robert, > > It has to do something with the ClipNode (mimiced code from the > osgprecipitation example). On one of the channels the rain is less dense > then on the other two, I managed to get some rain visible by reducing the > clip distance on the ClipPlane. > > Nick > > > On Mon, Dec 23, 2013 at 1:43 PM, Trajce Nikolov NICK < > trajce.nikolov.n...@gmail.com> wrote: > >> Hi Robert, >> >> machine per channel, the view matrices are set to blend on the edges. I >> will try your hint and let you know >> >> Nick >> >> >> On Mon, Dec 23, 2013 at 12:47 PM, Robert Osfield < >> robert.osfi...@gmail.com> wrote: >> >>> Hi Nick, >>> >>> On 23 December 2013 07:41, Trajce Nikolov NICK < >>> trajce.nikolov.n...@gmail.com> wrote: >>> I tried the default osgParticle::PrecipitationEffect accross three channels (left,center,right) with the default settings and the observation is that is shows ok on the center and right channels, however there is no effect on the left channels - same code only the view and projection matrix different. Any clue why? >>> >>> I don't have any ideas why this might happen. How are you setting up >>> the different channels? Different machines? Sample machine different >>> graphics contexts? >>> >>> What happens if you change the view matrix of the left channel to be >>> nearer to the centre direction? >>> >>> Robert. >>> >>> >>> >>> >>> ___ >>> osg-users mailing list >>> osg-users@lists.openscenegraph.org >>> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org >>> >>> >> >> >> -- >> trajce nikolov nick >> > > > > -- > trajce nikolov nick > -- trajce nikolov nick ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
