Le 17/01/2013 10:18, Klemens Schmiady a écrit :
Am 10.01.2013 18:38, schrieb Jack:
Le 10/01/2013 16:22, Max a écrit :
Am 10.01.2013 um 12:34 schrieb Jack jack@rybn.org:
Le 10/01/2013 11:57, Klemens Schmiady a écrit :
Am 10.01.2013 11:40, schrieb Cyrille Henry:
hello,
i can't reproduce this. is there anything special to do in the patch? cheers c
Le 10/01/2013 11:06, Klemens Schmiady a écrit : > Hey there, > > the sphere in combination with the lighting leads to strange > artefacts as shown in the .jpg. > Looking for a solution.
I think, this is because your viewpoint come too close to your geo (sphere). Put a [max 2] (for example) on your Z axis for the viewpoint to avoid collision.
i don't think it's that. i simplified the patch a bit more:
with GEM: ver: 0.93.3 GEM: compiled: Nov 10 2011 i have the same phenomenon here.
Yep, you are right Max. I think it was a problem with the near clipping plane of the frustum. But i was wrong... Cyrille, your patch seems to have the same problem if you put an alpha = 0.5, or i am wrong ? Here a patch that enable culling. Hoping it helps to solve _this_ problem. ++
Jack
PS : dont't forget to start by rendering the light before geos with an argument in [gemhead] ;)
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
I implemented your patch into what I'm working on at the moment but the light goes all crazy and I don't know anything about openGL. The lightsource should be in the center of the patch somehow. Can you please help me with that!
Thanks,
Klemens
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Hello Klemens,
Can you tell us your configuration (OS, Graphic card, Pd version and Gem version) ? Have you put an _argument_ in your [gemhead] ? See example (in Gem help) if you didn't put it. You have to render lights before the first sphere. And the first sphere before the second sphere. In fact, in your patch you can't know what is render first because you don't have this argument. See attached patch if it works for you. ++
Jack