Ali-RS

btw, @jseinturier I tested the `AWTComponentAppStateSample` on my Linux and it works fine.

Not directly related to this, but we had a similar issue reported about shadow disappearing with instancing (https://hub.jmonkeyengine.org/t/shadow-disapeared-on-instanced-objects-away-from-camera/46174) in which I ended up adding `InstanceCullingFunction` (https://github.com/jMonkeyEngine/jmonkeyengine/pull/1865) but I never provided...

Just a note, afaik `controlUpdate` method is called always even if not inside the camera view, it is the `controlRender` method that is not called when outside of the camera....

Pinging @tlf30 just in case, as he is the author of the extension.

>Could also be that the model definition is broken. Don't know yet. By the way, you can check this in an online gltf validator/viewer. https://github.khronos.org/glTF-Validator/ https://gltf-viewer.donmccurdy.com/ https://sandbox.babylonjs.com/

>Regarding the colors, in blender the lights are set to 0 watt. I think in that case we should not return NaN but ColorRgba(0,0,0,0) . So, are you saying blender...

>Radius is set to Infinity It seems you should set the range in the "Distance" field, not in the "Radius" field. See https://github.com/KhronosGroup/glTF-Blender-IO/issues/618

>but when in blender you set light power to 0 jme translates that to ColorRGBA(NaN,NaN,NaN,NaN) It seems that is because of Math.log(0) when intensity/lumens is 0: https://github.com/jMonkeyEngine/jmonkeyengine/blob/52568d76ffec063355bcff515066cb75887baf34/jme3-plugins/src/gltf/java/com/jme3/scene/plugins/gltf/LightsPunctualExtensionLoader.java#L250 Probably we should...

>When passing data to the shader the luminosity gets applied and passed as final value to the shader. Is the math simply a `color * intensity` or it must proceed...

Are you using assets from test data? I can not find probe-day.j3o and textures used in marble_01.j3m.