Loading... |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
This post was updated on May 21, 2011; 9:54pm.
It seems that not all jogl commands works in applets.
What work good in a local application works in some cases also in an applet. Skyboxes to get working in an applet seems to be a nightmare. I have 5 different skyboxes written, as Quads, as Cubemaps ....., that all worls perfect local in an application, but no one does it's work in an applet. The problem was not with normal objects, also shader does work correct in an applet, only the skyboxes does not work. (SkySpheres works correct) Can it be that the textures excedes the browsers capacities for memory or so? I don't know. Has somewhere have such problems in the past, and can give hints, what goes wrong? |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Administrator
|
What happens when you use your application with Java Web Start? Do you reproduce this bug?
Julien Gouesse | Personal blog | Website
|
Loading... |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No, with java webstart, all runs correct.
I have testet it with compressed and uncompressed textures. In applets, only a little texture artifact was seen. The rest from the cubemap textures are somewhere ?????? This problems exist only with cubemaps in an applet. |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Administrator
|
Can you provide a small test case?
Julien Gouesse | Personal blog | Website
|
Loading... |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
For a little test case, the library is to complex.
I can show you the calllist. Works very good in a local environment, but not in applets. There is no change between the code in the local and the applet environment. Also texture size, compressed and uncompressed are also the same. I guess, the problem is the coordination with more textures in applets, that does not correct work. A simple SkySphere with one texture works perfect, also all objects that requires only one textures does its work. But SkyBoxes created with 6 Quads or the CUBEMAP call works not in an applet, but in a normal local environment. Here a little overview list that works correct in a normal local environment (in java): // with GL2 gl // int vboIds[] = new int[2]; // relates to // gl.glGenBuffers( vboIds.length, vboIds, 0 ); int glName = gl.glGenLists(1); gl.glNewList( glName, GL2.GL_COMPILE); gl.glDisable(GL2.GL_DEPTH_TEST); gl.glCullFace(GL2.GL_FRONT); gl.glEnable(GL2.GL_TEXTURE_CUBE_MAP); gl.glEnableClientState(GL2.GL_VERTEX_ARRAY); gl.glBindBuffer( GL2.GL_ARRAY_BUFFER, vboIds[0] ); gl.glVertexPointer(3, GL2.GL_FLOAT, 0, 0 ); gl.glEnableClientState(GL2.GL_TEXTURE_COORD_ARRAY); gl.glBindBuffer( GL2.GL_ARRAY_BUFFER, vboIds[1] ); gl.glTexCoordPointer(3, GL2.GL_FLOAT, 0, 0 ); gl.glDrawArrays( GL2.GL_QUADS,0, 24 ); gl.glDisableClientState(GL2.GL_VERTEX_ARRAY); gl.glDisableClientState(GL2.GL_TEXTURE_COORD_ARRAY); gl.glCullFace(GL2.GL_BACK); gl.glDisable(GL2.GL_TEXTURE_CUBE_MAP); gl.glEnable(GL2.GL_DEPTH_TEST); gl.glEndList(); My assumption is, in applets the coordination for textures in the GPU works not correct anymore, or the texture coordinates are destroyed. |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Administrator
|
I think that your assumption might be correct. Do you reproduce this bug with Chrome? I don't like this browser but I had to look in its source code. The problem is that the browser should use its own OpenGL context, it should not cause problems to applications executed inside the browser. Maybe Sven could help you better than me.
Julien Gouesse | Personal blog | Website
|
Loading... |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Sorry, i must correct me.
The problem exists in all browsers. Tested with Mozilla/FireFox, IE, Opera and Chrome |
Free forum by Nabble | Edit this page |