picking

Some more love ..

JGO Picking Discussion
  <http://www.java-gaming.org/topics/picking-techniques/31301/view/topicseen.html>

OpenGL 'picking'
  <http://www.opengl.org/archives/resources/faq/technical/selection.htm>

Raycasting
  <http://antongerdelan.net/opengl/raycasting.html>

  Note: We do have/support:
  - AABBox
  - PMVMatrix incl.
    - inversion,
    - gluUnProject,
    - Frustum extraction
  - Frustum (planes)
  - ..
  - See com.jogamp.opengl.math.**,
 

I admit I didn't read the whole discussion, however let me leave just a link, that I found very useful about picking

http://ogldev.atspace.co.uk/www/tutorial29/tutorial29.html

We are right now emigrating from color picking to this new technique (with an hashmap of short-s)

This technique is less weak than color picking but it still uses glReadPixels and it looks like a bit the other slow technique I quoted ("Picking with an OpenGL hack").

To be honest Sven Gothel I'm worried,

You say I haven't "earned" the replies I got by Julien (gouessej I suppose). And that his replies to me are a token of "love".

And then you say,

"Don't assume evil forces are driving us when you don't understand.
Simply assume the best is given to you."

Have you lost your mind completely! This is typical sect behaviour! I'm glad I caught this before I committed to JOGL. I better understand now why Oracle will have nothing to do with you.

I asked a simple question about porting a picking solution from Direct3D to OpenGL. It's Julien who has refused to leave it at that and I have only reluctantly commented on his (outlandish) claims in order to be polite.

Julien doesn't want JOGL because it would be like reinventing the wheel. No, he wants scenegraphs, especially good ones like the one's he's promoting himself.

Hi

Please keep cool, everything wasn't intended to be read in the literal meaning. Oracle still has some things to do with our APIs, 2 of its employees used them during the latest Java One, I fixed a bug in Netbeans Platform, a maintainer fixed another bug I reported, there's no war even though there are some disagreements between us (about Java3D, JavaFX, OpenJDK, scenegraphs in general, Direct3D, deployment, security, ARM Hotspot JIT compiler, ...).

We gave you several indications even though they were in the middle of things you might consider as "off topic" and I quoted several examples and a paper to confirm all my claims. I have nothing against a good implementation of picking in shaders based on JOGL if it is really worth it. Moreover, I remind you that nobody answered you on the forum of JogAmp main competitor, our replies are better than nothing. Please stop making wrong suppositions about what I say, I already explained here in details that I'm not promoting a particular scenegraph. I spent several days in looking for a paper about a good implementation of picking in shaders and I found one as I said here, what else do you want? If you behave like a troll, you'll be treated like a troll. We respect you, I justify my claims, I make some researches each time I have a doubt, I spend some time in answering you but if you don't respect us, don't ask for help and stop making us waste our time as you don't take into consideration all our indications about OpenGL implementation details useful for your shader based picking (tessellation shaders, transform feedback, discard rasterization, primitive id).

Best regards.

On 01/11/2014 10:51 PM, Sven Gothel wrote: Now we have picking impl. - easy to use in general.

Add generic support for picking via raycast intersection and AABBox (or similar)
   <http://jogamp.org/git/?p=jogl.git;a=commit;h=8352cb7c5d8cc971c87f13fe9f61e346c4d2c541>

Demo ..
  <http://jogamp.org/git/?p=jogl.git;a=commit;h=5e728baa72517865d602580b920d9bdfdfb26c65>

~Sven