Hi fellow devs,
I try to implement a simple 3D position picking. My idea was to use x and y screen coordinates, add the depth value and unproject the whole thing to end up with a Vector3 in my world.
The first part works quite good, the depth value more of a problem.
I need to get the depth value at my mouse position. So first question: How do I get it out of my RenderTarget2D? Is that possible without rendering a depth image?
What I am currently doing is:
I render a depth image into a separate RenderTarget2D and do:
var depths = new Vector4[ 1 ]; _DepthRenderTarget.GetData( 0, 0, new Rectangle( screenX, screenY, 1, 1 ), depths, 0, 1 ); return depths[ 0 ].X;
Unfortunately this returns a value thats falling the whole time until it reaches 0 and then start from 1 again. And I have no idea why this is happening.
Any ideas? Thanks for reading.
Edit: Nevermind. I was drawing the depth buffer and positioning an object at the calculated position. The depths of that object of course where taken into account in every new loop - causing the object to move towards the camera and decreasing the depth with every loop. Sorry Solved.