jReality - the forum

Any plans for dual depth peeling? I've tried tricks with rendering order that sometimes work, but to work right with multiple cameras, it seems ordering the items in the scene-graph before rendering is not always going to work. It seems that the order of rendering of polygons needs to be different depending on the point of view of the observer.

Has any work been done on this? It seems like a really cool trick but if it's implemented I don't know how to turn it on in JReality.

Dual depth peeling is not implemented in jReality as far as I know. We looked at a "depth peeling" algorithm from an Nvidia white paper some years ago but nothing came of it. Please provide any algorithm/implementation details/references you can; that will improve the chances that someone begins to think seriously about doing something in this direction. It would be good to have correct transparency in the JOGL backend. And of course we welcome participation by developers not part of the TU team:)

There is an example application that does depth peeling in the jReality-svn.

On the screenshot you can see the depth layers that are being peeled from back to front and the image as it is drawn one layer at a time. Darker colors mean larger depth.

For Faces depth peeling is now part of the jogl3-backend (download via svn or the upcoming release).

What if you carried on the consistent shell depth to the low toms?

You could have 14 x 8 and 16 x 8, would there be a point where the timbre trails off?