Hey guys, I’d like to regenerate the contents of a TextureRenderer in a non-OpenGL thread and though this works sometimes it is somewhat of a hit-and-miss. Apart from generating a BufferedImage in another thread and then applying to a new TextureRenderer in the OpenGL thread, is there another route I can go to keep the regeneration load off the OGL thread? The reason I don’t want to create a BufferedImage and then apply it is because of memory restrictions for large textures because at creation time there will be two very large BufferedImages (the one I created and the TextureRenderer backing image).
Ah, I finally figured it out. Like an idiot I wasn’t calling “markDirty” after changing and re-using TextureRenderer (when possible) instead of creating a new one each time also helped a lot. Now all of my updates are quite smooth.