I want to create some GUI programmatically (it’s dynamicaly created based on a xml file) but I has to be pixel perfect (so I guess EZGUI is not the answer) and I wan’t to avoid the excessive draw calls (so I guess iGUI is not the answer too) and all the memory fragmentation created by the excessive “new rect” calls required by the Unity GUI.
Any advice?
I’m doing the same thing, right now. I’m using the 2 camera method, where one camera is orthographic, looking at the GUI and rendered on top of the main camera (Set the camera depth of your GUI higher than your main camera to do that, and set your camera’s depth pretty low). For the part of my GUI I’m populating via XML, I have every option I’m going to want on a single sprite sheet, and I’m offsetting the texture as I need to change it, and I have my offset x and y logged in the XML (along with some other stuff).
Hopefully that gives you some ideas.