Started pulling out the camera from the game side into the rendering layer. Not sure if this is a smart move so far. I’m doing this for making it possible to render the game in stereoscopic 3D. But also making it way easier to add light field rendering. Don’t know if it’s the right terminology, but instead of rendering the left and right eye. It would render multiple views from one list of draw calls. These multiple views can then be used for a more immersive 3D experience. A special display or headset can then show you the game from multiple angles. But it also enables for other viewers to see it in 3D as well. Some displays even enable actual real world depth of field through this type of rendering. So you would be able to see the car in focus or move your eyes to the trees and see those in-focus.
All pretty amazing stuff. Though of course not many people have these kinds of displays available to them. I see it as a good exercise for me and I want to gain some experience in these types of rendering. Of course the Nintendo 3DS is stereoscopic 3D, so that is the main goal here. Using the same list of draw calls would be need and probably necessary. Though I’ll just need to find a good balance as to what parts should be only on the game side. And what parts can bleed through into the rendering layer or even further, into the platform layer.
I sort of finished pulling out the camera from the game side. It’s definitely not the final code yet, but it’s a start. Especially the two views aren’t clipped correctly. And the area that gets paged in and updated should probably enlarged as well. You can already see the trees pop in and out because their dimensions are not correctly considered. But hey, there is actual stereoscopic rendering. Pretty cool to test some stuff this way 🙂