Not to undermine the work, which is really impressive, but also as someone that's reverse engineered a lot of different proprietary model formats, they all foll

Not to undermine the work, which is really impressive, but also as someone that's reverse engineered a lot of different proprietary model formats, they all follow a very similar structure. You could do it too if you knew You have your vertex buffer data, your index buffer data telling you how vertices are connected, and a list of commands telling you where in the index buffer to start/stop. It might take you a bit, but after a while you notice the telltale signs of model data (e.g. 0x3F800000 being 1.0 in floating point), and you can usually make a first pass at it pretty quickly. It's definitely a lot of fun if you haven't done it.