The watery surface example is actually a solid mesh. I regret not showing it better in the video, but it’s actually a displaced cube that happens to be slightly larger than the Cornell box so that it appears as a displaced plane. As I said in the video, I have had a bit of success playing with the X vector curve when using just a displaced plane, but it requires some more testing and I didn’t want to put my name on it until I was sure. I’ll play with it some more tomorrow and see what I can come up with. Might take a stab at reflective caustics as well while I’m at it, but I don’t want to make any promises there 
Anyway, I’m glad you enjoyed the video. I was starting to worry that no one was interested!