Notch's post assumes they are using voxel point clouds. They could be, since they have yet to release any information as to how they are doing what they are doing, but they also could be doing something completely different. Notch also fails to mention that you can compress point cloud data.
Regardless of whether or not the video shows actual progress on a potentially revolutionary change in graphics technology, the fact people are working on it is great. We need a breakthrough in either how we process graphics or in how we design video cards, because we can only make chips so small.
