Moore's Law has been around for a while now, and we haven't run out of things to do with the power yet. I'm reasonably sure that improvements in AI, rendering,

Moore's Law has been around for a while now, and we haven't run out of things to do with the power yet. I'm reasonably sure that improvements in AI, rendering, image processing and similar things will continue to absorb processing power for some time to come. Hell, on the rendering front, I can probably spend a few hours researching and tell you precisely which techniques will become usable with 2, 3 and 5 years' improvement at current rates. F'r example: true multi-bounce real-time ray-tracing through "Brave"-level hair and fur. Looks awesome. Is, erm, some way off being possible computationally speaking. Or, for that matter, the sort of hair physics simulation they used on the same movie - to the best of my knowledge, also Not Yet Doable In Realtime (and some way off).