Previous | Next --- Slide 41 of 46
Back to Lecture Thumbnails

It seems that it's not the Moore's law really predicts the technology front, but the hardware manufacturers are trying to match the "prediction"?


Yes, Moore was one of the cofounders of Intel and his law was basically a self-fulfilling prophecy because Intel (and other chip manufacturers) tried to match the predicted growth in order to stay on top of the competition.

As of late (last 5 years or so), I've seen quite a few sources claim that Intel is no longer trying to follow Moore's law due to physical limitations.


Do you think we will ever reach a point where super computers are considered fast enough, like for all tractable problems the super computer would be fast enough. I personally think you can split computation problems into 3 classes currently impractical to solve at scale(traveling sales man np complete stuff), doable but slow, or faster then it needs to be. I understand the faster the better and that energy usage could go down. You could also get more things done if the computer is faster. But really if your weather simulation completes in 1minute vs 2minutes does not have a huge effect on the world. Assuming the computer become much much faster.

TLDR; if computers were 10^9 times faster would that be fast enough