You don't need price control, there is going to be competition for the extra 1000$ that's for sure, but its the customers choice where they spend the extra money.
He’s basically showing data oriented design, where you, try to limit cpu cache misses by operating on the data.
This approach can be way faster, but is only relevant when you have a lot of entities you need to iterate over. If you have 3-100 objects it would of course still be faster but by negligible amount
I played a little with FFT Gaussian blur. It uses the frequency domain, and so does not have to average hundreds of points, but rather transforms the image and the blur kernel into the frequency domain. There it performs a pointwise multiplication and transforms the image back. It's way faster than the direct convolution.
The takeaway here is not “Divine intervention” it's either: A lot of people fall off boats and die while taking a leak, or there are so many people on the ocean, that if you go in a straight line, you're probably going to meet someone.