These two aims are diametrically opposed.
Compare performance per watt, P4, to Centrino, to M3 for example.
These two aims are diametrically opposed.
Compare performance per watt, P4, to Centrino, to M3 for example.
It hasn't been true for servers either, as reflected by the resale price of old server hardware. It turns out power over a long time frame dominates over the manufacturing costs. From what I've seen, the argument is just bad math and bad assumptions all the way down at best. At worst it is willful ignorance in service of validating their assumptions regardless of the truth.
Yeah, servers are far more power-efficient than they used to be, but that's not really what millions of households worldwide are constantly buying.
Here, since I didn't have any links/quotes initially, I figured I should spend the time to dig some up.
https://doi.org/10.1016/j.jclepro.2011.03.004 > The manufacturing phase represents 62–70% of total primary energy of manufacturing and operation.
https://web.mit.edu/2.813/www/readings/Williams%20-%20Energy... > life cycle energy use of a computer is dominated by production (81%) as opposed to operation (19%).