Frame from Sophie Wilson - The Future of Microprocessors Plays with YouTube's privacy-enhanced embed

Sophie Wilson - The Future of Microprocessors

Open on YouTube First HN thread

2 HN users recommended this talk in 2 different months between 2016 and 2018.

2007–2026
2 recommendations 2 citations HackerTube score 2

HN commenters cite it for the claim that process scaling beyond 28nm is uneconomic and for dark silicon.

Sophie Wilson, designer of the ARM instruction set, discusses the future of microprocessors, including the economics of scaling below 28nm and why much of a chip must stay powered off.

Pull quote: “Highly recommend watching this video” (HN, 2016)

Also tagged: microprocessor scaling, sophie wilson, arm, moore's law, dark silicon, chip economics

What Hacker News said

4 labeled comments from the HN archive; up to 20 are kept per video, favoring different commenters. Each one links back to the original thread.

Recommends dkarapetyan December 2016 under This AI Boom Will Also Bust
Explains why most of a chip is powered off.
Moore's law has been dead for a while now. Most of the chip in your phone is powered off because otherwise it would burn up. Highly recommend watching this video: [this video]
Recommends gens May 2017 under The Race to 10/7nm
Explains cost of going below 14nm.
It is hard (read expensive) to go under 14nm. Sophie Wilson talks about it much better then i could. [this video]

Citations, criticism and jokes

Cites it qznc February 2017 under What happened to clockless computer chips?
Source for claim that scaling below 28nm is uneconomic.
> when Moore's Law hits physical limits? It already did. Sophie Wilson said [0] its 28nm for ever. Scaling further makes no economic sense (unless you really need the space, e.g. in Smartphones). [0] [this video]
Cites it zengid May 2018 under C Is Not a Low-Level Language
Wilson makes similar point about scalar imperative languages.
I agree that this article hits pretty hard against a lot of assumptions about how our machines are working. I also feel like the author is trying to say something about how imperative scalar (meaning 'operates on one datum at a time') languages are causing more trouble than they're worth. Sophie Wilson said something similar in her talk about the future of microprocessors [1]. This implies that declarative and functional semantics would be more amenable to parallelization, as the author mentions in the article, as well as allowing the compiler more freedom to deduce a suitable 'reordering' of operations that would better fit the memory access heuristics the machine is using. [1] [this video]

Keyboard shortcuts

/
Search
j k
Next / previous video in a list
Enter
Open the focused video
l
Like (on a video page)
w
Watch later (on a video page)
?
This help
Esc
Close