qsmi
Joined 4 years ago
Karma
56
Submissions
18
Member since
Dec 29
Recent submissions
Comment · 4 years ago
Comment · 4 years ago
I am pretty shocked but I guess ASML delivers by 747. [1][2] But they do seem to know what they're doing. :)
"The current generation of EUV machines are already, to put it bluntly, kind of bonkers. Each one is roughly the size of a bus and costs $150 million. It contains 100,000 parts and 2 kilometers of cabling. Shipping the components requires 40 freight containers, three cargo planes, and 20 trucks."
[1] https://technical-news.net/euv-lithography-asml-delivers-100... [2] https://www.wired.com/story/asml-extreme-ultraviolet-lithogr...
Comment · 4 years ago
Increasing supply chain resiliency by having multiple, geographically separated, sites this more of a thing now.
https://hbr.org/2020/09/global-supply-chains-in-a-post-pande...
"The obvious way to address heavy dependence on one medium-risk or high-risk source (a single factory, supplier, or region) is to add more sources in locations not vulnerable to the same risks."
Comment · 4 years ago
> node names are entirely unrelated to any physical dimensions
I agree they're not necessarily related to any particular transistor dimension but are you sure it's entirely unrelated to any physical dimensions? If so, here is a question I have. 28nm, 22nm, 20nm, 14nm, 10nm, 7nm, 5nm, 3nm. What progression is that? Why skip 6nm and 4nm?
Comment · 4 years ago
Does anyone know why Intel would want to build a "mega-site" in the city of Columbus Ohio? Why not choose Cleveland Ohio where one has port access with an existing route to Europe? Fabs are international affairs, no matter where they're rooted, because just to keep the place running one needs a constant stream of parts from everywhere. It seems like being in a sea/rail/truck hub would be a logistics advantage.
[1] https://en.wikipedia.org/wiki/Port_of_Cleveland
Edit: On second thought, Arizona doesn't have port access either so I guess it's not really a significant consideration.
Comment · 4 years ago
If this analysis is even half correct those selling x86 server CPUs need to be concerned. Today there is a huge moat in existing x86 software, sure, but savings like that can justify many ports.
https://www.anandtech.com/show/15578/cloud-clash-amazon-grav...
"Cost Analysis - An x86 Massacre"
Comment · 4 years ago
The comment was in the context of a San Francisco City Ordnance, so the question is, does the City of San Francisco use Real-time continuous facial recognition? Today the answer is no, but that's likely to change; therefore, any forward-looking ordnance should take it as a given.
Comment · 4 years ago
> The ordinance basically presumes a context of continuous, 24 hour surveillance or access thereto.
I definitely agree with everything that you wrote but real-time continuous facial recognition is going to be here sooner rather than later. In the context of reforming this ordinance, it makes sense to just assume it's available. To me anyway.
Comment · 4 years ago
It depends, is the wafer passing or failing the manufacturing tests? This is an important consideration.
Comment · 4 years ago
Probably. From a follow on article,
"The biggest factor is heat. “Higher speeds tends to produce higher temperatures and temperature is the biggest killer,” says Rita Horner, senior product marketing manager for 3D-IC at Synopsys. “Temperature exacerbates electron migration. The expected life can exponentially change from a tiny delta in temperature.”"
https://semiengineering.com/aging-problems-at-5nm-and-below/
Comment · 4 years ago
> I’d much rather we over-design stuff to last decades, at least at the chip level where overedesigning is super cheap.
Engineers are working on it.
From a follow on article by the same author.
"An emerging alternative is to build aging sensors into the chip. “There are sensors, which usually contain a timing loop, and they will warn you when it takes longer for the electrons to go around a loop,” says Arteris IP’s Shuler. “There is also a concept called canary cells, where these are meant to die prematurely compared to a standard transistor. This can tell you that aging is impacting the chip. What you are trying to do is to get predictive information that the chip is going to die. In some cases, they are taking the information from those sensors, getting that off chip, throwing it into big database and running AI algorithms to try to do predictive work.”
https://semiengineering.com/aging-problems-at-5nm-and-below/
Comment · 4 years ago
I suppose depending on how extreme we're going with unfocusing on math.
But one doesn't really get to ignore math in life because of money. You'll still need to figure some things out. I can say I'm a poor painter so I'm just going to ignore that and things will be fine. The cost of ignoring math can be quite high.
Comment · 4 years ago
Definitely.
I think when people say they suck at math what they mean is; I find math homework extremely boring and unrewarding. Which perpetuates sucking at math and only compounds. Then things get worse as modern society works better for the individual when one does not totally suck a math.
Kinda like, I suck at piano because I find practicing the piano boring and unrewarding, but without the ramifications of sucking at math.
Comment · 4 years ago
Just to expand on this, when I worked as a grader in college I noted on engineering test keys the professor's answers were always roughly three short lines of handwritten equation. But the student's answer section were often totally darken with pencil they wrote so much. The heuristic was definitely the longer the answer the more wrong it was.
This is something I've taken to heart professionally too. If the equations are getting out of hand it's either wrong or I need some simplifying assumptions (like that cow needs to be spherical). Otherwise one just can't keep track of it and reason about it.
If I need a more precise answer, that's what computer numerics is for.
https://www.techcenturion.com/7nm-10nm-14nm-fabrication