"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."
That's similar to the method defense contractors use, they spread out their manufacturing locations to have influence over as many politicians as possible.
I'm a space nerd and the geographic distribution of the Apollo contractors was downright impressive. They built major systems from New York to California.
This is an interesting point I never thought of before. Now I wonder if the tech giants would have faired better politically if they weren't all co-located in one of the most politically polarizing areas of the country.
Honestly? They should be spread out. There is no reason for FAANGS to be located in mega campuses in the Bay Area. Let people who eg graduate from Midwestern colleges have the option of working in the Midwest (or returning there), pay them equally well. This would create so much economic activity in areas that really need it instead of simply enriching the landlords in the Bay.
It is a part of the reason why all the big government contractors are spread out all over the place. Even around a single metro area they'll have different offices in different congressional districts.
If you are on the winning end of that bloat life is pure happiness and joy. But if you are on the losing end (ie. not employed/a beneficiary of these firms) then you are paying dearly for the joy that others have.
I would think a lot of people benefit from the motivation for companies to geographically diversify. A lot of the bad things about the Bay Area boil down to people willing to put up with a whole lot because of a group delusion that a certain kind of thing can only happen there.
>I would think a lot of people benefit from the motivation for companies to geographically diversify.
Yes like I said there is the "in group". Is that "a lot" of people, I wouldn't say so. Namely the people who are employed or benefit financially from these orgs. If you are paying for the bloat via the government which is the customer of these defense orgs, it is effectively a tax on everyone else.
It’s not a delusion. While it’s true that you can start a business anywhere, being in the vicinity of people who have a ton of experience is very useful. There’s a reason why people move to LA to make it in entertainment or to NYC to make it in finance.
That said, for a non startup technology job, yes, no reason to be in the Bay. That work can be done from anywhere.
> 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?
A "full node" step is 1/2 the area of the previous. Since area is dimension-squared, 28 x 28 == ~800 area, while 20 x 20 == ~400 area.
So 28 -> 20 is a "full node" decrease (ie: twice the transistors).
---------
We can see the full-node step from 14nm -> 10nm as well: 14 x 14 == ~200, 10x10 == 100. So its a full node step. The next full-node is 7 x 7 == ~50, which is half of 100. After 7 comes 5, because 25 is half of 50.
The next full node is sqrt(12.5) or 3.5, smack dab between 3 and 4, so not really easy to round.
---------
The steps in between are "half-node" decrements, where you didn't quite achieve a full 1/2 area reduction. An incremental technology that represents something in between the full node step progression.
That's an interesting explanation and somehow I never heard it explained that way before. It makes sense if one thinks of the transistor count doubling relative to where that company previously was, and not where the industry is. In the transistor density plot each curve is roughly x^2, but with different initial conditions.
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.
So, I live in Columbus Ohio and I’m friends with two of the top people in our economic development department the broker this deal.
I can’t speak well to why not Cleveland but I can say that the actual city that they chose (New Albany, OH, 25 min to downtown and 15-20 to the airport) was built by Les Wexner, founder of L brands/Victoria’s Secret, is now home to data centers for Facebook, Google, and an AWS data center (there are two more within 30 minutes), and the general metro is one of the fastest growing regions in the US.
There is also a ton of available land on the fringes of the Columbus region and the sheer scope of acreage needed is bonkers.
I grew up near there. I guess I wouldn't quibble with the idea that Wexner "built" New Albany -- since it definitely isn't the same anymore -- but it was a town before he came along (two roads and a farm supply store, basically).
I used to go to summer camp at the old New Albany HS, and got my pickup truck stuck in the mud there many, many times while doing some odd job or another.
It has an international airport, intermodal rail freight depot and interstate highways.
I would guess that raw materials can be transported by road or rail. Access to a waterway is sometimes needed for very heavy indivisible parts (not sure if this is commons for fabs) but that is unlikely to be a regular occurrence. Just drive it down the interstate with a police escort at 2am.
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."
>Amid the recent chip shortage, triggered by the pandemic’s economic shock waves, ASML’s products have become central to a geopolitical struggle between the US and China, with Washington making it a high priority to block China's access to the machines. The US government has successfully pressured the Dutch not to grant the export licenses needed to send the machines to China, and ASML says it has shipped none to the country.
> needed for very heavy indivisible parts (not sure if this is commons for fabs)
It is not common. Semiconductor equipment is generally designed to be sent via air freight and assembled on site.
Some of the supporting operations (water and air purification plants, on-site chemical production, LN2 production, etc) may require some large parts but that's a case-by-case basis and probably avoidable.
> Does anyone know why Intel would want to build a "mega-site" in the city of Columbus Ohio?
My guess would be access to talent and costs. Columbus is more than 2x bigger than Cleveland. Building cars is also an international affair and we see those plants all over the place.
Columbus metro area is roughly the same size as Cleveland metro area so I don't think that is the reason.
However, it could still have a talent advantage. Having OSU nearby is helpful, and maybe it is easier to attract talent to move to a city with a big university.
Columbus grew 15% in the last decade, Cleveland shrank 6%. That could also have something to do with as there's reason to believe it will continue to attract skilled talent.
New Albany is a suburb of the capital of the third most active manufacturing state in the US. Intel is courting the federal government for direct semiconductor industry support that goes beyond tax breaks on semiconductor fabrication equipment.
> Edit: On second thought, Arizona doesn't have port access either so I guess it's not really a significant consideration.
From a pure logistics perspective, I'd say that Arizona beats Ohio. BNSF connects Arizona with both the Pacific Ocean and the Gulf of Mexico. (Not to mention Lake Michigan.) It's impossible to drive Route 66 from California to Texas without rolling alongside extensive trains hauling cargo through the desert.
For me, the trains are one of the most scenic aspects of that drive.
LAX is the eighth-busiest air cargo terminal in the world, and it has plenty of cargo flights to Asia where computer devices are often assembled. PHX is also one of the busiest cargo airports in the United States, albeit much less so than LAX or CVG.
Fabs have massive physical plant inputs (eg. lithography machines), substantial commoditized manufacturing inputs (eg. boules/wafers, freshwater, industrial gas), and core outputs with small size and high value (ie. chips). The former two can be pipelined without knowing the exact product-by-product breakdown of customer demand, and the latter cannot.
Infrastructure and labor concerns might also tip the scales one way or another. Water supply, wastewater management, energy cost, grid resiliency, labor supply, access to institutions for professional training, and other considerations can differ wildly between the two regions.
"Without identifying Intel, JobsOhio sent a request out to its regional economic development directors asking for potential sites that meet the company’s parameters. They had just three days to respond. One Columbus was the only regional partner to respond with a potential location: Jersey Township in western Licking County, near New Albany" https://www.dispatch.com/in-depth/business/2022/01/21/how-mi...
Port of Cleveland is not a big container port like you'd imagine using for consumer goods. It's for supporting industry, mostly steel, through moving commodities. Think a big open barge full of coal, and holding facilities for limestone or iron ore. There's probably a limit to the size of the sort of ship can navigate to the port of Cleveland as well, you definitely can't fit a huge cargo ship in that port.
Yep, the constraints of the St. Lawrence Seaway mean we get the cutest little container vessels every couple of weeks from Europe. Sometimes they have a portion of the deck set aside for windmill parts.
Presumably this plant will be trucking containers to Baltimore's port or sending air freight out of the cargo-only airport nearby.
It's hard to understate just how driveable this side of the country is from Ohio. In about 7 hours one direction you are in Chicago. 7 hours another direction you are in Toronto. 7 hours another direction you are in NYC, or DC, or Boston. Stretch it to 15 hours or so and you can drive all the way to New Orleans, or most of Florida in that time, and anywhere in between. Anywhere east of the Mississippi really is seemingly doable for a road trip in about a days drive, especially if you are driving in shifts. You can get flights pretty fast as well, but the issue is the ohio airports have limited direct connections; its frequent to fly to ohare or charlotte first. It's not like LAX where you can score a direct flight to half the western US from southwest for $59.
Highways do back up in Cleveland and Columbus in the cities during rush hour though, as well as when there is any construction or accidents or inclimate weather.
One thing I noticed is that the freeways in, say, Cleveland, are all 5 lanes wide or so in each direction, serving a population of 400k in the city and 1.2m in the county. You go to LA and what do you see? The same size freeways as Cleveland, 5 lanes wide or so, only its serving a city of 4 million and a county of 15 million. It's like, of course there is no congestion in Cleveland and tons of congestion in LA, it doesn't take a rocket scientist to see an issue with capacity. There's literally an order of magnitude more people using the same capacity of infrastructure.
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.
> 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.
Don't know what you are talking about. Real-time continuous facial recognition at the >98% level is used for about 5.000 crowds already, and they are just ramping up their infrastructures to detect 100.000 simultaneously. A football stadium. Only budgets are the problem, not tech.
Esp. with US government "services", but also Europe and Asia.
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.
The manufacturing pipeline is generally pretty long too, and you can stop wafers midflight if you're just going to throw them out anyway. So for instance, if you have some B0 wafers that have enough metal layers that you couldn't easily change them into B1 wafers half way through the process, you might end up with it being cheaper to end up with unusable wafers for plaques and what have you. The fab is generally happy to have the extra capacity even for tail end of the design. There's generally some cheap customers that'll slot in wherever there's gaps.
"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.”"
> 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.”
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.
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.
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."