HEADLINE: IMEC Discusses 2nm; EDS Wraps; What It Takes to Restore Notre Dame de Paris
This is Brian Santo, EE Times Editor in Chief. You're listening to EETimes on Air.
BRIAN SANTO：我是EE Times主编Brian Santo，你正在收听的是EETimes全球联播。
This is your EE Times Weekly Briefing. Today is Friday, May 17th, and among the top stories this week:
l A sobering roadmap of semiconductor process technology-- potentially coming to a halt at 2 nanometers
l We were at the Electronic Distribution Show in Vegas last week. On the last day, the news broke that the United States had escalated a trade war with China. How will that affect the supply chain?
l New advanced imaging tools that could help restore Notre Dame in Paris, but a big question remains: Can anyone else afford them?
Later in the show, we’re joined by two graphics technology experts, Jon Peddie and Kathleen Maher. We asked them to imagine the restoration of Notre Dame.
稍后，有两位图形技术专家Jon Peddie和Kathleen Maher加入我们的讨论。我们请求他们来构想一下巴黎圣母院的重建。
All that to come, but first, here's Rick Merritt, EE Times Silicon Valley bureau chief, who’s not in California this week, but on the road in Belgium.
所有这一切都将逐一呈献，但首先有请EE Times驻硅谷记者站主任Rick Merritt。他本周不在加州，而是在赶往比利时的路上。
Rick traveled to Antwerp to cover IMEC’s FutureSummits, where the world's leading R&D hub is strutting its stuff. The $64,000 question there is how much further IMEC and its partners can push their nano technology innovations. Here's what Rick found out.
I had a chance to talk with researchers at Imec in their Leuven headquarters and at an annual event they host in Antwerp here. and they showed a very sobering road map for the semiconductor industry, basically creeping ahead at a few nanometers per node and smaller gate lengths and metal pitches for the next few nodes pretty much coming to a halt at what they called a 2nm node at about a 40nm gate length and 16nm metal pitch.
The good news is there’s a wealth of new transistor designs, nanosheets, things they call forksheets and ideally this future complementary FET, a kind of vertical FET, as well as a lot of new packaging techniques to mitigate the fact we're slowing down and shrinking and we're getting to a point where shrinking will stop. One of the most interesting techniques was a new to dis-integrate the SoC into separate blocks for cores and SRAM and power circuits that each have their own very separate needs.
Interestingly, too, the observation was each of the leading-edge companies — TSMC, Samsung and Intel-- will probably grab their own unique mix of this basketful of techniques. So, it's kind of every man for himself as we get to what could become the end of Moore’s law in a handful of nodes ahead.
So, at the event here, there were some 1,800 people they said coming to talk about a wide range of topics in electronics. It's not just about CMOS scaling. But a couple of things I picked up from talking to people here, which thought was interesting. One was — I’m still checking this out with the companies — TSMC’s 7+ node supposedly still requires some double patterning because they're only using EUV on four layers. So I'm still trying to find out more about that.
And I also was told Samsung is heavily discounting its foundry prices, especially to some startup companies, providing a full mask set for less than what TSMC charges for what they call a multi-layer mask set, a very kind of reduced mask set, which itself costs 60% less than a standard mask set.
So, there’s a lot of price competition going on there, and a lot of unknowns about how these companies are using EUV. So I'm still trying to dig into more about that. Maybe a new story next week. If you know something about it, give me a call or drop me an email.
This is Rick Merritt reporting from the Future Summits event in Antwerp.
Next up, a report from Las Vegas, where component manufacturers and distributors gathered for the annual EDS trade show. Barb Jorgensen, EPSNews managing editor, observed a wild week at the EDS show. She explains the cause of all the excitement.
Component manufacturers and electronics distributors attending the annual EDS Show and Conference were cautiously optimistic about the second half of 2019 as the show began on Tuesday, May 7.
That lasted through Thursday.
Attendees that shared their thoughts about the second half of the year did so with one major caveat: that the U.S. and China would resolve their trade war.
Just as the show was wrapping up and executives began to scatter, the Trump administration raised tariffs on Chinese imported goods and reported that trade negotiations stalled.
A lot of EDS conversations are on-the-fly, but I did get one takeaway: The industry is still skittish about the rest of the year.
The electronics supply chain, while supporting the goals of IP protection and a less restrictive Chinese market, have not been in favor of tariffs. On Wednesday, the Institute for Supply Management released its Spring 2019 Semiannual Economic Forecast. U.S. manufacturing executives expect their revenue will grow by 4 percent in 2019. At the same time, nearly 60 percent believe tariffs have raised the price of goods on average about 6.8 percent.
Nearly 30 percent believe tariffs have caused delays and disruptions in the supply chain.
At least one U.S. OEM had weighed in on the new situation by Friday. George Whittier, President and COO of The Morey Corporation, said "As an electrical engineering and manufacturing provider, we rely specifically on imported resistors, capacitors and printed circuit boards. So the tariffs, combined with an already limited supply, can affect procurement costs, product design and production timing. The impact was small at first but has increased as suppliers' inventories deplete, and we're forced to increase the amount we import."
EDS attendees said they still have reasons for optimism.
Electronic content in automobiles continues to increase; the advent of 5G will require a brand-new infrastructure both terrestrial and in the Cloud; the robotics industry consumes a lot of components while it increases efficiency; and automation in general continues to grow.
Michael Knight, President of the TTI Semiconductor Group and Senior Vice President of Corporate Business Development, predicted 2019 is going to be bumpy a ride, but he also saw no indication that a recession is imminent. Notably, Knight said a resolution to the U.S.-China trade war could catapult the industry to 2018-level component demand. 2018 was a really good year for their supply chain.
The electronics supply chain is, of course, global. And the Asia-Pacific region is the biggest concern right now for the industry. Distributors' book-to bill-ratios in the region have declined. Demand has softened and margins remain tight. Although unit consumption remains high, companies aren’t making a lot of money in Asia-Pacific.
Even before the latest round of tariffs, electronics companies were tentative about their forecasts. As one expert once told me, “The market hates uncertainty.” But that’s exactly what the electronics industry is going to be dealing with, at least for awhile longer.
Even after a month has gone by, some of us still shudder at the horrific image of the slow blaze that consumed the spire of Notre Dame in Paris. While EE Times is no Architectural Digest, we figured that a publication that has chronicled the fundamental transformation of electronics technologies should find out how a new generation of digital technologies might be used not only for the reconstruction of the cathedral, but in civil engineering and archeological projects all around the world.
After Notre Dame in Paris recently burned, journalists who cover technology were falling all over each other to talk about the nifty high-tech tools that are newly available to help fix the cathedral. And there are a lot of new tools like that.
I wrote one of those stories too; it’s called, “Architecture is Still Mired in the Mechanical Age.” Yes, new tools exist, but there’s a problem: Hardly anyone can afford them. Notre Dame received pledges of nearly $1 billion to rebuild, and that will probably help its renovators fund the acquisition of advanced imaging equipment. But almost nobody else sees anywhere near that kind of money.
One of the people I spoke to for my article is the interim CEO of the World Monuments Fund, which has been working to preserve the world’s greatest heritage sites for half a century. Lisa Ackerman was genuinely pleased to see the money pouring in to rebuild Notre Dame, but she said that if she'd had her druthers, she’d like to see similarly large donations dedicated to research and development of tools and techniques that could be used to help preserve all the world’s antiquities, not just that one devastated cathedral.
The lack of money to preserve old ruins is disheartening, but the inability to afford modern tools is hobbling civil engineering, too. And that’s just stupid. It’s well known our infrastructure is crumbling-- ironically, much faster than many of world antiquities are deteriorating. And we obviously need advanced digital tools for more inspections and more thorough inspections.
I spoke with Jerry Hajjar, a civil engineering professor at Northeastern. He’s trying to create imaging systems for unmanned inspections of bridges and towers.
Now it’s great that someone is doing that, but I’m wondering why it falls to some engineering prof and a revolving cast of grad students to invent it.
What? There’s not a big enough marketing opportunity for creating cheap tools for civil engineering? Hey, before you answer that, get in your car and drive over one of the older bridges where you live, a bridge I can all but guarantee hasn’t been properly maintained.
I recently talked to a company called Cepton that creates lidar sensors that it maintains are far less expensive than others available. Its customers are mounting its lidars in handheld devices and even on backpacks so that researchers can continuously scan their surroundings as they’re moving across terrain. It goes to show that advanced sensors are beginning to become available in portable form factors at lower prices.
The fact is, there is dire need-- not just some market demand, a need-- for more tools designed and priced for the underfunded civil engineering, renovation and preservation markets. If we’re all lucky, some of you listening out there will accept that challenge.
Up next, computer imaging experts John Peddie and Kathleen Maher talk about the challenge of restoring the cathedral of Notre Dame.
Well you know, one thing we've seen is the way that technology has come together so it feels like art and technology are now melded. So that they're not separate anymore. The field of CAD has evolved way past the idea of drawings on paper, and now involve laser scanning and metrology, precise measuring, digital prototyping, simulation. So by the time that a project is being built, we already know how it's going to perform. And that's really important for something as important as the cathedral in Notre Dame.
And that data also can then be fed to a 3D printer so you can build things that way.
Well, it's true. One of the things I envisioned is drones flying over the thing, understanding where the building has to start. And 3D printers climbing on the roof of the building. But all this, actually the project has already started before with the work of Andrew Tallon from Vassar, who was able to make a complete scan of the cathedral from top to bottom.
And that's an extraordinary thing that he did. And in fact, it's an international treasure. Because he was able to scan the entire interior, as well as the exterior, down to a resolution of 5 millimeters. So, we now have an extremely accurate data base of what that church was like just before the fire.
You know, and that's an important challenge in design and CAD, is not designing the building or not even being true to the design, it's understanding what the thing is as it exists in the world.
Right. And so that brings up an interesting point as to how should the church be reconstructed. Should it go back to its original design? Some people have proposed that. Or should it just be reconstructed to the way it was prior to the fire? Or should it be reimagined and taken to the future?
Well you know, it's not just a technological challenge. It's a cultural challenge. Because this is something that means something to everybody in the world, but especially to the people in Paris. And Edouard Phillipe has suggested a design competition to consider this. Maybe the thing to do is to make it look just like it did. That's a very comforting idea. But maybe it makes more sense to reenvision it.
What are some examples of that?
Well, one of the things I think to keep in mind when you're reenvisioning is that the decisions that were made 300 years ago, 150 years ago are not necessarily the decisions you would make now. There's better materials, and there might be better reasons for... Put solar panels on the roof. Make it more functional. Two of the designs that came through, there was one from a Belgian company, Myasees, and from Studio NAB, and both proposed glass ceilings that would enable gardens or agriculture in the sky.
I know! It is, isn't it? And rebuilding the spire. So the spire was an after-thought. It was added in the 19th Century by Eugene Viollet-le-Duc, who was attempting to restore the cathedral and introduced his own problems because of the materials they had then. They didn't have good concrete. And the spire, some people were not crazy about it at the time. Though it's a typically Parisian-looking spire. So now the idea is to reenvision that and make it an apiary.
Yeah, for the bees. Because there were bees in the roof, and there were the hives up there. And miraculously, the bees survived. So Studio NAB suggested turning it into an apiary.
For the bees. Well, we need all the bees we can get. That's for sure.
I know! Isn't that a nice idea?
Yeah. I'd also like to see the church reconstructed with more modern materials. Part of the reason for the damage was that there was so much wood in it. And if we could replace those slats that held up the roof with...
Exactly. We can reenvision it now using steel and titanium and carbon fiber.
Which would be lighter and stronger and fireproof. We can look at the Louvre as an example, because the Louvre broke from tradition and put a glass pyramid in the courtyard to bring more light into the museum. But it turned out that the glass pyramid became a design feature of the museum as well. So there's nothing to say that you have to hold with tradition.
No. That was controversial at the time, but it's a lovely way to tie the future and the past. And I think that's the opportunity we have now.
And so, we're looking forward to see what they come up with.
Yeah. Aren't we though?
That was John Peddie and Kathleen Maher, talking about what it might take to restore Notre Dame in Paris after the cathedral’s roof went up in flames.
This has been your weekly briefing from EETimes. You can read all these stories and more at EETimes.com. Thanks for listening.