3D InCites Podcast

A Conversation About Reshoring Advanced Packaging in the U.S.

August 12, 2021 Francoise von Trapp, Bob Patti, Alan Huffman, Dick Otte Season 1 Episode 8
3D InCites Podcast
A Conversation About Reshoring Advanced Packaging in the U.S.
3D InCites Podcast +
Become a sustaining member and we'll give you shout out in an upcoming episode!
Starting at $3/month
Support
Show Notes Transcript

The acronyms involving funding for semiconductor manufacturing are flying around Washington. There is the Chips for America Act, focused on re-shoring, The Facilitating American-Built Semiconductors (FABS) Act that promises tax credits for investments in, either equipment or fabs. and then there’s the $250 billion US Innovation and Competition Act, which earmarks $52bn for new chip manufacturing and research. This is now being debated in the House of Representatives. There’s a lot to unpack here – what we’re focusing on today is how all of this could impact advanced packaging and test in the US – To talk about this, we invited 3D InCites Community Members. Bob Patti, of Nhanced Semiconductor, Alan Huffman, of Micross Components, and Dick Otte, of QP Technologies. 

 The conversation covers a wide range of questions: Will the $52Bn be enough? How should those funds be best allocated? What will the impact be on small entities? Where will the workforce to support it come from? How do we create a US supply chain? Are we politicizing chip manufacturing? How do we ensure that Foundry 2.0 adds IP value and attracts new talent? Is the answer a vertically integrated advanced packaging company?  Should we create a consortium of advanced packaging companies and use the government funding to fill in the gaps? 

 Tune in to hear the answers to these questions and much more. 

If you'd like to learn more about some of these 3D InCites Community Members, visit their profiles:

·      Nhanced Semiconductor 

·      Micross 

·      QP Technologies 

You can also contact the guest directly via email:

Bob Patti, Nhanced Semiconductor 

Allan Huffman, Micross 

Dick  Otte, QP Technologies

 

 

Support the show

Become a sustaining member!

Like what you hear? Follow us on LinkedIn and Twitter

Interested in reaching a qualified audience of microelectronics industry decision-makers? Invest in host-read advertisements, and promote your company in upcoming episodes. Contact Françoise von Trapp to learn more.

Interested in becoming a sponsor of the 3D InCites Podcast? Check out our 2023 Media Kit. Learn more about the 3D InCites Community and how you can become more involved.

Françoise von Trapp:

Hi there. I'm Françoise, and this is the 3D InCites podcast. The acronyms Involving funding for chip manufacturing are flying around Washington. There is the Chips for America Act focused on reshoring semiconductor manufacturing, the Facilitating American Built Semiconductors or FABS act, which promises tax credits for investments in either equipment or fabs. And then there's the$250 billion US innovation and competition act, which earmarks 52 billion for new chip manufacturing and research. This has passed through the Senate, but is now being debated in the House of Representatives. Now there's a lot to unpack here. What we're focusing on today is how all of this could impact advanced packaging and test in the U.S. So to talk about this, I invited 3D InCites, Community Members, Bob Patti of NHanced semiconductor, Alan Hoffman of Micross, and Dick Otte of QP Technologies. So I'm going to let them introduce themselves and tell us a little bit about their companies. So, Bob, let's start with you.

Bobb Patti:

Hi I'm Bob Patti, and I've been around the advanced packaging industry now for more than 20 years. I started in this area as part of Tezzaron and NHanced was a spin-out of Tezzaron in 2017. We took some of the engineering team and one of Tezzaron's fabs, and spun it into NHanced, which focuses on 2.5D and 3D advanced packaging. We basically apply semiconductor processes b ack e nd of line technologies to a dvance packaging. So we're in the very semiconductor-like parts of advanced packaging, and we're a small company. The fab is located in Raleigh, North Carolina, and we do a design and test in Naperville, Illinois.

Françoise von Trapp:

Okay, great. So you're more of working more on what was the backend of the front end moving into the advanced packaging side of things?

Bobb Patti:

Right. It's it's the direct bonded interconnect, the oxide-oxide bonding. We do custom back end of line work, basically anything that can be done in the metal stack we build interposers, uh, insert TSVs, do those kinds of things. It is probably the most rapidly growing area of advanced packaging today, at least in my opinion.

Françoise von Trapp:

Okay. And now, Alan how about you?

Alan Huffman:

I'm the director of engineering for Micross Components, advanced interconnect technology division. We're located... People are actually from Bob's facility in Raleigh, we're in Research Triangle Park. My background is a 26 year involvement primarily in advanced interconnect. I've worked in a lot of wafer level packaging over the years and over the course of the development of our 3D interconnect technologies as well here at Micross somewhat parallel, but a bit of a different direction than Nhanced- Bob's focus in his company. We still focus a lot on 2.5D and 3D integration, methodologies- chip packaging. We do a lot from the products and services division of Micross, servicing standard packaging as well. So chip and wire and molded packages and so forth. Micross Components also has a lot of competency in, front end, or sorry, backend advanced tests. And we have the largest die distribution business globally in the world. So, Micross as a company covers a lot of the different microelectronics industry in different ways. And my division in particular is supporting all of the advanced interconnect and packaging.

Françoise von Trapp:

Okay, awesome. Um, Dick, tell us a little bit about yourself and QP and Promax.

Dick Otte:

Yes. QP Technologies used to be called Quik Pak until recently when we rebranded it, as people say, it is one of two divisions in the parent corporation called Promex Industries. I personally am the CEO of that parent, and my office is in Promax, the division, which is located in Santa Clara and QP Technologies that does a lot of semiconductor packaging is located in Escondido, California in the San Diego region. The Santa Clara division specializes in medical and biotech devices that i nclude, that are physically small, that in the jargon of our industry today incorporate a lot of heterogeneous integration kinds of processes. So we do a lot of things that incorporate optics and MEMS devices, ultrasound, f luids, chemistries, all leader, or a very broad range of processes as unique to Santa Clara. And, yeah, I've been involved in, in these classes of technologies for many years before I became CEO of this company in 95, I was the president of a company called Advanced Packaging Systems. That was a joint venture between Corning lass and R ay Count that specialized in what in those days w as c alled a thin film, multi chip modules. So I've seen a long evolution of these various packaging techniques over the years.

Françoise von Trapp:

Okay. So you've been along for the ride as they say. So I sent you guys some questions to review, but I'm actually going to dive kind of into the middle of these...the Senate approved the$52 billion for chip manufacturing and research in June. And it's now in the House of Representatives. Since you're all in the advanced packaging space, I really want to talk about how this push for onshoring is going to impact you and your companies. Is the$52 billion enough?

Alan Huffman:

You could argue no amount of money is ever enough for the industry.

Bobb Patti:

I think, the 52 billion is a lot of money and what it'll come down to is, how it is spent and how it's focused. I think the value may be in more of the l ong t erm. I think the immediate influx of cash may or may not be ultimately worthwhile. A longer- term policy of stimulating the technology and bringing it on shore is probably better for the nation. You know, I, like the rest of us on this podcast, would really like to get some of that money. I certainly think that it is going to bring some additional fab capacity on shore that otherwise wouldn't occur. I've talked with several fabs that are busy lobbying already to get access to that funding when it is approved by Congress. I think for those of us who are further downstream and especially at the leading edge of technology, which has advanced packaging is... You know, we're hoping that we get some of it, but the lion's share of it is probably going to go to the larger semiconductor companies that are doing multi-billion dollar programs. Most of us on the phone call are much smaller entities, compared to an Intel or a GlobalFoundries or some of the other large fabs. We're unlikely to get, a large amount of money directly out of this, where I think, c e rtainly from my perspective, I'm hoping to be one of the fleas at the end of the dog's tail that does get some additional funding to help expand capacity. We recently announced that we are expanding the capacity in our facility, but it's more difficult, I think for the smaller entities, bec ause we don't have the concerted lobbying efforts, which I think are key to accessing a lot of the money that's available in this initiative. I think the FABS initiative, which is the tax credit, is probably something that will do small companies like NHanced, a m u ch better service. It's more valuable to us because it's something that we can count on without having to have a lobbying effort in Washington DC. Uh, I' m going to circle back to my initial point, which is the 52 billion is great. What would be better is a consistent policy that looked forward over a decade of how we're going to continue to grow and induce the industry to stay on shore. A one-time quick fix is, is not going to occur... If this is an indication of forward moving policy that the U.S. government recognizes the need to incentivize companies to be on shore. And that being the U.S. Is no longer just enough to bring high technology into the U S... This is a radical philosophical change for, I think the U S g o v e rnment, because the US government has always been of the opinion that they don't need to incentivize companies to come to the US... to do manufacturing in US...to do high technology in the U S because we're THE place. And what has happened is the rise of China, the continuous leak of high technology offshore has, I think really concerned the US government from a strategic, mili t ary... a world position that this has to change. I'm hoping that it becomes a long term change, not a one-off: here's a big pot of money. And my concern is, is it get s consumed by a small handful two years from now, everybody forgets and they go back to offshoring again.

Françoise von Trapp:

Wow. Okay. That's a lot. And thank you for setting up the conversation so nicely. Alan and Dick, do either of you have anything to add to those opening thoughts?

Dick Otte:

Yes. I would concur with sort of the view that it has to be a long-term thing. This cannot be, it will not, i t'll be a waste of money if it's a one-time thing. So the question gets to be where is those funds and how do you invest those funds in the long run to achieve what I'll call significant progress in the next, roughly three to five years. That would be my objective for this kind of an investment here. And the key issue, I think, is sort of that we're talking about h ere is packaging. And if we're going to support within the U.S. All of these semiconductor fabs that a re r eally already committed. A nd then we have to address, what are we going to do for even conventional packaging. The old BGA, what have you, and QFNs, who's going to fabricate these things for this, a half a cent, a lead kind of thing that the people out of Asia talk about and w ill produce. We will need that capability here and exactly how that's going to get put in places is a really tricky issue, u h, because companies like us are not particularly interested in investing in these old technologies c ause you're fighting every day, the economics against Asia. S o, hence even the discussion here, we're talking about. The easier thing to do is to make sure that we build a strong foundation in the emerging technologies for packaging, and that we really have a strong position and that we protect and defend that, and maybe subsidize it with the kind of funding that is being talked about. So that, that gets you into then where are the directions i n, what, how was that going to be done? And realistically, it does in fact, start with smaller companies like us that are traditionally the pioneers of these new d irections, like interposers and what have you, and then has to proliferate, to ultimately grow, to be bigger companies. And as the volumes t hat these things grow and t hey're more widely adopted-- because th e t hing we all know that ar e i n this industry is these trends in advanced packaging are not one year trends. You're talking five, 10 year kinds of things. My personal vi ew w h ere t he technologies to invest in today beyond just sort of supporting the classic packages, I see three directions first, there's this move to chiplets-- the move to chiplets in my estimation is a fundamental change in the manufacturing process for packaging it's new to the world. And if we get our act together here in the U S a nd build a substrates that are required, the interconnect technologies that are needed, continue to control the intellectual property, that's critical to making that kind of technology that works. We can build a real strength for ourselves. The second area that's emerging is the co-packaged optics- data rates are getting up into the multi terabit data rates- Its become clear that the get the information in and out of these die, we're going to have to go to rather complex optical technologies. And those are now just emerging. The U S is doing well. The AIM facility in SUNY New York, th a t was funded by the government six years ago now is making major contributions to that field. Those are several of the technologies that I see that fit into this category of the enhanced areas that we should be investing in, The biggest issue that I see we have as a country is not money. We got lots of money around here. The big problem we have, and that we're having is finding technical talent. The younger generation all wants to do things in software. And the reality is, is this enhanced packaging, and much of semiconductors is a physical hardware kind of thing. And that, for whatever reason, it's kin d of out of favor, if you will, as a career direction, we have to do something. I think, as an industry, as a culture, even to get more enthusiasm for the younger generation to move into this physical space. How about you, Alan?

Alan Huffman:

Yeah. I'll echo Bob and Dick's comments about the fact that this can't be a one-time only program it's, you know, it should be, I think, viewed as the beginning of a mega trend where the United States has a more relevant role to play across the entire microelectronics spectrum. You need to have- as we've kind of seen over the last year o f the pandemic- additional sources of technology and supply chain to even out this mentality that we've gone to, in at least the commercial world, about just in time delivery of things. And you don't have to look very much further than your email box every day. And the news story aggregation services that send things out to us to k now that all the foundries, the big major global Foundry players are lining up with their stories about how th ey're t hey can't do this And they can't do that. You know, that that's the that's, I think where a lot of this initial money is go ing t o g et invested. And I think a lot of that is being justified on the us l evel as a matter of national security, whether or not they're saying it or not. Because there is this concern about how the United States for aerospace and defense is going to continue to be able to acquire leading edge technologies and have access to the leading edge CMOS I spend my career doing this. Bob Dick had be en involved in this for a very long time, too. W e look at what happens on the other side of the Foundry and want to know how it is that we' re go ing to be able to push the advanced packaging rat ion as pects of this. And all three of our respective companies have been working in this field for 20 plus years, developing a lot of technologies, implementing them. And a lot of, a lot of what have generally been in my case, at least, u h, g o vernment sponsored research and development, kind of things put a lot of pieces of technology in place that are now ready to be moved forward to the next step of being able to support larger scale production. You're goi ng to sp end, you know,$5 billion to put a new leading edge fab facility on the ground. You can do that with a l e ading edge advanced packaging facility for a f r action of that cost. And there's going to be, It's going to be very interesting when, when these programs eventually get released and I think they will get released and funded where the propensity of, of our lawmakers is going to be to allocate funding away from the things that have the largest visibility in, i n the press, fro m, f rom the fou ndries an d to the smaller companies who are really going to be looking to put in place the next level of support for those advanced packaging capabilities. Those of us who've been working in packaging for a long time hav e wa tched the development of technologies that started, you know, in large part here in the United States, gradually moved to Asia, become very difficult to compete with from a cost perspective. And now we're talking about pulling all of that back here and creating that capability in the United States. There's going to have to be some way of allowing that to be cost competitive... the novelty of having those capabilities here in the US and available to a wider group of customers than they are right now I think it's a good starting point, but there is, there is this question about the sustainability of this going forward. If it's just the sa m e pr ogram, I completely agree that won't be the case. You will have to continue to support thi s ei ther with direct funding or subsidies or, o r you r ta x incentives and things like that that are goi ng to ma k e in dustry competitive globally.

Françoise von Trapp:

Yeah, it would be somewhat ironic if all of the major semiconductor manufacturers get the big pieces of the pie and, and start manufacturing all these chips on shore, and then we have to send them off shore for the packaging part. Um, that doesn't really make sense. Does it?

Alan Hoffman:

The reality is, a lot of those people interested in getting those chips don't have that choice. If you're going to build your chip in a three na nometer l eading edge fab, and you're going to continue to try to package those things in an infrastructure in a, US wide capability that isn't commensurate with that you're, you're spending a lot of money on the front end without thinking about how you're going to do the integration back in the last de cades. An ything y ou can't take those two things apart.

Françoise von Trapp:

Well, TSMC is, has been talking about building also an advanced facility in the United States, right. And, um, Intel, I think also, and...

Alan Hoffman:

in Japan.

Bobb Patti:

Actually, they mentioned doing an Arizona as part of that fab complex. Yeah.

Françoise von Trapp:

Well, they kind of would have to wouldn't they? I don't know if you read yesterday, I saw a sponsored piece by Pat Gilsinger from Intel talking about, using those funds for US companies, not other nations who were g oing t o invest in the United States because they feel like they're going to take that technology back to their country. Yet Intel is also investing$20 billion in Europe, ri ght? If politicizing chip manufacturing is, are th ey're d angers to that?

Bobb Patti:

Well, I, I think it's fundamentally politicized because his has been built on the foundation of we're afraid of China's rise. And at some level, the, the nation has realized how the leading edge technology has moved off shore. We all believe that Intel was at the forefront and for years, decades, they were a couple generations ahead of every everyone else in the race. And all of a sudden we woke up one day and Intel, isn't a couple generations ahead. As matter of fact, Intel is now a couple generations behind. Um, so I think there's this very cold shower that has happened in regards to where our position is in the world. And the fact that other nations have been paying attention to this and putting money to it. It is fundamentally a political concern because for the same reason that China is pushing so hard to onshore its own semiconductor or in country it's semiconductor, we feel like we need to do the same thing. And the recent shortages, which I believe have their foundation in many different areas. COVID exposed i t. But it's t he result of many years of shortening the, well lengthening the supply chain and reducing inventories, making everything just in time. There's no rubber band and the pandemic caused a reallocation of resources. The semiconductor business didn't shrink last year, it moved. So what happened when the economy started to come back is all of a sudden there was this need for this massive growth very quickly because the move in the industry i n what production was all of a sudden came back, and that was an enormous gap in capacity. So the shortage we're seeing today is the fact that we didn't build capacity last year. And most people didn't plan for capacity builds this year. I think all of these things, they do have a political basis to answer your question. U m, and I think that it would be simplistic to assume that that politics is not going to play a significant role in how, microelectronics policy a nd h igh-tech policy is set moving forward.

Dick Otte:

I would add to that, that the next level, once you deal with the political issue is what I'll call the financial orientation of American culture and business these days. I think one of the reasons the supply chain got tightened so short is, is to enhance return to investors. And I think one of the things that we're going to have to do if we really want to do this effectively, is we're going to have to loosen up on the financial side. And that's what the government funds allow us to do is to spend money on things other than maximizing return for investment, and to become more sustainable, to invest in bringing things back on shore, putting more rubber band into this is Bob suggested earlier. So there's more flexibility in it. We're gonna have to train more people. We've got to bring back on shore the material sources. One of the issues that we in the packaging industry face is, that no longer are there any U.S. Sources for molding compounds, lead frames, bonding wire, all of those technologies have gone off shore here, and we're going to have to bring all of that back to, and, and boy, there's a lot of stuff that has to come back here and we sort of gave away for a 5% or 10% reduction in costs.

Bobb Patti:

And w I'll just, uh, put out there that I, one of my big concerns about the, the investment that is being made and its political focus is we're fighting the last war also. It is a big concern of what I call about Foundry 1.0, the traditional model and pushing More Moore, technology going down the curve, which is very expensive. All of us on this chat really are in the, what I call Foundry 2.0 Business. It is the Chiplet. I I'm a strong believer that we're going to see a bifurcation in the industry to Foundry 1.0, which is entirely driven by cap ex and Foundry 2.0, which is driven by advanced packaging chiplets- by fundamentally innovation. And this shift in the semiconductor business is going to be revolutionary. I have some concern that putting all the money in that they are may stifle some of that because the focus will be on more and more technology, not more than more technology. I'm hoping that isn't the case, but it's important that the, the politics of this understand what the new frontier is, where, where does the road go? We're busy right now, concentrating on the problem we have today and the problems we've generated over the last decade. We can't lose sight of where the industry is going to be. And what do es t he industry need to be in five years, 10 years. And as we go through this process of onshoring, it's very important that we recognize what do we need to onshore to make sure that Foundry 2.0, which I think is where the growth is going to be. It's where the intellectual property value is. It's what creates those slots that you were talking about Di ck f or, for talent, you know, into the marketplace, tr aining a bunch of people to be semiconductor technicians today I'm not sure is th e best investment because the fabs become more and more automated and larger and larger. It's going to be, in this Foundry 2.0, which is going to be a step back to almost the way the world was when we were doing printed circuit boards with 7,400 series logic. I look at the next generation of being, you have catalogs of chiplets. You have some specialized IP that may be we do in a g o back to a Sea gates kind of approach. And instead of building these SOCs that require a thousand engineers for three years and a quarter of a billion dollars in development, it is now more like the small semiconductor company of 20, 30 years ago that really could innovate on a 10 or$20 million budget and have a new product and drive the next PC, the next AI, what that next frontier is. The politics around all of this, we have to make sure that it doesn't lose sight of winning the war just so that they could win the last battle.

Françoise von Trapp:

Yeah, you bring up some really good points. And one of the things that I'm thinking about right now is that, I've noticed a lot of the news out there, that all of these fab expansions are going to solve... They're not going to solve the current chip shortage. These foundries aren't going to come online t ill 20, 23, but using that argument, A lan, you made me think of that when you were talking about how the media, how they're setting themselves up in the media, if the$52 billion goes for that, that's not helping anything. That's a problem right now.

Alan Huffman:

You're, you're not solving that problem today, tomorrow or in a year. I mean, again, it's pretty clear that the major global players and Foundry technology are positioning themselves to when this money becomes available. And I think it will to say, well, we were already investing in the solution for this. So in order for us to carve a piece of that out for you guys to have as well, here's what we're going to ask for. And that's fine. I think we all understand t that that's going to be a piece of this. And I completely agree with Bob, you know, there, there, there will be some continued investment and it's going to be a large amount of money that gets put into More of Moore, but the More than Moore approaches that companies like ours have been working on now for the last 15 years, those are going to be the technologies that are, that are going to push forward fundamentally the capabilities of the microelectronics industry, you are just not going to make things work faster by spending more money on smaller gates. You're going to need to start integrating things together in different ways. You've talked about it and you're on your side. We've all talked about it in our own respective publications. My hope is that we don't lose sight of that because, you know, certainly I would imagine that all of our companies are somewhat active in having our eyes on, on this and possibilities, but we really don't have places that you can go buy interposers as a commodity today. Right. And that's a fundamental question. Um, and in the United States, if you want to make that a ground base piece of technology for that More than Moore generation, you have to address that question. So it's going to have to be part of the solution as much as the, Foundry technology and the partitioning of devices into chiplets. I think that's going to be something that this sort of program can help jumpstart because those are all economic decisions, right? If there's not a profit in it for Intel to start making chiplets for everybody to buy off the shelf, then they won't do it. But if you can somehow jumpstart this with, with some seeding technology money and make that available for people to start utilizing, kind of take a step ahead. And that's what I'm hoping will happen, but it's, it's very geopolitical, it's very national security centric and not just from a defense perspective, but for individual countries kind of protecting their own industries and trying to make sure that they are showing that they're supporting.

Françoise von Trapp:

I don't know if you saw the E E Times article labs over fabs, but in the House of Representatives, they're pushing more for a more holistic approach in funding- the national science foundation, I think- as part of that chip development versus handing over a lot of the funds to establish companies to do manufacturing. So do you feel like you're more aligned with that approach?

Bobb Patti:

I think I'm more aligned with that approach. And I think fundamentally it's the better return on investment. AT some level, having a lot of economic resource, having a lot of money, like Dick said earlier, the U S has lots of money. In some ways it's almost a curse. Um, if you look at how China has spent some of their money, as well as other countries, as they've come up, the technology curve, a lot of the time, they have a much smaller amount of resource. So therefore they're much more strategic about what kind of ROI can they get. And they do look forward to what is the next generation, how can I get in front of the problem? Can I, can I be the guy who invests, you know, a little bit of money, but I'm there first and therefore I have more leverage and I become the nexus for that next technology. Uh, the, the approach by the national science foundation, or what's been mentioned in that regard really, I think, is in that direction, let's, let's focus on generating the next industry. It's more about winning the war, not, you know, refining the last war or trying to win the last battle. I'm not suggesting that we just simply see what I was calling Foundry 1.0 earlier to, foreign competitors. I think that there is a need for a national effort to onshore those kinds of things. It's just, we can't focus on only that as the answer, because in five years we're going to have the same conversation it's going to be, well, we build all the chips and then we send them to China for packaging and for chiplet assembly. It will be the same thing. We're going to miss the opportunity to nurture that industry domestically and have it be here from the get go. Instead, we're going to have a conversation about trying to reassure that because the expertise has moved.

Dick Otte:

So, I mean, I can see two ways to go forward to address that problem, which in my mind is really critical. First, we can go back to the historic model. The electronic industry had of vertical integration, so we could set up a big corporate, a new corporation to do what I'll call advanced packaging that makes its own ceramic. Starting with sand makes its own polymers, starting with oil, builds its own machinery like was done in the early days and build a vertically integrated company that it addresses advanced packaging. The alternative is to stay with the model we have today, whereby some magic that's a little unclear, a bunch of small companies pop up that provide all of these lower levels of the, of the enhanced packaging component supply chain. And to me that that is very hard to do today because, what's the approach that we use as what I'll call an industry structure here and how do you direct this money to cause the, the capabilities to really emerge here in the U S.

Bobb Patti:

I think the ship proposal that the Navy had put out there when, when it came out in my mind, it afforded an opportunity to kind of create this pseudo public, private partnership corporation to do exactly what you were saying, Dick. And I know, I was one of I'm sure dozens that submitted proposals on that. In my view of it, which was shared by 40 companies was that you create a consortium effort almost along the lines of what AIM was or the power electronics Institute was, and use the government money to fill the gaps. Basically, we have lots of bricks that we could put together, but we needed mortar. And the using the government to help fund the mortar, I think is something that the government can do. It can create, uh, in infrastructure. It can, it can do what the big corporation did. Okay. But I'm not in favor of trying to replace all the small businesses because it's where the innovation occurs. It's, uh, the nimbleness, the ability to adapt to new technology. You really need the small companies, but you need, you need somebody to fill the gaps like you're saying, and there's the potential to put together a advanced packaging Institute, for instance, that would concentrate on how we do this. How can you nurture, fill the gaps, and until the industry can develop half the government subsidize the gaps.

Françoise von Trapp:

Other than the material suppliers and the equipment, Isn't this what Intel TSMC, Samsung have all been doing by developing advanced packaging technologies. And, um, the, the more advanced the chip, I mean, a lot of the chiplet stuff is coming out of these companies, right?

Alan Huffman:

Those guys capitalized on what I considered to be an advantageous business position from their Foundry, right? They said, we can give you seven nanometer stuff. U m, and if you want us to package it and guarantee that its all going to work i n an area that a lot of the OSATs aren't really capable of working on now to the level that we've developed it, then you can keep it here and we'll have an integrated solution that we can give to you. Right? So, you know, T SMC, in my opinion was the first one that realized that thihs i s the way they need to go and Samsung followed them. And now Intel is doing the same thing as well. So do I think that we need a vertically integrated single kind of OSAT here in the United States that iterated out of this sort of large scale investment by the government? I would argue no, but I would put my hat i n, in Bob's corner, I think, and say, you have a situation where there, I mean, there are a lot of technology companies who have been working on things like this, including all of ours for the last 20 years. And the reason that you don't have one of us in an Intel size sort of facility these days is because there's not the demand for that technology yet. And so y ou h ave t his fundamental problem, right? You have the technology that people want utilize. You haven't realized a way to scale that up from a financial perspective, because your investors are looking at that situation and going that many millions of dollars is a bit speculative. So what's a better way to do this, right? Obviously, if you can create, and this is the b uzzword today, the public private partnership, where you can help support yourself and subsidize over that period of time that these companies are going to start utilizing this more and more it's, it's, it's sort of a chicken and the egg problem, right? And until there's so meplace, that can support these types of advanced integration technologies on a bigger scale than they are being done today, there's always going to be this viewpoint that that's a risky proposition for companies to begin to start adapting and embracing that, interconnect paradigm. So on our side of this, the more than more side of this that we exist in, I like the idea that, the public private partnership is a way to help expand the capabilities and fundamental technology capabilities that a lot of companies have developed. And then if you continue to nurture those over the years, while the economy for those technologies continues to grow, and it will definitely move in that direction. I don't think there's any question. I think that's a good investment when you look at this as a lo nger-term i nvestment and as this initial investment.

Françoise von Trapp:

And short on time, there's one question I want to make sure we get to critics of the chip subsidies question, the need to throw billions of taxpayer funds at a profitable industry and warn that the incentives arms race could create a glut of production. In other words, getting back into being at over capacity, what, what do you think about this?

Bobb Patti:

I think it certainly is a risk, but it's foolish to ignore the way the world's been since probably the early eighties, when Japan was kind of the first to incentivize industry to go there. So shaking your fist at the tidal wave and saying, you know, I don't want to participate in what everyone else is doing, does you know good. The industry has created incentives. I think it's up to us to make sure we spend the money wisely and that we get a good return on that investment. I think what people don't realize is the semiconductor industry is extremely conservative. Investment in semiconductors is a very conservative business. Building in the US is a big risk because of the economic imbalance that exists in the marketplace. Uh, many years ago, I talked to the head of a large semiconductor company, and he talked about the risks that they have. They build billion dollar factories. Every day that a fab line is down costs them millions of dollars. They are risk adverse. What we're describing the business that Alan and Dick and I are in we're risk takers, where the innovation is really semiconductor risk. And even though we're small dollars in our markets compared to building new fabs, a large semiconductor company looks at what we do as a very high risk proposition, even though it's small dollars. Large semiconductor companies spend billions of dollars on fabs, but for them, it's like putting money in the bank. They know what's going to happen. They projected their... They have 50, 60 years of history of how this works with more and more technology to back them up the need for the government here, I think, and this is my concern is I don't want to put all the money there because it is risky to bring it on shore. And we want on shore I'm with you, but I don't want to spend all the money there because really the risk adverseness of this industry is preventing the expansion of the More than Moore technologies, which is the future. And if we don't invest in that today, I fully expect China. I know China is investing in this kind of work. Um, they have gone to it because it was a way to get around a lot of the rules, the regulations, their lack of onshore capability, they're going to promote it. And if it, if it isn't China pick another small country, um, or another country not necessarily small, um, yeah, it is important that the government incentivize industry spending the dollars to bring the industry, create the momentum on shore is important, but it will also require a consistent technology or a consistent approach to high-tech and incentivizing high-tech. And, and that extends to whether it's taxes or subsidies or incentives and the education it's the workforce development. We have to be able to staff the high tech when we bring it here. And so this touches on a number of policy areas that we really have been delinquent in managing as a nation for decades at this point.

Dick Otte:

You know, a comment I would make on this whole issue of how do you justify this to take an altruistic view of it? I think the US building these capabilities onshore, so we are less dependent on Asia gives us what I will call a greater independence and less incentive to go to war if China invades Taiwan, to put it simply in the pressures that it seemed to me inevitable that there's going to be a conflict over Taiwan. And that's the day that we lose access to semiconductors out of Asia. And that's the day that the current semiconductor shortage looks like a picnic compared to the mess we're going to have. Yeah. So, as somebody said at the very beginning of this, it's a, it's a national security issue because now we are extremely vulnerable. Our whole us economy is extremely vulnerable here. And the result of that is it's generating a politically militarily unstable environment here. And so I, I see that as one of the major reasons for trying to develop independence at the very high level of how do you justify this?

Alan Huffman:

I completely agree with that, Dick. I think the question of, of, you know, if a critic is saying, well, aren't you worried about creating excess capacity and in saturating the industry, that's a really good reason to say that that that's, that this is a worst worthwhile investment. If you have, you know, again, advanced Samos capabilities within the United States, does that add additional capacity globally? Yes, it does. Is that going to happen in a short amount of time? No, it isn't. And there are, you know, the geopolitical and national security and supply chain and industry questions that, that come attendant with with making that decision.

Alan Hoffman:

Um, but you guys know as well as I do that, there aren't a whole lot of foundries that sit empty these days and micro electronics consumption and integration into all kinds of things and move forward. So, um, you know, I completely agree with Bob, it's a risk averse industry and, and at the upper end of this, people spend a lot of money, but these facilities in place, and they don't do it without a very high, uh, without a very high expectation of return on their investment. The, the history of the semiconductor industry has been that they can generally do that and, and, and do it pretty well, but it's an extended period of time that you do it over. So, you know, I don't think there's a big risk of, of over-saturating the industry with supply and, and having a problem there. But it goes back to the basic tenant that we were talking about this entire discussion is it's not just Foundry technology. It's got to be what happens when you get wafers out of Foundry and the integration side of this thing, at least as far as I'm concerned, and I've made my career out of this, the integration is where things are going to get driven, going forward. And it is absolutely the area that has the least amount of investment in all of the areas of microelectronics from, from the U S perspective, I think.

Dick Otte:

Well, yeah, and, and compare compared to what it's going to take to put in place significant a state-of-the-art semiconductor fab. The ammount of the money's involved in the enhancing, the semi enhanced packaging technology industries. It's probably an order of magnitude less, and it's almost as equally important. So in some sense, it's, it's really critical. And in some ways it probably has the best return on the best benefit here. In terms of the degree of.. We break our dependence upon Asia for molding compound, lead frames, wire bond wire, all of those things suddenly a re back. And there's a whole bunch of weak links that are eliminated if we do this, whereas building the semiconductor fabs just eliminates that one big, really big key, but man, this whole industry, the supply chain is called a chain for a good reason. I t's everything is in series and every link in the chain has to work and be available. And w e've got to make sure that we fill in all these other little links calls, ceramic, et cetera.

Bobb Patti:

Yeah. In building on what you're saying it exactly. If we spend that 52 billion to onshore wire bonding, great, except it has no usefulness where semiconductors are going to be. We have to use common sense in it. And, and I'll also put out there, I think I have a fair amount of skepticism about how many of these foundries ultimately get built. I think that right now it's in vogue to talk about onshoring and announcing, but you know, there's, uh, a lot of time between now and when that happens, I think a lot of these companies are queuing up to be in line for some of the money that's going to be available. And it, isn't only in the United States, it's in China, it's in Europe. Um, and what's ultimately going to happen is when it kind of shakes out and they see where the money is and how much is available, We probably are going to see far fewer fabs ultimately built. And some of them that actually physically get built, maybe empty shells for quite a while. It, uh, yeah. You know, semiconductor, they're conservative. They're not going to go out on the limb, Intel TSMC, Samsung, GlobalFoundries. None of these guys are going to go out and build a whole bunch of capacity. If they see everyone else building lots of capacity, unless they have it tied up. And I'll point out that a lot of the semiconductor guys have changed the paradigm with their customers and they have long-term contracts. I have heard about six year contracts now with large customers for capacity. So they're going to build fabs based on a low risk. I have a captive customer, and I know what my risk is. If I build the fab, how I'm going to fill it. Um, I don't expect any of these guys to go spend billions of dollars in hopes that some customers are going to show up.

Françoise von Trapp:

Okay. So I think this is a good place to stop, but I also want to say that this should be a, to be continued because we've made a lot of predictions. There's a lot going on right now. And I would love to come back in a year and see where we are with this. And, and what, you know, how many of those projects have moved forward. How many are stalled, um, where the money went, um, or where it's going. So I really appreciate your time today. This was, um, a really wonderful conversation. Thank you. Thank you. Good to talk to you guys. There's lots more to come. So tune in next time on the 3D InCites podcast.