ads mdx semiwiki building trust gen 800x100ai

How France’s Largest Semiconductor Company Got Stolen in Plain Sight

How France’s Largest Semiconductor Company Got Stolen in Plain Sight
by Doug O'Laughlin on 01-23-2022 at 10:00 am

SOITEC Stock Price

Originally published on Fabricated Knowledge

Soitec is a semiconductor materials company known for its smart cut and Silicon on Insulator (SOI) technologies, which are critical in 5G, Silicon Photonics, and Silicon Carbide (EV) end-markets.

Yesterday, they announced that current CEO Paul Boudre will retire and be replaced by Pierre Barnabé in July of 2022. Barnabé is currently an SVP at Atos, a French IT consulting company. This seemingly routine CEO transition sank company shares by over 15%. It’s not what it seems.

First, the entire existing management team is opposedThe management team sent a letter to the board that does not mince words. Translated to English, it reads:

Soitec’s Management Committee deplores the takeover of Soitec by the Chairman of the Board of Directors for 3 years, which culminates today with the incomprehensible appointment of a new CEO.

What exactly is happening at Soitec? Well first let’s just start with the current story of the now outgoing CEO, Paul Boudre.

The Story of Soitec (and Paul Boudre)

Soitec, like most firms, was no overnight success. This timeline of the companies history is the best place to start. Soitec originally pursued other markets but failed to gain traction. In 2015, the company Soitec shuttered its solar business as part of a hard pivot away from its existing failures.

The situation was dire. The company had more debt than the value of its market capitalization.

For context Soitec today is a ~5.8 billion euro company now

In January of 2015, with shares trading at measly $200 million market capitalization, Paul Boudre was appointed CEO. Paul initially joined the company from KLA as an Executive VP in Sales before transitioning to COO in 2008. Drastic changes were needed.

Soitec’s rather bleak 2015

In the midst of significant layoffs in the solar division, a hefty financing package company to keep the company afloat, Paul focused the company on SOI and made it what it is today.

The details of the turnaround are unimportant, but the results today speak for themselves: Soitec now boasts a ~$5.8B market capitalization (~29x higher than the start of his tenure) and has 4x+ annual revenue. This is a legendary turnaround that will surely be sung of in Semiconductor Valhalla.

At the age of 63, Paul is approaching retirement. Given such a track record, one imagines a thoughtful process involving existing management, not the overnight skullduggery that actually occurred. Further, the new CEO candidate has no experience in the semiconductor field. Why did the board choose newcomer Pierre Barnabé over other qualified internal candidates?

To answer this, we need to learn about the Chairman of the Board.

Enter Eric Meurice (and Pierre)

Eric Meurice is best known as the former CEO of ASML. In 2013 Eric was appointed Chairman of the board of ASML after 8 years of being CEO. Note the exact lanaguage of the Chairman announcement.

Eric Meurice will be Chairman of ASML Holding and act as adviser to the new leadership and the Supervisory Board until the end of his contract on 31 March 2014, ensuring a smooth and comprehensive transition of critical tasks and processes, customer contacts and relations with strategic suppliers.

The current CEO gets the job during Eric Meurice’s contract and Mr. Meurice gets bumped to Chairman. It’s pretty customary for the outgoing CEO to spend time as the current Chairman. It’s less customary to not have the contract renewed. Despite European CEOs having shorter tenures, this is atypical.

As the former CEO of ASML, Eric Meurice is now an ideal candidate for board memberships. Here’s a list of the few boards he’s been a part of.

Let’s hone in on Eric’s stint at Soitec. Eric Meurice joined the board of Soitec in 2018 as the chair of the Nomination Committee. The committee was tasked with nominating a new Chairman, so else does Eric Meurice nominate but himself?

Eric, overqualified as he is, gets the job. But that’s not enough. In 2019, he picks up two more important roles as the Chair of the Strategic Committee and Chair of the Compensation Committee. He now holds both the keys to the kingdom and to its treasury.

Joining a company, becoming the Chairman of the Board, and then taking on additional roles as the head of the Compensation and Strategic Committees fit into the typical model of a high-powered executive. That’s standard.

How does the French Government fit into Eric’s rise to power?

Eric Meurice’s Extraordinary Actions

Let’s look at the specific actions Eric Meurice took that prompted significant backlash from the executive committee. In the executive committee letter, they listed out a few specific complaints that were (importantly) falsifiable. Their (translated) list of grievances is as follows:

  • Takeover of the interim compensation committee that has become definitive, creating an omnipresence in all committees and at the head of several committees
  • Interference in social dialogue without consultation with management.
  • Double language regarding the opposition of management on the implementation of PAT (Action Plan for All) in 2021.
  • Establishment of internal regulations granting exceptionally extensive powers to the Chairman of the Board of Directors and establishing the keys to his takeover.
  • Alteration of evidence in the context of the investigation of a governance drift.
  • Intimidation, vexatious practices towards members of the Executive Committee.

The board granted itself extensive power through a list of resolutions added to the typical bylaws of the company at the so-called “extraordinary shareholders general meeting”. It’s rare to see an extraordinary resolution, so it’s outright mind-boggling to see 35 resolutions. This is one of the broadest power grabs I’ve ever seen. Let’s consider a few of the resolutions.

There are a total of 35 resolutions, each giving the board more power than a typical board would have.

The resolutions are technical, but the gist of them is that the board now has a whole set of new powers that are usually reserved for the CFO. They can issue shares, buy back shares, decide who gets shares, and all of these powers are directly granted to the board. This creates extraordinary power to the board and the people on it.

This explains why they had 3 CFOs since the extraordinary resolutions began. Remy Pierre was replaced in September 2019 by Sébastien Rouge who was replaced one year later by Lea Alzingre. That’s high turnover for the job.

Maybe these resolutions could be kosher, but the reason they are not is all the extraordinary resolutions are new for Soitec. Take 2017 for example, when there were only 6 plain resolutions. Something has changed.

6 routine resolutions in 2017.

But besides the mundane extraordinary powers of resolutions given to the board, there’s more at play. I now want to focus on the executive committee complaint around the compensation board because this is where the other players (France) start to enter the scene.

Board Power Politics

Let’s talk about the composition of the board. The board power politics make sense when you can see who is sits on what board committees. There are 5 board committees in Soitec, and the restricted strategic meeting is an ad-hoc group for acquisitions or other events. So that means there are really 4 standing committees and 1 ad-hoc committee. They are the Strategic Committee, the Audit Committee, the Nomination Committee, and the Compensation Committee.

These committees are filled with 14 members, many of them are supposed to be independent. This breaks down with further examination as many of the “independents” are clearly affiliated with the French Government. Now let’s start with chairs of the five committees.

Eric Meurice is the Chairman of the BoardChair of the Compensation Committee (most powerful committee), and the Chair of the Strategic Committee.

Laurence Delpy is Chair of the Nomination Committee, the chair that Eric Meurice previously held before he became Chairman of the Board. She’s independent, yet she sits on all 5 committees.

Lastly, Christophe Gegout is chair of the Audit and Risk Committees. He’s supposed to be an independent member, but he used to work for CEA aka the large French consortium with a meaningful stake in Soitec. He doesn’t work there anymore, but they’re clearly playing fast and loose with the definition of the word “independent”.

Let’s look at the actual composition of the committees and identify 1) who is in what committee and 2) which committees matter. I made a simple graphic based on the filing with a legend that explains where everyone’s allegiances lie. In order of importance, it’s the Compensation, the Nomination, the Strategic & Restricted Strategic, and lastly the Audit committee. I broadly categorized the board members into the 4 “teams”, aka Team France, Team China, Independent, and Employee Directors. Note the legend in the picture below.

Notice the legend in the photo above. Team Blue / France is the one to watch

Let’s start backward. The green team is employee directors and is part of a movement to get labor union leaders on the board so employees have more say in the company writ large. For the sake of this analysis, I consider them non-players, as this is their first year on the board, and they don’t sit on committees.

Next is the team “actually independent”. Note that Satoshi Onishi works for Shin-Etsu, so he’s independent as in he’s representing his company’s JV with Soitec. He is not a big player. In Shuo Zhang’s case, I cannot make any meaningful connections to anyone else. Paul Boudre of course is the outgoing CEO. Notice that he does not sit on any important committees.

That brings me to Team China. Team China is the NSIG (National Silicon Industry Group) block, which invested 14.5% at the time which now is diluted to 10.34% stake (this is lower now) into Soitec in May 2016. NSIG, like many large Chinese firms, is an extension of the CCP. They hold two seats that I represent in Red. Kai Seikku is actually sitting on the powerful committees, aka the nomination and compensation committees. But importantly Jeffrey Wang has been pushed out to the not-important audit committee.

Last is Team France. With the exception of Francoise Chombar, they are all French Nationals. I put Francoise Chombar on Team France because she shares a board with Eric Meurice at Umicore, so I assume she’s on his team.

Everyone else either works for CEA (French Alternative Energies and Atomic Energy Commission) or Bpifrance (French Public Investment Bank), formerly worked there, or suspiciously looks connected (Laurence Delpy is clearly important but I have no links other than she worked at Alcatel Lucent where Pierre is from). Thierry Sommelet works at Bpifrance for example.

Importantly Team France consists of 6 out of 8 members of both of the most powerful Nomination and Compensation committees. And everyone else who is not affiliated with Team France conveniently sits outside of these boards with the exception of Kai Seikku, who represents the powerful ~10.34% share block from NSIG. Team France is clearly in control here, and the BPI and CEA seats are permanent, with rotating members but consistent committees.

What I’m trying to say is that Soitec’s board is controlled board by a very small number of players, all of which can be linked to France. Obviously, these members want to protect the interests of France, and what’s more most of the moves taken by the board pre-date Eric’s arrival. So it’s clearly not Eric in charge, but rather the representatives of France that are driving this bus!

What’s France Got to Do With It?

When I first started down the rabbit hole of the Eric Meurice takeover I thought the motivation was pretty simple. Ousted CEO Eric Meurice was looking for another kingdom to rule and found it in the form of Soitec. A clear power play, as highlighted by the management letter. After all, we knew he already had ambitions to sit in the CEO seat again, per this ST Micro rumor. In a series of successive moves, he rose from Director to Chairman, and he pushed his way to the top.

There are a few problems with that theory and it comes in two bold flavors. First, the year (July 2018) that Eric Meurice got appointed to the board as a director was the first year of expanded extraordinary resolutions. The previous year it went from 8 total resolutions to whopping 23 new resolutions. So the expanded board powers actually pre-date Eric’s arrival on the board. Eric Meurice was just the conduit for the control of the board.

The second thing was this extremely crucial disclosure about a standstill agreement with NSIG that made clued me in that something else was happening. When NSIG (National Silicon Industry Group aka China) bought the 14.5% stake in Soitec, they agreed to a standstill agreement on the shares.

The standstill agreement is an agreement at the time so that NSIG (China) didn’t continue to raise their stake in Soitec and effectively take over the company. It’s a takeover provision that stopped their large influence from increasing. The French board members were clearly aware of this.

But that standstill agreement ended on June 7, 2019, which was 1 year after Eric’s rise to power. This explains why he entered when he did. However, this tidbit made everything become clear(er) regarding the resolutions.

Should NSIG Sunrise S.à.r.l acquire shares in the Company before the expiration of the Shareholders’ Agreement at the close of the Shareholders’ General Meeting called to approve the financial statements for the fiscal year ended March 31, 2021, it would lose its rights relating to the Company’s governance

Team France knew they needed to lock down the company from a governance standpoint before March 31, 2021, or risk further influence from NSIG aka China. And Eric Meurice was the perfect man for the job. Win-win.

Besides Paul Boudre was ready to go. He terminated a paused employment contract so that the company wouldn’t have to pay him a termination fee, and they rewarded him with an incentive (I find this weird). He clearly signaled he was on the way out. Readers of the filings could have spotted his retirement as early as 2020, it was just the executive team that was blindsided.

But that brings us to what really happened. France Nationalized Soitec through a series of board moves, just in time before the Chinese government could push for more control. The CEO put in place is just a placeholder for the board.

National Champions Need Nationalization

How are we surprised that France wanted to nationalize Soitec?! This is France we are talking about here! It’s clear that there’s a vested interest in keeping Soitec French-controlled. The expanded board powers also expanded government influence over Soitec. Soitec has effectively become a State Owned Controlled Enterprise.

Look at the shareholder base. The ~17.67% French-controlled block was large, but not completely dominant given the large NSIG (aka China) block. A Chinese takeover of France’s semiconductor star would be devastating (not to mention embarrassing). Team France could not let this happen.

All of the board actions in conjunction means that this was likely premeditated by the controlling shareholders – the state of France – to further control Soitec. Paul’s retirement was just the catalyst to push the changes that have already been made ahead. So what about our underqualified CEO candidate, Pierre Barnabe?

The thing that I found really curious was that Pierre is on the board of INRIA, or the Institute for Research in Computer Science and Automation. It’s a government lead entity that’s core purpose is to further France’s technology interests. From the French National control lens, Pierre is a perfect CEO.

What Now?

So now France effectively owns Soitec. What are they going to do with it? Paul left the company on a strong financial note and Soitec is a springboard for other national pursuits, like manufacturing Silicon Carbide. Manufacturing Silicon Carbide would be terrible for Soitec the business, but great for France’s national ambitions.

Another likely option is that the board can use its expanded powers to purchase a fab in Europe, which would be perfect given Soitec produces wafers. Now France produces wafers and chips! There is a small design team within Soitec, and expanding that could also further French national interests. That’s a full stack semiconductor company with just 2 additional acquisitions.

All of this makes sense in the now geopolitically driven world of semiconductors. As of late, there have been multiple announcements for new mega-fabs, like the Intel Ohio fab or the new TSMC Japan fab. Every country is doing what it can to shore up its semiconductor businesses, and France couldn’t let China steal their national champion.

Conclusions and Questions

Soitec is effectively nationalized through board control. It makes sense given that China had a window to push for control, so instead, France just took the whole thing. France, Soitec’s largest shareholder, put everyone in a place of power in order to achieve this. What looked like a power-hungry move by a single actor, Eric Meurice, really was a coordinated win-win to control the company for France.

While I’m sure Soitec’s nationalization will be disappointing for free-market capitalists (where they at?), it’s not surprising at all given the current semiconductor climate. It’s a de facto-controlled company now. The takeaway: national-level politics continue to matter in the semiconductor industry. This theme is not going to go away any time soon. Soitec is just the latest and greatest in the series. Goodbye Soitec, Hello French National Semiconductor company.

There are some loose ends. How does NSIG feel about this? Did Paul Boudre know about any of this? There are a lot of other interesting threads in this entire story, but it’s clear what happened. France nationalized their largest semiconductor company!


If you enjoyed this piece, please consider subscribing. Even the free tier gets occasional posts. I try to write about semiconductor companies broadly from an investment perspective, so this investigative journalism is a bit different.

Oh by the way if you are either a past ASML employee during Eric’s time at ASML or a current Soitec employee who would love to talk, you can reach me at Doug@fabricatedknowledge.com.


Some Unfinished Business

I wanted to discuss parts of the story that didn’t flow well with the rest of the power politics at Soitec. The question I presume most people would be wondering is about Pierre Barnabe. Why is he not qualified again?

I really had a problem with this statement from Soitec management.

He draws upon a remarkable track record that includes a threefold revenue increase at the Atos Big Data and Cybersecurity division in the space of a few years, in a highly competitive market requiring deep cooperation with the ecosystem.

Part of that three-fold growth was 38 acquisitions along the way. Is that execution or is that just buying three times more revenue? And also the stock and business are horrid. It’s down 63% over the last 5 years and revenue growth is tepid at best.

Oh, and STMicroelectronics has a deep bench of French National semiconductor executives that would have made a perfect fit. Given Pierres in with the government, Pierre definitely did not get the job on the basis of merit.

What about Paul Boudre?

One of the weirdest side exits is Paul Boudre. By voluntarily abdicating his contract agreement and then getting paid to do so, I think Paul had wind of changes but didn’t care as he knew he was on the way out. And even if Paul did want to change things, he was locked out of the committees that matter (compensation and nomination). He wouldn’t be privy to what happened anyways.

I think the biggest blindside is the executive committee. I get why their reaction is so strong, but I think their outrage of Paul’s replacement and the lack of internal promotion is missing the bigger national interest story in what they thought would be a routine succession. The COO likely expected promotion, and other executives in turn would be promoted to COO. I feel bad for them, as it’s frustrating to not be a part of the company they clearly helped build.


Appendix: A bit more on SOI

Silicon on Insulator is a technology that embeds the insulator below the surface of silicon. Usually, this is done via Ion implantation, a “smart cut” and a flip of the substrate so that the buried insulator is now below a device layer. This improves overall performance meaningfully.

Soitec is the only volume manufacturer in the world, and their competitors license their technology. Soitec believes that their market share in SOI wafers is ~77% globally.

Subscribe to Fabricated Knowledge

Related Blog


Musk: Colossus of Roads, with Achilles’​ Heel

Musk: Colossus of Roads, with Achilles’​ Heel
by Roger C. Lanctot on 01-23-2022 at 8:00 am

Elon Musk Achilles Heel 1

For Tesla, 2021 was an amazing year. A blindspot looms in 2022.

Critics cheered the National Highway Traffic Safety Administration for opening multiple investigations into fatal and near fatal Tesla crashes.  Legislators decried the de facto beta testing of Tesla’s Full Self-Driving beta on public roads. And in December, click-bait headlines shined a spotlight on a distracting in-dash video game function – which Tesla had enabled and later disabled – ending the controversy.

While the critics howled, fans flocked. Tesla closed out the year by reporting 936,000 units sold globally and garnering a top safety pick from the Insurance Institute for Highway Safety for the Model Y, following up a similar assessment for the Model 3. For now, for Tesla, it looks like nothing but green lights – but there is trouble ahead.

Tesla’s CEO, Elon Musk, is famous for poo-pooing technologies that he holds in ill regard – among them: hydrogen fuel, lidar sensors, and wireless V2X communications. His dim assessment of lidar stands in stark contrast to an industry-wide embrace of the technology which might have served to help Tesla vehicles avoid crashing into police cars and emergency vehicles parked on the shoulders of highways.

Musk claims that lidar is unnecessary. What he is really saying is that lidar is expensive and he is trying to control the cost of his vehicles. For Musk, cameras and perhaps a little bit of radar are good enough.

In the same way, Musk has routinely dismissed vehicle-to-everything wireless communication technology enabled by cellular V2X (C-V2X) or dedicated short range communication (DSRC). Musk’s position is that autonomous or semi-autonomous vehicles must work without a network.

Musk’s opposition to wireless-enhanced autonomous operation puts him and his company outside of a vast industry collaboration working toward the leveraging of wireless technology to enhance vehicle safety. The onset of 5G, which brings with it V2X functionality, promises to enable a wide range of collision avoidance applications including the protection of vulnerable road users and the communication of the signal phase and timing of traffic lights.

These capabilities are arriving today in cars in China and, soon, in the U.S. as C-V2X technology sees swift adoption from car makers including Ford Motor Company and Audi, among others. These car companies understand that C-V2X will allow their cars to communicate their location and avoid collisions.

Meanwhile, manufacturers of infrastructure equipment are increasingly shifting to C-V2X tech to enable infrastructure-to-vehicle communications. Here, too, the objective is safety and collision avoidance.

Musk is resisting and Tesla is steadily diverging from the rest of the industry. While Tesla has led the way toward the widespread adoption of wireless-based over-the-air software updates, the company has neglected using wireless technology for safety purposes.

Tesla still lacks an automatic crash notification function (outside of Europe) – equivalent to General Motors’ OnStar. And Tesla’s traffic light recognition application that is part of the full self driving beta is entirely dependent upon vehicle mounted cameras and the vigilance of the driver for safe operation.

Musk’s camera-centric approach (enhanced with ultrasonic and radar sensors), which has helped propel the company to the forefront of semi-autonomous vehicle development has clearly reached the limits of its efficacy. We’ve all seen the videos of Teslas mistaking Burger King signs or even the moon for a stop sign. The multiple crashes involving fixed objects in the roadway speak volumes – to consumers and regulators.

The market leader in camera technology, Intel’s Mobileye, has dropped the camera-only pretense and is developing its own lidar while cooperating, in the short term, with lidar supplier Luminar.  That’s a multimillion dollar shift by Mobileye. Tesla has made amazing strides with camera technology, but has now clearly reached the end of the road.

To keep up with developments coming fast throughout the rest of the automotive industry, Tesla needs to embrace the integration of wireless technology into safety applications for collision avoidance and emergency response. What Tesla is missing by failing to leverage wireless is the ability to extend the safety sensing horizon of the car beyond the line of sight of its cameras – including over hills and around corners.

Without wireless, Tesla will also remain blinded to emerging wireless alerting solutions for everything from the movement of emergency vehicles (Haas Alert Safety Cloud), vulnerable road users, wrong way drivers, and, most notable of all, the presence and signal phase and timing of traffic lights. Tesla has proven that it cannot solve these challenges with cameras alone.

There isn’t much that Tesla is getting wrong – from fast charging to direct sales to software updates to battery gigafactories. Wireless connectivity, for Tesla, remains a weakness.

Also read:

RedCap Will Accelerate 5G for IoT

Traceability and ISO 26262

Siemens EDA Automotive Insights, for Analysts


ASML Too Much Demand Plus Intel and High NA

ASML Too Much Demand Plus Intel and High NA
by Robert Maire on 01-23-2022 at 6:00 am

ASML High NA Intel

Too much demand- A “good” problem-Managing supply & capacity-Intel & Hi NA

ASML great Q4 results-Demand off charts-Supply constrained
-Dealing with chain issues, putting out fires, expediting
-Looking forward to next gen High NA in 2024/2025
-Intel’s order doesn’t give advantage, just joining a long line

Great Q4 results

ASML announced Q4 results of Euro5B in revenues and Euro4.39 per share in earnings. This was $5.656B of revenue and EPS of $4.98 (in dollars) versus street estimates of $5.85B and $4.25 per share, so a strong beat on EPS and a slight miss on revenues.

Most importantly orders for systems came in at Euro7.1B for the quarter including one High NA tool. Demand and business was strong across both EUV as well as DUV tools. 73% was for logic and 27% for memory. Taiwan was far and away number one at 51% of ASML‘s business followed by South Korea at 27% and China at 22% which leaves the US at zero/doughnut hole/bupkis%.

For the year Taiwan was 44%, South Korea 35%, China 16% and USA at 5%. This is obviously very reflective of our most recent article about the dominance of Taiwan/TSMC and how far behind the US is. And 2021’s numbers are prior to the huge jump in TSMC spend recently announced.

This is obviously very reflective of our most recent article about the dominance of Taiwan/TSMC and how far behind the US is. And 2021’s numbers are prior to the huge jump in TSMC spend recently announced

Can’t keep up with demand

ASML is the one and only game in town in defining the leading edge of semiconductor technology. There simply is no alternative, they have a monopoly.

Demand is off the charts and every semiconductor company that cares is ordering as quickly as possible. The geographic mix of business is reflective of the leadership in the semiconductor industry.

Despite supply chain issues in general and a fire at their stage manufacturing in Berlin, they have still managed todo a great job on shipments. The fire in Berlin seems to have little impact as ASML was likely able to move inventory and spares around to make up for the loss.

Its also clear that demand remains strong for DUV systems for second tier applications and memory applications. While ASML has done a great job at maintaining the pace, it is much harder to increase the pace as many subsystems are just very constrained.

Perhaps the biggest constraint is in lens manufacture as Zeiss in Germany doesn’t want to go as fast as demand would otherwise take them, likely for fear of the cyclical nature of the industry.

All of this is quite good for ASML’s gross margins as price is a secondary concern as compared to just getting a tool.

In our view, this limitation of capacity is not at all bad and may help not only pricing over the longer run but also reduce cyclicality as production simply can’t be ramped up as fast as the industry demands thus limiting the inherent volatility.

Expediting may add to confusion

ASML is being asked by customers who are so desperate for tools to short circuit the normal final assembly and test in the Netherlands and instead ship the systems directly to customers for final assembly and test.

This is akin to “don’t bother test driving my new car just deliver the pieces to my driveway and we’ll take it from there”. This adds to confusion on the financials as revenues that are usually counted on shipment now have to wait until final test at the customer site.

Investors will have to learn to focus on shipments rather than revenue and we may see numbers swing back and forth between quarters. At least Euro2B is expected to be delayed in recognition initially.

High NA starting to come into focus

Next generation , High NA tools are coming into view. ASML said they have orders for four model 5000, High NA R&D tools and Intel just placed an order for a 5200 “production” tool. Hopefully we start to see the first of these in 2024 and 2025.

I think we can safely assume that of the four tools on order, TSMC gets one, Samsung gets one and Intel gets one. Maybe someone else gets the fourth tool or someone gets two. If the pattern follows history maybe TSMC ordered two.

Intel claiming High NA EUV “advantage” is simply nonsense

Right now its unclear who will get the first of the four High NA EUV R&D tools on order. It could be TSMC, Samsung or even Intel. The only thing we know is that Intel placed an order for the first “production” tool, the 5200. So who is really first to High NA is a question of semantics. More importantly the first “production” tool is not until at least 2025.

As we have seen with the original roll out of EUV it took more than two iterations to get to real production. More importantly, right now, TSMC has roughly 10 times the number of “real”, “productive” EUV tools. This also means that TSMC likely has 10 time the staff trained on complex EUV, ten times the experience (likely even more).

Most importantly, TSMC has likely more than 10X the capacity and experience in building EUV masks which are the “hyper critical” negatives from which the chips are printed by ASML tools.

Given the combination of the current EUV tool count competition, the recently announced hugely expanded TSMC capex budget and the constrained capacity of ASML it is physically/mathematically not possible for Intel to even come close to catching TSMC.

Even though Intel was the first to order a “production” version its totally unclear if they will get it first let alone get it any meaningful time ahead of TSMC. This isn’t even until 2025 at best! This means that Intel ordering a 5200 is more of a PR stunt which also helps ASML create competitive tension between already desperate customers.

The Stocks

ASML already trades at a very large premium as it is one of the very few successful European large cap tech stocks. It trades at a premium to US semiconductor equipment companies. Stocks have been faltering already and ASML’s announcement was into a soft trading day with the group down.

ASML is down less than the group so the results were taken positively as they should but the results were not overwhelmingly positive such that the stock could break through the general weakness.

Semiconductor stocks in general have been facing more resistance and good news is not driving them which is a reflection of the overall market sentiment. At this point we are less inclined to buy into weakness even though we still like the name. Momentum is a bit negative and we are in a critical earnings period. We think other semiconductor stocks will be focused on execution and supply chain issues with less focus on the positive aspect of huge unprecedented demand and record results.

Also Read:

Forty Four Billion Reasons Why TSMC Remains Dominant

“Too Big To Fail Two” – Could chip failure take down tech & entire economy?

Semicon West is Semicon Less


Podcast EP58: A look at 2021 and 2022 with Wally Rhines

Podcast EP58: A look at 2021 and 2022 with Wally Rhines
by Daniel Nenni on 01-21-2022 at 10:00 am

Daniel Nenni is joined by popular podcast guest Wally Rhines. Dan and Wally explore 2021, some of the expected results and some of the surprises. COVID, supply chains, chip shortages and international trade are just a few of the topics.

Wally and Dan then turn their attention to 2022. What kind of year will it be? What will be the drivers, the successes and the surprises? This is a far-reaching discussion covering many relevant topics. We exceed our 30-minute length on this one by a bit to cover it all.

The views, thoughts, and opinions expressed in these podcasts belong solely to the speaker, and not to the speaker’s employer, organization, committee or any other group or individual.


The 5G Rollout Safety Controversy

The 5G Rollout Safety Controversy
by Matt Commens on 01-21-2022 at 6:00 am

Plane Landing

There has been a lot of attention in the news recently about AT&T and Verizon rolling out their first implementations of sub-6GHz 5G radio access networks (RAN). Notably, the FAA and airline industries have voiced serious concerns about potential safety issues for aircraft autopilot and landing systems. As a result of these concerns, some international and domestic flights were  canceled, and the Customer Services Providers (CSP) did not turn on these new 5G systems near select airports around the U.S.

Here are the details: The wireless spectrum auctioned off by the FCC back in Dec. 2020 in the 3700-3980 MHz C-band frequencies is deemed to be too close to the 4000-4400 MHz bands reserved for aircraft radar altimeter systems. These critical, sensitive radio frequency (RF) systems bounce RF signals off the ground to determine the plane’s relative altitude for use by the aircraft’s autopilot and automated landing systems. Obviously, erroneous altitude information coming from this system could cause significant problems, especially during takeoffs and landings.

On the surface, the problem seems innocuous because the two systems are not operating in the same frequency bands. But, when discussing RF systems composed of active components, this is not so clear-cut. Active devices inside RF systems can generate spurious out-of-band signals. Although these signals may be of a much lower amplitude than the primary, intended, frequency, they can still be large enough to interfere with nearby systems operating in other bands. In addition, the radar altimeter systems are very sensitive by design,  since the reflected signal from the terrain can vary considerably in amplitude depending  on the altitude as well as features and shapes of the terrain.

A somewhat surprising aspect of this controversy is that it was not anticipated and coincides with the rollout of the 5G networks. Given the significant amount of time between the auction of the frequency band in Dec. 2020 and the rollout in Jan. 2022, it would seem that these concerns could have been raised earlier to avoid flight cancellations and other problems.  The crisis is especially surprising considering that there are simulation tools, such as Ansys EMIT, which can predict these interference effects and provide guidance for mitigation.

The image below shows an analysis of a typical RF system, with multiple antenna and radio devices  analyzed in one system. The system can be the plane itself or, in this case, a combination of the plane and the terrestrial 5G system  near the airport. EMIT uses models from a variety of sources. In this case, it can use rigorous field coupling models generated by field solvers such as Ansys HFSS and its related asymptotic electromagnetic solver, SBR+.

For difficult interference problems, the Ansys EMIT toolkit, an integral component of the Ansys Electronics Desktop and part of the Ansys HFSS portfolio, is designed to consider wideband transmitter emissions and assess their impact on wideband receiver characteristics. In addition, any issues found are visually identified by a trace-back method, allowing the designer to easily understand the source of the interference.

EMIT considers both in-channel and out-of-band effects. Beyond transmitters and receivers,  antenna systems must also be taken into account. Ansys HFSS is the industry standard for modeling the physics of antenna systems, their installation effects, and their couplings, even over long distances with ground and terrain reflections.

For more information about predicting radio frequency interference with Ansys EMIT, see this webinar on demnd. (https://www.ansys.com/resource-center/webinar/ansys-in-action-ansys-emit)

Stay tuned for a closer examination of this potential 5G conflict with the radar altimeters employed in aircraft safety systems!

Also Read

Can you Simulate me now? Ansys and Keysight Prototype in 5G

Cut Out the Cutouts

Is Ansys Reviving the Collaborative Business Model in EDA?


TSMC Earnings – The Handoff from Mobile to HPC

TSMC Earnings – The Handoff from Mobile to HPC
by Doug O'Laughlin on 01-20-2022 at 10:00 am

https3A2F2Fbucketeer e05bbc84 baa3 437e 9518 adb32be77984.s3.amazonaws.com2Fpublic2Fimages2Fc9a34855 deed 40fb 904d 95104dc445a6 1480x672

Hello! The most important semiconductor company in the world reported earnings last night. It’s been something of a tradition to post Taiwan Semiconductor Company (TSMC) earnings posts not behind my paywall, and I think that I’m going to continue that to kickoff each earnings season.

There are so many threads in the TSMC call that I want to talk about, but the big one that I think we will look back on 2022 for TSMC is that this is the year HPC will become the largest part of the business. Let’s expand.

TSMC’s current business mix

Smartphones and HPC are neck and neck in the largest buckets for TSMC, but HPC is growing faster. The reason for sluggish smartphone growth was laid out pretty well by TSMC, and that’s volume growth in smartphones has topped out.

Yes, I think — let me add. The global smartphone unit growth last year is about 6%. So some of the — you see some of the company smartphone revenue may grow, it could be due to the pricing. But we — our pricing strategy, as you understand, is strategic, not optimistic. So we’ll grow with the smartphone units in our business.

Well TSMC just guided for “high 20s” growth, and long-term growth of 15-20% CAGR. They additionally guided to an accelerating 2022, and strong sequential growth into Q1. Given that they think that their business will grow in-line smartphone units, the only logical growth driver is HPC. I did some pretty simple math to back out what the HPC segment should look like given their assumptions on growth.

Let me answer the platform question. In 2022, we expect the HPC and automotive to grow faster than the corporate average. IoT, similar. Smartphones close to the corporate average. That’s the platform growth.

I’m a bit more bearish on smartphones growing 20%+ unit growth so let’s say smartphones grow at 15% next year, and DCE / Other grows at 10% as well. I grew automotive at 50% and IoT at ~28% – and the result is that HPC revenue crosses over smartphones in 2022. I use HPC revenue as the plug to then hit the ~28% revenue number. It looks like this is the year HPC is finally larger than Smartphones at TSMC.

For the longest time, I have believed that the largest incremental dollar pool of revenue growth will be HPC. It’s nice to see it come true and TSMC confirms that this is their belief as well. I first wrote about when I suspected meaningful growth in the data center in 2022 after Facebook’s earnings. I got it confirmed shortly thereafter by AIchip’s results. I had a suspicion that data center would be strong, but hearing the largest fab in the world expect something akin to ~40%+ growth in this segment is pretty mind-boggling even to a huge bull like me.

Another point to the leadership of the data center going forward is that HPC is starting to adopt the smaller nodes faster than smartphones, which used to be the premier first adopters of TSMC’s newest node. In the past, HPC would adopt the newest node a year after smartphone, but now HPC is in the driver seat and will be adopting N3 at the same time smartphone is. Beyond just node adoption, I’m pretty bullish on data center exposed stocks like Marvell and Nvidia.

Speaking of Marvell and Nvidia one of the questions on the call was “how can you grow your revenue faster than your fabless customer’s expected revenue growth”. TSMC answered that they believe it’s pricing and share gains.

This is C.C. Wei. Actually, the growth in 2022 is all the above you just mentioned. It’s a share gain, it’s the pricing and also its a unit growth. Did I answer your question?

Part of this is that Intel is starting to outsource to TSMC and that foundry likely will grow faster than memory this year. But I have a hard time believing another obvious answer is that the fabless estimates are too low.

Given that TSMC just guided to accelerating revenue (25% 2021 growth to 28%+ 2022 growth) and has over 50% of global market share, I have a hard time believing that the industry is going to meaningfully decelerate while TSMC revenue explodes. The numbers don’t reconcile. And that is why I believe that the fabless companies’ revenue estimates are likely a bit too low. Also that their 9% industry growth number is likely too low. Getting the theme here?

I believe that 2022 is going to be another strong year, and that almost every fabless company’s numbers will be revised higher. Let’s turn next to the capex side of the equation.

TSMC Expects to Spend $40-44B on Capex

Not only was growing faster for longer a surprise but the $40-44b capex was a real shocker. For context, the most bullish estimate on the street was at ~$40 billion. The upside is now the new downside case. Given that WFE grew by ~40% last year, and TSMC grew capex by ~77% in 2021 over 2020, this is pretty meaningful growth. In absolute terms, they are adding more spending in 2022 than in 2021. But of course, this is a deceleration on a larger base.

I think that the preliminary read-through is that WFE is going to have yet another good year. I believe that WFE likely is more to the tune of 20% growth than to 10% growth. Speaking of 10% growth – this estimate by SEMI came out on January 11th called for 10% growth and after TSMC’s spending estimates it already seems like this will be false. 2 days and it’s already out of date! The true number is going to be higher.

As we discussed on the VLSI semicap comparison of numbers bottom-up to top-down, it seems like estimates need to move higher. I think this is great for semicap broadly (surprise!). If you’re a long-time reader of the substack, one of the core beliefs is that the rising capital intensity of making a semiconductor accretes to fabs and even more so to semicap companies (ASML, LRCX, AMAT, KLAC, TOELY, etc).

This is just another indication that the thesis is correct given that Capex is growing faster than revenue. Which brings me to an interesting question – how could TSMC ever support this kind of spending indefinitely? The answer is that they are either utterly wrong about their growth and are going to throw the entire market into overcapacity, or that demand is still being underestimated. I believe that it’s the latter, as I wrote in my cyclical to the secular thought experiment. I believe that TSMC believes this as well, and given how they are investing and guiding, I want to call this TSMC’s bold bet.

Growing for Longer – TSMC’s Bold Bet

A recurring theme of the analysts calls with TSMC is that every quarter analysts pepper management with “how can you maintain the margin with this investment?” and “you’re spending a lot on capex will this ever normalize?” questions. The answer that TSMC answers each quarter is somewhere along the lines of “We are going to grow trust us”. This first long-term guidance in a while is an indication of that.

We expect our long-term revenue to be between 15% and 20% CAGR over the next several years in U.S. dollar terms, of course, fueled by all 4 growth platform which are smartphone, HPC, IoT and automotive.

The staggering thing I want to point out to you is their 10-year revenue growth CAGR is 14%. That’s the kind of growth that got them to the largest fab in the industry, yet their long-term revenue guide is now actually a call that their revenue will accelerate on a larger base. It’s impossible for them to gain share at the rate they used to so the only answer is the entire industry must accelerate as well.

TSMC is probably one of the best management teams in the entire industry with the most credibility you can ask for. They are prudent, ROIC focused, conservative in their node shrinks yet aggressive in their capital spending. Simply put they do not miss. If they are investing in larger amounts for accelerating growth they believe will come, I am going to believe them.

This is the definition of long-term thinking and bold bets. They are pushing forward at an even faster pace at the peak of their dominance in order to ensure they continue to hold share. And everything is pointing to the diversity and strength of the entire semiconductor ecosystem, and I think that the answer is clear. The 2020s are going to be a better decade than the one before it for the entire semiconductor ecosystem.

Passing Price

I want to briefly mention the gross margin part of the equation. Every quarter there is a lot of hang wringing about the sustainability of the gross margin at TSMC. Last quarter analysts got really hung up on “51% or greater” long-term margins and asked in as many ways as possible if that margin was sustainable.

This quarter of course they put up 53% gross margin and now are guiding to “53% or greater” margin longer term. The bar of course has shifted higher. The right answer and framing around the gross margin sustainability debate are that TSMC really is one of the only games in town, and the demand for their capacity is intense. I mentioned briefly that TSMC can just pass price as much as they want in the Rising Tide of Semiconductor Costs and I think that will continue.

No matter how much capex spend is required and how much depreciation and amortization will grow as a part of TSMC’s cost, TSMC is simply not a price taker. They will raise prices and pass their costs onto their customers, and in this case, it seems like they are able to pass on more than just the cost they take. If they can maintain 53%+ margins against rising CoGs, this means that customers will be taking price raises on the chin. Because what other choice do they really have? Intel’s Foundry business is still more of an idea more than a meaningful business, and Samsung is growing but relatively small. TSMC will get the money that they are due.

There’s a lot more in the transcript itself, which I recommend reading if you have some time. TSMC continues to believe I think that the continued prepayments by their customers are another indication that the fabless companies get it as well. They want more capacity because their businesses are well but they are capacity constrained.

An interesting idea I had was that the capacity precommitments in order to secure capacity feels a bit like the ASML investment by INTC / TSMC. It’s clearly a greater good, there is really only one company that can achieve it, and it’s going to cost a lot of money. In order for the economics to work at TSMC will need a lot of money.

Anyways that’s it for today. I just wanted to cover these points for now, and I’ll be posting a lot more content like this but for the ~100s of other semiconductor companies that will be reporting in the next month. I just always love to start with the biggest and baddest first.

From Fabricated Knowledge

Let’s learn more about the world’s most important manufactured product. Meaningful insight, timely analysis, and an occasional investment idea.


Podcast EP134: The Impact of Using a Physically Aware NoC with Charlie Janac

Podcast EP134: The Impact of Using a Physically Aware NoC with Charlie Janac
by Daniel Nenni on 01-20-2022 at 10:00 am

Dan is joined by Charlie Janac, president and CEO of Arteris IP. Charlie’s career spans 20 years in multiple industries, including design automation, semiconductor capital equipment, nanotechnology, industrial polymers, and venture capital.

Charlie discusses the benefits of using network-on-chip, or NoC IP on several types of design projects. He also discusses the substantial benefits of using physical awareness for the NoC and how to set up this valuable capability.

The views, thoughts, and opinions expressed in these podcasts belong solely to the speaker, and not to the speaker’s employer, organization, committee or any other group or individual.


Forty Four Billion Reasons Why TSMC Remains Dominant

Forty Four Billion Reasons Why TSMC Remains Dominant
by Robert Maire on 01-20-2022 at 6:00 am

Forty Four Billion Reasons Why TSMC Remains Dominant

-Chips for America is better than nothing, but not much
TSMC $40-44B Capex crushes competition (Intel & Samsung)
-Additional efforts other than handouts are necessary
-Could/should TSMC be “adopted” as a US company?

Competition can’t keep up with $40-$44B of Capex

TSMC’s recent announcement of $40-$44B of capex spending is a mind blowing number that is a large increase from last years huge number and makes them far and away the biggest spender period.

Both Intel and Samsung can’t hope to spend anywhere near that number.
If we assume some is for expansion of existing non leading edge capacity but the majority is aimed at the leading edge it means a dominant share of capacity and a very dominant share of EUV and other advanced enabling tool sets.

$52B Chips for America is barely a rounding error

When you assume that the Chips for America act is a one time, one shot disbursement spread over a number of years and a number of companies it becomes clear that its not much against TSMC’s spend.

It also does not compare to what China as a whole is spending on semiconductor technology.

Basically the US is being outclassed and outgunned by both China and Taiwan ( probably part of China in the not too distant future).

Even if Intel got the whole $52B it still couldn’t keep up as the spend would be over several years. Never mind that only $10B of the $52B is for fab projects with a $3B limit per project. Essentially the $52B will be spread so thin as to be ineffective versus the focused sharp spend of TSMC.

Can the US fabs being built make a difference?

Intel announced two fabs in Arizona at $10B each along with TSMC announcing a 5NM fab in Arizona which by the time its operational will be a drop in the bucket trailing edge fab perhaps meant to mollify the US.

Samsung has announced a $17B in Texas in addition to existing facilities there.
It looks like Intel has chosen Ohio for its “megafab” project and Micron is eyeing North Carolina.

While details are scarce, it sounds like the Intel Ohio and Samsung Texas fabs are the most impactful on the US. Samsung would be somewhat less impactful as we assume that bleeding edge technology R&D will continue to be done in Korea making the Texas fab a “fast follower” much as the existing Samsung fab in Texas is today. That leaves Intel Ohio as the only trail blazing R&D facility in the US.

It also remains to be seen if the brain trust in Portland can either be moved or shared with Ohio or if Portland remains the R&D center with Ohio for production.

Excellent CNET article

All this begs the question as to why we can’t get TSMC in the US?
(In a real way)

Obviously TSMC’s brain trust is in Taiwan but could it be both re-created and/or moved to some extent with the right incentives? Rather than a token trailing edge fab in the US, a real leading edge state of the art complete with R&D?

Maybe grant citizenship to those employees willing to move to escape a potential China take over or at least hedge against it.
Throw a lot of money at them, along with a nice life in the US.

It might be a much more reliable and expedient way of fixing the US’s chip problem as compared to trying to rebuild Intel or take hand me downs from Samsung.

Right now time is not on the US’s side and we need chip capacity and expertise fast. We cannot wait ten years for Intel to build a fab and get critical employee mass in Ohio starting with a literal greenfield.

We need to think outside the box or get left behind

Incentives and disincentives are needed in addition to Chips for America

The cost of a semiconductor fab is similar around the world as the majority is the cost of the equipment which costs the same wherever you are (or should it?)

Could/should the US put export taxes on US semiconductor equipment (whether made in the US or a foreign subsidiary but with US technology)
Maybe a 50% tax on tools going to China, maybe 25% on Taiwan. Penalize those companies who would try to trans ship equipment to China. Use the received funds to support the US chip industry.

Incentivize all the companies, Intel, Micron, TSMC and others to build fabs in the US because the primary cost, equipment, would be cheaper here.

We could also go back to prior export restrictions that limited US tool exports to equipment that was several generations old. This technology export restriction worked well with SMIC and China for years but was removed over the years as relations improved (even though relations are now souring again).

The US seems to forget that the majority of semiconductor tools have their basis in US technology (including EUV made by ASML).

Is it better to throw money at the problem as compared to controlling our own fate through our own technology?

Where are the non-handout protections and incentives in the Chips for America act?

In a zero sum game equipment makers don’t get hurt, consumers pay a little more for US made…

As the world needs a certain amount of semiconductors and it takes a certain amount of equipment to make them. Equipment makers would sell the same amount of equipment just to different end locations.

Any chip maker, foreign owned or US owned would benefit by having a fab in the US.

End consumers would likely have to pay a little more for chips made in the US rather than China , but I would rather pay a little more and be able to buy a car and a computer when I need one and not held hostage.

Removing an attractive target- “Broken Nest”

From a political and global strategic perspective it is the existence of TSMC in Taiwan that likely creates additional instability in the region vis a vis China. If Taiwan were a backward Island of farmers China would not be so focused on getting hold of it and would likely not be doing as much saber rattling.

It is the mere fact that Taiwan contains exactly what China desperately needs right now, semiconductor dominance, that makes it such an attractive target, worthy of a lot of effort and bloodshed to obtain.

Remove semiconductors from Taiwan and suddenly the target’s value is greatly diminished and maybe not worth starting a war over.

This is not unlike the middle east and its supply of oil that made it so strategically important. If it were just a desert with camels, no one would care.

Could Taiwan be less attractive if China wasn’t getting TSMC? Could that cool the current tensions?

Broken Nest: Deterring China from invading Taiwan

The stocks

The stocks have lost a little steam as the rapid run up has lost some momentum with less new news and concerns about some sort of peak.
The TSMC capex news underscores that 2022 will continue to be a very good year for chips and likely better than 2021.

Although there is the possibility of slowing, it appears that current momentum remains strong enough to make it a very good year.

We remain most positive on the arms merchants…the semiconductor equipment stocks as its clear that “wartime” spend in the chip industry continues to increase benefitting the tool makers.

While TSMC stands to make a huge amount of money from the continued supply demand imbalance it seems to be plowing most of that money back into capex so its unclear exactly how much will result in increased profitability.
This is even more so the case with Intel that has more spending projects than they currently have money for. First Intel has to figure out where the money is coming from. It looks as if they have already committed their entire market cap in spend over all the projects they have announced.

Even if the Chips for America act is a “gift” to Intel, they are still billions short of what they need.

Samsung will not be left out and will likely double down as well.
The main risk we see, at least in the near term remains supply chain issues limiting tool makers and other kinks in the supply chain like fires in Berlin and new variations of Covid locking down China again.

The usually seasonally weak Q1 is not

Perhaps one of the best signs we see is that usually Q1 is the weakest quarter of the year for chips as we are in a post partum depression after the selling frenzy of the holiday season coupled with a couple of weeks for Chinese new year. Q1 is also usually the low point for memory pricing again related to slower demand.

We are not seeing as much of the normal “seasonal” slowing in the industry. Demand for chips seems undaunted. We had previously thought that we would start to see an easing of the crunch in 1H of 2022 but it seems to be longer lasting.

We still question how long it lasts as the industry remains a cyclical one by its very nature but demand remains overwhelming for the time being.

We may see some psychological weakness as investors are concerned about a cyclical peak or Covid news. Even though business increases may be as good as 2021, we don’t think the stocks will move quite as quickly as they did last year and may prove more volatile on the way.

For now, memory seems stable, but that is usually the first thing to fall. Memory makers seem to be maintaining a rational spend and technology pattern that continues to support profitability. If anything , Micron is probably more undervalued given the current circumstance.

Chip equipment companies need to be razor focused on execution as the demand is there if they can make product.‌

Also Read:

“Too Big To Fail Two” – Could chip failure take down tech & entire economy?

Semicon West is Semicon Less

Supply Chain Breaks Under Strain Causes Miss, Weak Guide, Repairs Needed


Embedded Logic-NVM Solutions for Al Chips

Embedded Logic-NVM Solutions for Al Chips
by Kalar Rajendiran on 01-19-2022 at 8:00 pm

What is Analog NVM

Last month, eMemory Technology hosted a webinar titled “eMemory’s Embedded Logic-NVM Solution for AI Chips.” While the purpose was to present their embedded Logic-NVM solution, the webinar nicely sets the stage by highlighting Analog NVM’s value as it relates to neural networks. Of course, the algorithms of neural networks are  core to implementing AI chips, especially in weights storage. Dr. Chen, the presenter is a manager of one of eMemory’s many technology development departments. Following is a synthesis of the salient points I gathered from the webinar.

Market Trends

There is a massive migration of AI processing from the cloud to the edge, enabled by emerging AI algorithms. Fast growing AI applications are many, such as data inference, image and voice processing and recognition, autonomous driving, cybersecurity, augmented reality, etc. In order to develop efficient AI chips for these applications, it is important to implement various types of Al processing elements (PEs) with low power consumption and high computing performance.

Neural Networks

Artificial Intelligence (AI) is about emulating the human brain. Human brain, of course consists of many neural networks and neurons are the structural and functional units of these networks. It is neurons that perceive changes in the environment and transmit information to other neurons through current conduction.  The neurons can be divided into four parts, namely the receptive zone, trigger zone, conducting zone and the output zone. The basic architecture of an electronic equivalent of the human neural network must include corresponding zones/layers.

Analog NVM

Implementing a Neural Network electronically is achieved through a Multi-Layer Perceptron (MLP) structure. The MLP consists of an input layer, hidden layers, and output layer that are all connected via weights (the electronic equivalent of synapses). The input layer is mapped to data inputs and weights, the hidden layer to the net-input function, and the output layer to activation function and output.

Interestingly, this kind of mapping architecture is comparable to a non-volatile memory array architecture. Refer to Figure below.

For an NVM array, the data input of the NN is the WL data input, weights are stored in the NVM units to do a multiply-accumulate (call MAC) process, and finally, the output data is through the activation function(ADC)  to generate. These output data could be for making decision or transferring to the next PE system. The trick with emulating weights behavior is accomplished with different current levels by leveraging the NVM cell’s data retention capabilities.

Why Analog NVM

Refer to Figure below for two different architectures for designing AI Inference chips.

The Von Neumann architecture approach consumes lot of power due to the SRAM-based processing elements. As high power consumption cannot be tolerated by edge computing applications, In-Memory Computing architecture is the preferred approach for now. By leveraging the analog NVM characteristics, this approach can lower the power consumption and simplify the implementation at the same time.

The power consumption savings on a data inference application could be 10x-1000x using in-memory computing architecture implemented using analog NVM. Fast growing AI applications that were mentioned earlier in the market trends section can all benefit from lower power consumption.

eMemory’s Analog NVM IP Offering

eMemory’s Analog Memory is floating gate-based, and is built on embedded logic compatible memory process that uses a single poly layer. It allows precision current controllability using a smart PV function circuit to support multi-level cell current that can support 4bits~5bits accuracy. The NVM IP demonstrates good data retention and a very low error rate in eMemory special analog IP design.

Refer to Figure below for details about eMemory’s NVM IP.

 

This was developed in collaboration with one of their customers. It is important to note that no extra masks were needed and the manufacturing followed the foundry’s baseline process.

Realizing CIM with eMemory’s Analog NVM IP

eMemory’s team built neural network (NN) processors using floating gate-based NVM to emulate MLP and model a compute-in-memory (CIM) TensorFlow. eMemory’s analog NVM demonstrated excellent control of current with low standard deviation and error rate.

 

The four- major step process for realizing CIM using eMemory’s Analog NVM IP is as follows:

  • Choose the appropriate NN model for the application
  • Use “Software” such as the open-source TensorFlow to help build the new specific NN model as well as the training model in the AI Chip
  • Design the memory cell/array to fit the type of weights and various degrees of accuracy needed such as 2bits/4bits/5bits or higher.
  • Build the peripheral circuits such as smart PV function, activation function, precision ADC/DAC part

Summary

The flow presented in the webinar is one way to implement CIM using analog NVM in an AI application.  Alternate flows can be implemented for one’s specific application through collaboration between software and hardware engineers. The Q&A session after the webinar provides some guidance on handling different types of neural networks and common questions you may have. You can access a recording of that entire webinar from eMemory’s Resources page.

For more information about their IP offering, you can contact eMemory Technology.

Related Blog


Conquering the Impossible with Aspiration and Attitude

Conquering the Impossible with Aspiration and Attitude
by Daniel Nenni on 01-19-2022 at 10:00 am

Conquering the Impossible with Aspiration and Attitude

Cornami is an interesting company. Their leader is also an interesting person. I’ve interviewed Wally Rhines many times on various topics. All of them have been big hits. Cornami is focused on enabling fully homomorphic encryption (FHE) in a commercially viable and widespread manner. Many say this isn’t possible, but Wally and several other high-profile supporters of Cornami believe otherwise.  I covered some of the technical details behind Cornami here. This discussion isn’t about what Cornami is doing. Rather it’s about the people and the process that facilitate conquering the impossible.

Cornami recently released a white paper on the backstory of how Cornami is doing what most say can’t be done. It’s written by a good friend and long-time collaborator, Mike Gianfagna. Mike digs in to understand why some very prominent folks got behind Cornami and its mission. These are folks who judge the efficacy of business plans and technical ideas for a living. Mike explores why these folks bought in to the vision. Mike has seen his share of business plans in his career as well, absolutely, so I wanted to understand what he discovered. The story takes a few unexpected turns and delivers a few surprises along the way.

If you are interested in the benefits of FHE, or if you just wonder what it takes to do the impossible, you’ll want to read this white paper. Here is a summary of some of the twists and turns in the story and the key stars of the show. Don’t worry, I won’t spoil the story. My aim is to shine a light on the how unpredictable life can be when you aim to do the impossible.

Why Should I Care?

About FHE that is. The answer has to do with data security. Most will understand the importance of protecting data. Compromises can cause all sorts of havoc. The actual value of data is growing exponentially and so protecting that value is another aspect of the problem. One of the stars of this story coined the term, “data is the new oil”. There’s a lot that goes with that. You can dig into more details about data security. Just remember FHE is the most bullet-proof way to protect data. Even quantum computing can’t break it.

An Evolving Collaboration

Gordie Campbell founded Cornami years before Wally joined as CEO. Gordie’s story could occupy a white paper all by itself. Gordie founded and became chairman and CEO of CHIPS and Technologies, the world’s first fabless semiconductor company. That innovation drew the famous comment from Jerry Sanders of AMD, telling Gordie publicly that “real men have fabs”. The rest, as they say, is history.

Years later, Gordie and Wally’s paths would cross. The reason had to do with software discounts. You’ll need to read the white paper to learn more. After Wally left Mentor/Siemens, he was doing some work for the Defense Advanced Research Projects Agency (DARPA). The agency had a particular interest in, you guessed it, FHE. The Department of Defense (DoD) was convinced this was a critical technology for the future. After a lot of research, Wally concluded that no one was working on FHE. It would be at least ten years before anything remotely resembling a solution could be in hand.

How did Wally go from “no way” to CEO of Cornami? There are other star performers involved. You’ll have to read the white paper.

Smart Money

My overview will end with Eric Chen of Softbank. Eric led the most recent funding round for Cornami. Eric has a background in physics and a PhD in electrical engineering from Stanford University. He is first and foremost a technologist. He has fueled that passion as both an entrepreneur and an investor. Way before meeting with Gordie and Wally to discuss Cornami’s funding round, Eric had done substantial research on data, its impact on the economy and how to protect it. He knew a lot about, you guessed it, FHE. Why did he invest?  You’ll have to read the white paper to find out.

To Learn More

There’s a quick synopsis of a rather engaging white paper on what’s involved in conquering the impossible. I highly recommend it. You can get your copy here.

Also read:

CEO Interview: Wally Rhines of Cornami

I Have Seen the Future – Cornami’s TruStream Computational Fabric Changes Computing

Podcast EP65: Trust But Verify – The Backstory of Applied Materials and Cornami with Wally Rhines