feature article
Subscribe Now

Unlikely Pairings

Recent Atypical Mergers

Companies come; companies go. I don’t focus a lot on who buys whom – there are plenty of folks breathlessly watching that stuff, so I mostly leave the drama to them. After all, it’s an age of consolidation and accumulation of immense corporate power. So your typical low- to mid-level merger may not be particularly noteworthy.

But lately, there have been a couple of mergers/acquisitions that have had some unusual features to them. Add to that the fact that they’re companies we’ve looked at before, and it seems worth spending some time on them.

An EDA Merger… Or Is It?

You may recall a company we’ve looked at several times that lies, you might say, at the fringes of EDA: Coventor. They’ve made their name in two spaces: process simulation (if you’ve seen videos of a chip being built, layer by layer, it may well have come from their SEMulator 3D tool) and MEMS, with their MEMS+ tool.

Being a smaller company – and not being directly involved in the design of high-volume CMOS chips – has positioned them outside the EDA mainstream. But we all know how things go in EDA: new companies start up, and then they either die, putter along, or, most frequently (if they’ve got good ideas), they get bought by one of the Big 3: Synopsys, Cadence, and Mentor.

So you might think it natural that this would be the future for Coventor. Synopsys already does some of what Coventor does with their TCAD-oriented tools, so perhaps they could swallow an otherwise pesky competitor – if they were feeling the pressure. Alternatively, one of the other guys could have bought them in order to compete better with Synopsys on that front.

So it was not at all obvious when they were acquired by… Lam Research? Lam makes semiconductor equipment; they play in a market with other major players like Applied Materials, ASML, and KLA-Tencor. In some areas, these companies compete; in others they have particular pieces of equipment that dominate the market.

You could imagine a dynamic where, as with EDA, a small company might come up with some clever new tool (although it takes much more investment for a hardware tool than for even a complex software tool). And that might make them attractive to one of the Big Guys. But Lam buying Coventor doesn’t fit that model at all.

So what the heck is afoot?

It turns out that Lam made the purchase based on what they perceived as particular synergy. They feel that owning Coventor will improve their customers’ use of their tools. And that’s because they have all of the detailed models of the reaction chambers on their equipment, and they can meld that with the Coventor tools to provide particularly accurate representations of processes that use Lam tools.

Is this available to Lam’s competitors? Well, maybe. Lam will continue to sell the Coventor tools exactly as they’ve been sold so far, so Coventor customers will presumably see little or no change. The familiar management faces that they (and I) are used to seeing – CEO Mike Jamiolkowsky, VP Dr. Stephen Breit, and CTO Dr. David Fried – will still be doing their thing.

Furthermore, the roadmap isn’t changing, and they say that the only operational change that has been reflected in customer questions relates to the protection of customer proprietary information that both companies might acquire in the course of doing business. Both companies have systems for this: Lam’s is auditable, and Coventor’s will be made auditable.

So everyone – including Lam competitors and their customers – will still have access to the Coventor tools. Could competitors do what Lam is doing? In theory, yes. But they’d be on their own. Lam perceives owning Coventor to be a strategic advantage.

This is why, to Lam, their acquisition of Coventor makes more sense than an acquisition by, say, Synopsys would. Lam has the reaction chamber info, and therein lies the synergy. Synopsys doesn’t (nor do the other EDA guys).

Vertical Integration – Upside Down

The typical EDA merger – or, in this case, semi equipment and tools – is a horizontal play: you’re basically dealing with the same customers at the same level of abstraction. Vertical integration, by contrast, would normally involve a company buying one or more suppliers so that they can capture more of the profit and gain some control over critical portions of their supply chain.

In a typical vertical scenario, dollars grow as you go up. That’s kind of obvious, since the higher-level companies are incorporating products from the lower-level companies, so the higher-level revenues need to account both for higher-level profits, administrative costs, the costs of manufacturing, and for the costs of the supplies. The lower-level supplier’s revenues reflect only their supply, which, by definition (in a well-behaved market), is smaller.

Now let’s lay out a vertical market that doesn’t seem to play by these rules: gadgets that use sensors. Near the bottom of the supply chain lie the sensor makers; sensors get incorporated into systems, along with tons of software that makes it all work right. The software is critical for end users, since having to start with nothing but a raw sensor means a hard row to hoe before you’ll have a product.

Here’s what’s wrong with this scenario:

  • Companies like to pay for hardware. You can see hardware; you can touch it; you can toss it at someone in a fit of pique. Software? Meh… not so much. Try wadding an algorithm up into a ball to defend yourself with when getting mugged on a mean street. And getting paid for it can also be, all too often, harder. Just ask anyone that’s been involved with – or declined to get involved with – FPGA design tools.
  • MEMS and other hardware companies require at least an order of magnitude more investment to spin up than a software company does. You might be able to buy a software company for a song compared to what would be needed to buy a going hardware company (without taking an enormous haircut).

So, rather than a higher-level company buying a supplier, in our case of interest today, the supplier has bought the higher-level company: mCube has acquired Xsens. Turns out that Xsens was acquired in 2014 by Fairchild, which was then acquired by ON Semiconductor in a nerdy version of that bigger-fish-eats-smaller-fish-chain image. But ON decided that software was outside their area of focus – and so now mCube has it.

While Xsens sold modules largely used by motion-modeling companies (like animators), their real value to mCube is in the fusion algorithms. That software will be made available as a SaaS tool offering on their majority-owned subsidiary, Ten Degrees. They won’t be selling the software outright, in an effort to keep much of its strategic benefit out of the direct hands of competitors. The SaaS approach cloaks the innards better.

Both Ten Degrees and Xsens will remain sensor-agnostic – mCube says that they will compete with the other sensor makers for use in those platforms. mCube’s edge comes in controlling the algorithms and in leveraging Xsens’s module expertise for planning future sensor features. Which, if theory plays out as they hope, would help them in that competition against other sensors.

 

More info:

Coventor

Lam Research

mCube

Xsens

One thought on “Unlikely Pairings”

Leave a Reply

featured blogs
Oct 5, 2022
The newest version of Fine Marine - Cadence's CFD software specifically designed for Marine Engineers and Naval Architects - is out now. Discover re-conceptualized wave generation, drastically expanding the range of waves and the accuracy of the modeling and advanced pos...
Oct 4, 2022
We share 6 key advantages of cloud-based IC hardware design tools, including enhanced scalability, security, and access to AI-enabled EDA tools. The post 6 Reasons to Leverage IC Hardware Development in the Cloud appeared first on From Silicon To Software....
Sep 30, 2022
When I wrote my book 'Bebop to the Boolean Boogie,' it was certainly not my intention to lead 6-year-old boys astray....

featured video

PCIe Gen5 x16 Running on the Achronix VectorPath Accelerator Card

Sponsored by Achronix

In this demo, Achronix engineers show the VectorPath Accelerator Card successfully linking up to a PCIe Gen5 x16 host and write data to and read data from GDDR6 memory. The VectorPath accelerator card featuring the Speedster7t FPGA is one of the first FPGAs that can natively support this interface within its PCIe subsystem. Speedster7t FPGAs offer a revolutionary new architecture that Achronix developed to address the highest performance data acceleration challenges.

Click here for more information about the VectorPath Accelerator Card

featured paper

Algorithm Verification with FPGAs and ASICs

Sponsored by MathWorks

Developing new FPGA and ASIC designs involves implementing new algorithms, which presents challenges for verification for algorithm developers, hardware designers, and verification engineers. This eBook explores different aspects of hardware design verification and how you can use MATLAB and Simulink to reduce development effort and improve the quality of end products.

Click here to read more

featured chalk talk

Solutions for Heterogeneous Multicore

Sponsored by Siemens Digital Industries Software

Multicore processing is more popular than ever before but how do we take advantage of this new kind of processing? In this episode of Chalk Talk, Jeff Hancock from Siemens and Amelia Dalton investigate the challenges inherent in multicore processing, the benefits of hypervisors and multicore frameworks, and what you need to consider when choosing your next multicore processing solution.

Click here for more information about Multicore Enablement: Enabling today’s most advanced MPSoC systems