What surprises me with Infineon is that they seem to not work on a LLM on a chip. In particular that here in Linz Infineon, Apple's 5G modem designer and the Johannes Kepler University with Sepp Hochreiter (Inventor of LSTM) are just next to each other.
I watched a interview of Infineon head of AI just recently, and I was like excited to hear about some cool hardware. And all he did is talk about how he uses LLMs.
Yeah sadly. I think they have all the know how and people and connections to pull that off. Yet as so often, Austrian (Linz) companies are all about being a supplier not a disrupter.
> Yet as so often, Austrian (Linz) companies are all about being a supplier not a disrupter.
That makes a lot of sense when you consider the origins. After all, Infineon used to be a subsidiary of Siemens, a famously relatively conservative company
Infineon has these weird automotive tricore microprocessors. Going into consumer market it would be expensive and not competitive. Infineon tried that with XMC microprocessors, but they weren’t cheap, had thermal issues and weren’t well received by customers.
Does somebody work on combining flash storage technology with analog computers? After all, a flash storage cell is "just" a FET, so in theory it should be possible to use the cell directly as a weight of an input to some summing amplifier. In other words combine some kind of opamp with a bunch of flash cells to make up one artificial neuron. Should massively increase density and energy efficiency compared to traditional digital stuff. And slight inaccuracies of analog circuits won't matter that much in artifcial neural networks. AFAIU.
Someone needs to still keep making the "literally everything else" chips in the world. I'd rather it be an old name like Infineon instead of them handing their entire business to China in favor of chasing a bubble.
Going into something so far out of their expertise would be (rightfully?) punished by the market. That kind of innovation is more likely to succeed in a startup than in a traditional lifestyle company like Infineon.
Anyone know why it took so long for them to transition to 300mm? As mentioned in the article, they're mainly using existing 300mm equipment designed for silicon, which transitioned years ago. I imagine some process steps are different (mainly growing to larger boule, but probably others), but it doesn't seem like they would be huge blockers.
Making a SiC crystal that's 300mm large is much more difficult than silicon. SiC can't use the Cz process that silicon uses for one. SiC doesn't even melt at normal temperatures. This is great for thermals but bad if you want to use molten phase techniques.
I did my PhD in a SiC lab, working in a tangential topic but in that lab, so I heard a lot about SiC. My take away was that the similarities and transferable aspects of silicon and silicon carbide is that they share the same first seven letters, after that all bets are off. It’s a nightmare of a material to work with.
did anyone say anything about larger die? I think this is a more-die play, so a given defect density won't affect yield. (maybe a small improvement due to less circumference for given area.)
As you say, the main reason for using a wafer with a double area is to double the number of transistors that can be made on a wafer, reducing their fabrication cost.
Nonetheless, the reduction in fabrication cost is greater for large-area devices, for 2 reasons.
One is that on bigger wafers less area is wasted at the periphery, where in many places a full die does not fit inside the circle, so an area corresponding with a fraction of a die remains unused. The amount of such wasted area grows quickly with the size of the die.
The second reason is that the dies that are close to the wafer edge, even if they fit inside they have worse yields than the dies that are far from the edge. Again, the proportion of such dies with worse yields increases with the die size and it is lower for bigger wafers.
Not really. The article is about a larger wafer, not larger dies.
They are mostly cutting transistors, voltage regulators, and such out of those. There's already no practical limitation on the size of those components.
While this is not the principal reason for using bigger wafers, bigger wafers reduce the cost of large-area devices more than they reduce the cost of small-area devices.
European investors and governments are risk averse. DRAM is one of the riskiest silicon products. They talk a lot about independence but European governments lack the raw investment power and willpower to make decade-long investments that are guaranteed to lose money in the next decade as DRAM. If Europe wants DRAM made in European soil, the governments and people should be ready for losing tens to hundreds of billions of taxpayer money.
When Korean and American factories ramp up, the price will plummet. Basically whatever business will not make a single Euro profit for years.
European silicon manufacturers target industrial and automotive for a reason. They want large, guaranteed contracts for older silicon nodes which are cheap and profitable. They let the US, China and Taiwan to take risk of developing new ones. When they do research, they prefer incremental ones.
When it comes to real talk Europe doesn't really care about independence more than it cares about capital and profits. Only an actual war will make the governments care and it will be too late. Without a strong drive to create one's own nuclear arsenal, don't expect any groundbreaking risky investment out Europe.
Then why are they investing 800 billion in weapons? I mean, I get it for the investors. Governments are investing in weapons, so investors build what governments require. In the EU the problem is not building factories, it's selling what's being built while being sabotaged by the EU and it's import treaties letting Chinese vendors massively undercut their own industry to the point of bankrupting entire sectors (solar panels, plastics, now cars). But WHY do EU governments do this? (and why are they lending very large amounts of money to do it)
I mean, they must be aware of the basic facts: if you're making weapons of war best possible outcome is they sit unused in a basement. If they are used, the other guy's weapons will be going in the other direction and your weapons will not just not contribute to the economy, they'll literally have very strong negative contributions. Whichever of the 2 situations you're in, there will be no improvement for the population, no economic growth beyond the initial investment, no nothing. Just a LOT of tax money poured into a black hole in the ground.
I get that weapons lets a lot of political appointees select their own companies and personally make a lot of money in the short term, at incredible cost to everyone around them.
For the governments buying the weapons this is a guaranteed zero-return-on-investment, with loaned money. Does it really need explaining how stupid this is from a financial viewpoint? In a few years this will become a drag on the EU economy of something like 10% of all taxes raised in the EU. And that's assuming they aren't actually used for anything, because then it'll be a 50%+ drag on the economy.
Europe now has war spending 60% as high as it's famously ultra high social spending.
Surely a DRAM industry is cheap in comparison, plus it actually has a chance of paying off.
I doubt it will. You hv to consider memory tech may shift a bit towards LLM-first, so a kind of more computations in memory, and other innovations will be common and demand will not be lower especially if you consider average lifetime of a GPU card is 5 years, there will be demand like just to utilize all those racks to pay out all the money they borrow now. In memory compute will make AI more decentralized (still a hopium).
I think you missed the part where they(SK, Samsung and Micron) are a cartel forming a monopoly, and can control market prices regardless of demand via controlling the supply.
Sure, cartel behavior is illegal, but if you're in bed with the right political administrations, then blind eyes can be turned.
They have been a cartel for decades, the have had boom and bust periods all the same. There's only so much money you can leave on the table before you snap and stab your cartel pals.
Perhaps an uninformed take, but there seems to be nothing about memory packaging that European industry would have much trouble with - except not having an already existing factory and specific know-how. Otherwise, it seems to be precision mechanics (for bonding) and chemistry (for the materials), both of which are bread and butter of European industry.
There is a lot of tribal knowledge and a huge ecosystem of vendors needed to support packaging.
Additionally, most countries in Asia as well as individual states in the US are giving massive subsidizes to help retain the existing ecosystem of vendors.
It is difficult for the EU to directly compete head on when the ecosystem has largely existed outside the EU for almost 30 years.
>You can likely live without a new gaming rig. You likely cannot live without replaceable parts for trucks and cars.
Bad logic for an argument. With this logic you can say you can live fine with your 30 year old diesel car that doesn't need these new cutting edge SiC wafer chips, and that making diesel engines is more important for national security than making chips, and so on and so forth till you reach knowing how to make fire, swords an arrows as the most important for national security.
Truth is fabbing CPUs and GPUs is miles more profitable than what Infineon and EU are fabbing, and they're just as important for national security/sovereignty if not even more:
Did Infineon's engineers use EU or US CPUs and GPUs for the complex CAD/CAE simulations needed to develop those SiC wafers? Why else do you think the US got TSMC and Samsung to make cutting edge fabs on US soil if CPUs and GPUs are not important for national security? When you need to process TB of live battlefield data in case of war, do you not need domestic CPUs and GPUs?
Ten years old CPUs are "okay" and there are fabs in Europe that could make them. Faster is better, but slower isn't entirely useless and hugely better than no CPU. Like a 100 hp heavy truck is greatly underpowered, but almost infinitely better than no truck.
Yeah but in a war, be it a trade war or kinetic one, against a power WHO HAS modern chips and can process data and intelligence faster, who has better leverage? The one with cutting edge tech, or the one with "okay" tech?
Seeing how the orange bafoon slapped the EU around with his trade war and received no retaliation that could cause even paint in return, I assume the answer is obvious.
Another point being how Ukraine can beat Russia in drone warfare simply because they have access to the superior tech of all western allied nations despite making lower numbers than Russia which makes more but can't easily get embargo chips.
>and there are fabs in Europe that could make them.
Which EU fabs make 14++nm CPUs and GPUs? Which EU domestic CPUs and GPUs IP designs are there at the moment that can be fabbed in case of emergency without parts or IP from rivals?
The truth is EU is painfully absent from this field. Sure, in theory they can duct tape together some RISC-V CPU and some ARM MALI GPU production on Dresden 16nm fabs, in case of a war, but since the EU never built and shipped a commercial domestic CPU and GPU from scratch in the last 20+ years, it's not gonna be as easy as turning a screw, as it is for nations that ship new cutting edge compute chips every year: US, Korea, Taiwan and to an extent Japan and China. When you've been out of the game for so long, you get rusty and comebacks are slow and painful. EU would take decades to catch up and rebuild this industry in case of a war cut them off from these parts. It's basically another supply chain vulnerability it failed to address.
Regarding the fabs (well... fab - not good), GlobalFoundries Dresden, and regarding IP, ARM. Or old AMD or Intel designs, because intellectual property goes right out of the window in a war.
Which was the last ARM CPU that's been full-stack designed and made in the EU, that they can pull out of the mothball and iterate on if need be?
Just because you own an ARM license and know how to make one in a lab or university research group, doesn't mean you know how to make and ship competitive chips as the nations that actually do this for a living daily.
> Or old AMD or Intel designs
Firstly, where will the EU get old Intel and AMD designs? Are the Verilog files, netlists and photomasks just floating around?
Secondly, if you need to bootleg the obsolete 20 year old designs of your adversary to fight them with, then you already lost the war before you even started. Which is why the EU lets itself bullied and pushed around by everyone. You have no sovereignty when you're at the mercy of everyone for the vital things you need.
ARM is a UK company, of course they have the IP. While their highest performance CPU cores have been designed in the US in recent years, they also have design teams in the UK and France. I don't doubt that their European part can adapt an otherwise finished design to a different process.
Intel or AMD IP would need to be stolen somehow. GlobalFoundries might still have some old AMD materials around. Yeah, this does seem really difficult - it would be easier and less risky to run x86 code on ARM if necessary.
Lastly, a factor of 5 or so in computing power does not decide a war.
>they also have design teams in the UK and France.
That's not enough. You need teams that have experience and history of shipping complex full stack designs from start to finish, because that's what the US (and China) has, but the EU does not. Thinking you can just turn a screw and have those teams get proficient and catch up with decades of lost experience in shipping complex silicon designs is wishful thinking.
At least the US had enough self reflection and foresight to realize they de-industrialized too much and so have on-shored a lot of manufacturing back, from cutting edge semi to ship building, using their Korean and Taiwanese partners. The EU has not and remains much more venerable, with its only strategy being wishful thinking, hopes and dreams that the status quo and favorable alliances last forever, which history proves they don't.
>Lastly, a factor of 5 or so in computing power does not decide a war.
What about 5x in everything else, not just compute?
The eu doesnt plan to go to war, so the strategy revolves around that. They just need to be thorny enough to make attacks painful and then sit behind their nuclear Shield and allies
>The eu doesnt plan to go to war, so the strategy revolves around that.
As an EUropean I'll say that's where the big problem lies with the EU's fairytale idealist pacifist strategy: that its limp-dick approach being the key to endless peace and prosperity, it only (barely) works inside Europe with pacifist liberal equal-minded nations sharing a common identity and values, but doesn't actually work in real life outside the EU in a world where you're surrounded by hyper-nationalistic predatory thugs with imperialistic tendencies who want to see you take you down so that their empire can dominate you, and it ends up affecting us anyway leading to EU being forced to make concessions to these extortionist thugs (Trump, Putin, Erdogan, Aliyev, Modi, Xi etc) who know the EU will always sacrifice the financial welfare of its own citizens in order to pay up racketeering fees to avoid any and all direct confrontations, leading to a slow but constant decline in its global influence and economic prosperity.
That's why guerilla soldiers in conflict areas around the world typically kidnap EU tourists but not American ones, because they know EU nations are so weak they'll just pay them ransom, while they're afraid the US will send some military team or a drone strike and kill them.
The only guaranteed thing in real life, is that (someone else's) war always finds you anyway, whether you want to or not (see Trump's tariffs and his conflict with Iran how much that affected EU economy and consumers). So it might as well find you well prepared for it, which the EU sadly never is, because it holds on to an outdated mindset from the peaceful period of the post-USSR collapse, that was only a once in a lifetime exception in human history, in a world slowly returning to default factory settings, which is "might makes right".
So "better to be a warrior in a garden, than a gardner in a war."
>They just need to be thorny enough to make attacks painful
How did that work against Trump's tarif war? The truth is that you're right, but the EU isn't thorny enough, and that's the problem.
>and then sit behind their nuclear Shield and allies
And what happens when your allies don't feel like being your allies anymore and instead turn into economic adversaries working to undermine your economic prosperity to further their own in a zero sum game? How does your nuclear shield help then? Is the EU gonna nuke the US over an economic trade dispute? Who do you think has more nukes and better shields?
The main reason why guerillas kidnap EU citizens is that the US have a policy of not paying ransom. Which is game-theoretically smart, but gets really bad PR the first few times, and if I knew the people sacrificed the first few times, I might have a different opinion, too.
That also encourages everyone else to go kidnap your citizens knowing you're a free walking ransome bag waiting to be seized, thereby making the problem worse and increasing the chances that you'd be in such a position.
Now SiC MOSFETs are the most efficient devices for the power converters used in electric vehicles, where DC voltages of 800 V or 1000 V are used in the latest models, which require, respectively, 1200 V or 1400 V SiC MOSFETs.
GaN HEMTs (high-electron-mobility transistors) are the most efficient devices for the power supplies that are fed from mains electricity, where 650 V GaN devices are used for the universal power supplies, which must accept up to 240 V AC on input.
I'm curious about RF uses (esp phased-array radar / AESA). usually GaN is treated as the the best for power/efficiency reasons, but does SiC enter into the picture? not fast enough?
The GaN itself is grown on SiC (instead of Si) for high performance RF, so SiC production capability matters if you want a homegrown supply chain (but you won't use SiC in the signal path itself). Many fabs are on 100mm or 150mm wafers trying to jump to 200mm ASAP.
That's all motherhood and apple pie, but to the GP's remark, why is this supposedly "a huge deal"?
In my mind, it's a process improvement towards cost-effective economies of scale, not a fundamental change in the end product. The PR also mentions Infineon will be "further executing on its 200-millimeter SiC manufacturing roadmap."
I mean nice milestone accomplishment, but are SiC products a major supply chain bottleneck, or can the market reasonable expect unit costs to be chopped in half any time soon?
For now, SiC MOSFETs remain more expensive than silicon IGBTs, even if the difference is today much smaller than a few years ago.
So at any new design, one must ponder whether to use the better SiC MOSFETs or the cheaper silicon IGBTs.
Even if they are more expensive, SiC MOSFETs ensure economies elsewhere, i.e. they enable a higher energy efficiency, which reduces the operational costs for the end user, and also the use of cheaper devices in other parts of the system, which can compensate the higher cost of the SiC devices.
So now any price reduction of the SiC devices is welcome, expanding the range of products for which they will be chosen, thus their market share. The price reduction from using wafers with a double area can be very substantial, not only due to producing a double number of devices with a not much greater cost, but also by improving the yields for large-area devices, because usually the chips that are located close to the edge of the wafer have worse yields.
There already are many recent electric vehicles that use SiC MOSFETs, inclusively from Tesla and this is a market that grows quickly and SiC MOSFETs are a critical component for it, by enabling much better efficiency, size and weight of the electronics, in comparison with the old IGBTs.
I am pretty certain that the new EV chargers that can charge a vehicle in less than 10 minutes, or even in only 5 minutes, by working at voltages of up to 1 kV and currents of up to 1 kA, are made with silicon carbide MOSFET modules and it would have been difficult to make them in the past, when only IGBTs were available.
Vacuum tubes have huge conduction losses, i.e. the voltage drop on them is very high at high electric currents.
This is why for high-power switching they were quickly replaced by gas-filled tubes, long before WWII, e.g. with thyratrons (General Electric) and ignitrons (Westinghouse Electric).
Semiconductor devices have much lower conduction losses than even gas-filled tubes.
While conduction losses are much less a problem with semiconductor switches, many of the semiconductor switches have high switching losses, which limit their use to low switching frequencies.
Silicon carbide devices have lower switching losses than the older semiconductor devices, and gallium nitride have even lower switching losses than SiC devices, so both enable great increases in switching frequencies, which has many benefits.
As a bonus, in the future (when SiC crystals with a better quality will be available) silicon carbide devices will enable the production of high-voltage switches that require much fewer devices connected in series to reach working voltages of hundreds of kilovolt, or even of a few megavolt.
Nowadays, thermionic vacuum tubes or thyratrons remain sometime competitive in pulse switching applications, but only for very high voltages, i.e. above 50 kV. Even for such applications they have a too short lifetime, and there are better semiconductor alternatives, except that this market is too niche, so the big semiconductor device manufacturers do not bother to make such devices, which sometimes forces the use of obsolete vacuum tubes or thyratrons.
Only for the generation and amplification of electromagnetic waves in the microwave frequency range and at very high power, special vacuum tubes may remain preferable to semiconductor devices (e.g. the magnetrons of microwave ovens or the traveling-wave tubes still used in many satellites).
Wooooo let's go back to 1827 for a moment and have a chat with Anyos Jedlik. Or maybe Andres Flocken in 1881. Let's see how they did electric vehicles.
I mean sure, motor control existed before solid state switching devices, but it was also much lower efficiency and used much lower voltages. Power electronics absolutely are critical for the efficiencies and handling characteristics of modern electric propulsion.
Point still stands - no vehicles electric or otherwise were using SiC until around the 1970s (and even then that was really niche applications) and widespread use wasn't until the 2010s.
Until then, we were predominantly using SiO2 MOSFETs and IGBTs.
You will note that timeline also maps quite well on when electric vehicles started to become viable in the market . Of course battery technology also had a lot to do with that, but efficient high voltage power electronics absolutely are an important part of the equation.
Yes?! Infineon's fabs and new SiC wafers aren't built for the small feature sizes of CPUs and GPUs.
Making power FETs uses different process nodes than CPUs and GPUs. All semi fabs are highly specialized: some for power ICs, some for RF, some for DRAM, some for logic, some for NAND.
Wolfspeed/Cree does not have a fab and does not create wafers, let alone LEDs. They are purely a research/licensing company with a side business of selling luminaire fixtures.
Wolfspeed just barely started making anything. Their plant only broke ground recently in 2023, and they did not have anything fabrication-wise going when they first broke off from Cree in 2021.
Wolfspeed didn't break off from Cree, the Cree name was sold off as they moved away from LEDs and became pure SiC/compound semiconductors.
The Marcy fab has been in HVM since mid 2023. Prior to that they had a large 150mm fab in North Carolina since at least 2016. Unless you are referring specifically to 300mm SiC?
Different kind of applications. GaN has no reverse recovery at all and a way smaller gate source capacity and threshold voltage so you can crank up the frequency higher.
But GaN conducts heat worse and also anything that can withstand more than 650V becomes a huge die due to no vertical mosfet tech in GaN yet.
SiC is more or less like super Silicon, almost negligible reverse recovery, very fast switching, way smaller gate charge and a 1200V 1700V is economical and very performant.
For high power like 10s-100s of kW SiC is the best. Higher than that passive components become so big you don't need as high a frequency so you would go with Si IGBTs, lower than that GaN allows you to increase frequency by a lot and reduce passive components size AND in a lot of cases total losses.
So yeah, don't expect to see GaN in a car power train and SiC in a server power supply.
GaN is (much) better than SiC for certain FOMs, but worse in terms of thermal conduction and technology maturity. In addition, you can't yet make "vertical" GaN devices (there's a few flamed out startups that tried) which forces certain undesirable tradeoffs at high blocking voltages. So GaN's sweet spot is 100-650V devices, for 48V and mains voltage applications.
There is an overlap between SiC and GaN at rated voltages between 600 V and 750 V, but at lower voltages than that only GaN devices are used and at higher voltages than that only SiC devices are used.
Even at 650 V where there are available both GaN and SiC devices, there is little direct competition between them. For such voltages, SiC devices are used for controlling greater powers, where also the power dissipated by the device is great enough to require a big package that enables efficient cooling, while the pure GaN devices (i.e. not the hybrid silicon-GaN cascode transistors) are available only in small surface mounted packages, which enable much higher switching frequencies, but at lower output powers, because it is more difficult to cool them well enough.
For now, the SiC devices are only unipolar devices, i.e. MOSFETs and Schottky diodes. But in the future, when SiC crystals with fewer defects will become available, there will also be SiC bipolar devices, i.e. thyristors and IGBTs. The bipolar devices will be available for much greater powers than the current devices, while also being much faster than the silicon bipolar devices. On GaN, bipolar devices are not possible, so GaN will always be used only for higher switching frequencies, but lower output powers than SiC.
Silicon carbide MOSFETs are the best devices for the SOTA electric vehicles and also for their chargers, which work at 800 V (where 1200 V SiC MOSFETs are used) or at 1000 V (where 1400 V SiC MOSFETs are used).
Infineon marks innovation milestone with world’s first 300-millimeter power silicon carbide (SiC) wafer
PBusiness & Financial Press
Nov 09, 2026
Infineon presents industry-first functional power devices processed on 300-millimeter silicon carbide wafers
Technological breakthrough underlines Infineon’s innovation strength and leadership in power semiconductors and wide-bandgap technologies
300-millimeter SiC process feasibility lays the foundation for secure, future-proof SiC supply to customers in automotive, renewable energy, industrial and AI data center applications
Infineon will present the first processed 300-millimeter SiC wafer to the public at the electronica trade show starting 10 November 2026 in Munich
What surprises me with Infineon is that they seem to not work on a LLM on a chip. In particular that here in Linz Infineon, Apple's 5G modem designer and the Johannes Kepler University with Sepp Hochreiter (Inventor of LSTM) are just next to each other. I watched a interview of Infineon head of AI just recently, and I was like excited to hear about some cool hardware. And all he did is talk about how he uses LLMs.
AFAIK, they do power and analog stuff and digital logic for automotive and industry, not cutting edge digital logic at the smallest feature sizes.
Yeah sadly. I think they have all the know how and people and connections to pull that off. Yet as so often, Austrian (Linz) companies are all about being a supplier not a disrupter.
> Yet as so often, Austrian (Linz) companies are all about being a supplier not a disrupter.
That makes a lot of sense when you consider the origins. After all, Infineon used to be a subsidiary of Siemens, a famously relatively conservative company
Infineon has these weird automotive tricore microprocessors. Going into consumer market it would be expensive and not competitive. Infineon tried that with XMC microprocessors, but they weren’t cheap, had thermal issues and weren’t well received by customers.
Does somebody work on combining flash storage technology with analog computers? After all, a flash storage cell is "just" a FET, so in theory it should be possible to use the cell directly as a weight of an input to some summing amplifier. In other words combine some kind of opamp with a bunch of flash cells to make up one artificial neuron. Should massively increase density and energy efficiency compared to traditional digital stuff. And slight inaccuracies of analog circuits won't matter that much in artifcial neural networks. AFAIU.
Answering my own question, googling yields a bunch of links to different approaches doing that and the technological challenges that are involved.
Someone needs to still keep making the "literally everything else" chips in the world. I'd rather it be an old name like Infineon instead of them handing their entire business to China in favor of chasing a bubble.
What's interesting to me in the last ten years we've seen a shift to silicon carbide and gallium nitride power components.
Those are key components in the shift to renewable power. Inverters for solar, motors, battery chargers, even inverter based power transformers.
Going into something so far out of their expertise would be (rightfully?) punished by the market. That kind of innovation is more likely to succeed in a startup than in a traditional lifestyle company like Infineon.
Their black magic is a different breed.
Anyone know why it took so long for them to transition to 300mm? As mentioned in the article, they're mainly using existing 300mm equipment designed for silicon, which transitioned years ago. I imagine some process steps are different (mainly growing to larger boule, but probably others), but it doesn't seem like they would be huge blockers.
Making a SiC crystal that's 300mm large is much more difficult than silicon. SiC can't use the Cz process that silicon uses for one. SiC doesn't even melt at normal temperatures. This is great for thermals but bad if you want to use molten phase techniques.
From the wiki, SiC monocrystals are made in an "Acheson graphite electric resistance furnace."
Scaling this up does not appear to be trivial.
CVD and the "Lely" process are also mentioned.
https://en.wikipedia.org/wiki/Silicon_carbide#Production
SiC is just in general an extremely unforgiving material to work with. The stars just didn't align for 300mm wafers until now.
It's about temperature gradients forming over a much larger surface. You've got 2.4x more area to worry about at 300mm.
Growing the crystals is very slow, so you have a lot more time within which to make a mistake per wafer.
I did my PhD in a SiC lab, working in a tangential topic but in that lab, so I heard a lot about SiC. My take away was that the similarities and transferable aspects of silicon and silicon carbide is that they share the same first seven letters, after that all bets are off. It’s a nightmare of a material to work with.
Larger dies == more material == more chance for impurities to wreck the crystal.
Chasing 9’s is very hard already (both uptime and purity), and by the time you start getting into 6+ digits of 9’s? Really really hard.
did anyone say anything about larger die? I think this is a more-die play, so a given defect density won't affect yield. (maybe a small improvement due to less circumference for given area.)
As you say, the main reason for using a wafer with a double area is to double the number of transistors that can be made on a wafer, reducing their fabrication cost.
Nonetheless, the reduction in fabrication cost is greater for large-area devices, for 2 reasons.
One is that on bigger wafers less area is wasted at the periphery, where in many places a full die does not fit inside the circle, so an area corresponding with a fraction of a die remains unused. The amount of such wasted area grows quickly with the size of the die.
The second reason is that the dies that are close to the wafer edge, even if they fit inside they have worse yields than the dies that are far from the edge. Again, the proportion of such dies with worse yields increases with the die size and it is lower for bigger wafers.
It says right in the title?
Not really. The article is about a larger wafer, not larger dies.
They are mostly cutting transistors, voltage regulators, and such out of those. There's already no practical limitation on the size of those components.
Ah, my mistake. I of course meant wafer in my comment.
See my other comment above.
While this is not the principal reason for using bigger wafers, bigger wafers reduce the cost of large-area devices more than they reduce the cost of small-area devices.
That was not the question I was answering.
Can you guys restart DRAM production in Europe again? If 20 or so years ago it was possible, today is the best economical opportunity.
European investors and governments are risk averse. DRAM is one of the riskiest silicon products. They talk a lot about independence but European governments lack the raw investment power and willpower to make decade-long investments that are guaranteed to lose money in the next decade as DRAM. If Europe wants DRAM made in European soil, the governments and people should be ready for losing tens to hundreds of billions of taxpayer money.
When Korean and American factories ramp up, the price will plummet. Basically whatever business will not make a single Euro profit for years.
European silicon manufacturers target industrial and automotive for a reason. They want large, guaranteed contracts for older silicon nodes which are cheap and profitable. They let the US, China and Taiwan to take risk of developing new ones. When they do research, they prefer incremental ones.
When it comes to real talk Europe doesn't really care about independence more than it cares about capital and profits. Only an actual war will make the governments care and it will be too late. Without a strong drive to create one's own nuclear arsenal, don't expect any groundbreaking risky investment out Europe.
Then why are they investing 800 billion in weapons? I mean, I get it for the investors. Governments are investing in weapons, so investors build what governments require. In the EU the problem is not building factories, it's selling what's being built while being sabotaged by the EU and it's import treaties letting Chinese vendors massively undercut their own industry to the point of bankrupting entire sectors (solar panels, plastics, now cars). But WHY do EU governments do this? (and why are they lending very large amounts of money to do it)
I mean, they must be aware of the basic facts: if you're making weapons of war best possible outcome is they sit unused in a basement. If they are used, the other guy's weapons will be going in the other direction and your weapons will not just not contribute to the economy, they'll literally have very strong negative contributions. Whichever of the 2 situations you're in, there will be no improvement for the population, no economic growth beyond the initial investment, no nothing. Just a LOT of tax money poured into a black hole in the ground.
I get that weapons lets a lot of political appointees select their own companies and personally make a lot of money in the short term, at incredible cost to everyone around them.
For the governments buying the weapons this is a guaranteed zero-return-on-investment, with loaned money. Does it really need explaining how stupid this is from a financial viewpoint? In a few years this will become a drag on the EU economy of something like 10% of all taxes raised in the EU. And that's assuming they aren't actually used for anything, because then it'll be a 50%+ drag on the economy.
Europe now has war spending 60% as high as it's famously ultra high social spending.
Surely a DRAM industry is cheap in comparison, plus it actually has a chance of paying off.
In a few years dram will be a losing money segment again, why bother?
I doubt it will. You hv to consider memory tech may shift a bit towards LLM-first, so a kind of more computations in memory, and other innovations will be common and demand will not be lower especially if you consider average lifetime of a GPU card is 5 years, there will be demand like just to utilize all those racks to pay out all the money they borrow now. In memory compute will make AI more decentralized (still a hopium).
I think you missed the part where they(SK, Samsung and Micron) are a cartel forming a monopoly, and can control market prices regardless of demand via controlling the supply.
Sure, cartel behavior is illegal, but if you're in bed with the right political administrations, then blind eyes can be turned.
They have been a cartel for decades, the have had boom and bust periods all the same. There's only so much money you can leave on the table before you snap and stab your cartel pals.
Memory packaging is non-trivial and Europe currently doesn't have players in that space.
The primary players are South Korea, Taiwan, the US, and China and most of their capex in the space is in Asia and the US.
Goodram exists. Last sticks I bought were from them because they are not that expensive now.
Goodram doesnt do packaging, they do assembly.
as far as I recall it, packaging wasn't done in Europe even then
Global Foundries (fab) is in Germany and Amkor (packaging) is in Portugal
I almost forgot abuot GlobalFoundries and I wonder if they could profitably make DRAM at current prices...
Nope. They don't have the tech.
I think some packaging (not necessarily memory) goes on in Ireland
Perhaps an uninformed take, but there seems to be nothing about memory packaging that European industry would have much trouble with - except not having an already existing factory and specific know-how. Otherwise, it seems to be precision mechanics (for bonding) and chemistry (for the materials), both of which are bread and butter of European industry.
There is a lot of tribal knowledge and a huge ecosystem of vendors needed to support packaging.
Additionally, most countries in Asia as well as individual states in the US are giving massive subsidizes to help retain the existing ecosystem of vendors.
It is difficult for the EU to directly compete head on when the ecosystem has largely existed outside the EU for almost 30 years.
This may not appear sexy but this is a huge deal. Power semiconductors are critical for most NatSec and industrial usecases.
You can likely live without a new gaming rig. You likely cannot live without replaceable parts for trucks and cars.
>You can likely live without a new gaming rig. You likely cannot live without replaceable parts for trucks and cars.
Bad logic for an argument. With this logic you can say you can live fine with your 30 year old diesel car that doesn't need these new cutting edge SiC wafer chips, and that making diesel engines is more important for national security than making chips, and so on and so forth till you reach knowing how to make fire, swords an arrows as the most important for national security.
Truth is fabbing CPUs and GPUs is miles more profitable than what Infineon and EU are fabbing, and they're just as important for national security/sovereignty if not even more:
Did Infineon's engineers use EU or US CPUs and GPUs for the complex CAD/CAE simulations needed to develop those SiC wafers? Why else do you think the US got TSMC and Samsung to make cutting edge fabs on US soil if CPUs and GPUs are not important for national security? When you need to process TB of live battlefield data in case of war, do you not need domestic CPUs and GPUs?
Ten years old CPUs are "okay" and there are fabs in Europe that could make them. Faster is better, but slower isn't entirely useless and hugely better than no CPU. Like a 100 hp heavy truck is greatly underpowered, but almost infinitely better than no truck.
>Ten years old CPUs are "okay"
Yeah but in a war, be it a trade war or kinetic one, against a power WHO HAS modern chips and can process data and intelligence faster, who has better leverage? The one with cutting edge tech, or the one with "okay" tech?
Seeing how the orange bafoon slapped the EU around with his trade war and received no retaliation that could cause even paint in return, I assume the answer is obvious.
Another point being how Ukraine can beat Russia in drone warfare simply because they have access to the superior tech of all western allied nations despite making lower numbers than Russia which makes more but can't easily get embargo chips.
>and there are fabs in Europe that could make them.
Which EU fabs make 14++nm CPUs and GPUs? Which EU domestic CPUs and GPUs IP designs are there at the moment that can be fabbed in case of emergency without parts or IP from rivals?
The truth is EU is painfully absent from this field. Sure, in theory they can duct tape together some RISC-V CPU and some ARM MALI GPU production on Dresden 16nm fabs, in case of a war, but since the EU never built and shipped a commercial domestic CPU and GPU from scratch in the last 20+ years, it's not gonna be as easy as turning a screw, as it is for nations that ship new cutting edge compute chips every year: US, Korea, Taiwan and to an extent Japan and China. When you've been out of the game for so long, you get rusty and comebacks are slow and painful. EU would take decades to catch up and rebuild this industry in case of a war cut them off from these parts. It's basically another supply chain vulnerability it failed to address.
There's Intel 4 in Kildare in the Republic of Ireland.
Regarding the fabs (well... fab - not good), GlobalFoundries Dresden, and regarding IP, ARM. Or old AMD or Intel designs, because intellectual property goes right out of the window in a war.
>and regarding IP, ARM
Which was the last ARM CPU that's been full-stack designed and made in the EU, that they can pull out of the mothball and iterate on if need be?
Just because you own an ARM license and know how to make one in a lab or university research group, doesn't mean you know how to make and ship competitive chips as the nations that actually do this for a living daily.
> Or old AMD or Intel designs
Firstly, where will the EU get old Intel and AMD designs? Are the Verilog files, netlists and photomasks just floating around?
Secondly, if you need to bootleg the obsolete 20 year old designs of your adversary to fight them with, then you already lost the war before you even started. Which is why the EU lets itself bullied and pushed around by everyone. You have no sovereignty when you're at the mercy of everyone for the vital things you need.
ARM is a UK company, of course they have the IP. While their highest performance CPU cores have been designed in the US in recent years, they also have design teams in the UK and France. I don't doubt that their European part can adapt an otherwise finished design to a different process.
Intel or AMD IP would need to be stolen somehow. GlobalFoundries might still have some old AMD materials around. Yeah, this does seem really difficult - it would be easier and less risky to run x86 code on ARM if necessary.
Lastly, a factor of 5 or so in computing power does not decide a war.
>they also have design teams in the UK and France.
That's not enough. You need teams that have experience and history of shipping complex full stack designs from start to finish, because that's what the US (and China) has, but the EU does not. Thinking you can just turn a screw and have those teams get proficient and catch up with decades of lost experience in shipping complex silicon designs is wishful thinking.
At least the US had enough self reflection and foresight to realize they de-industrialized too much and so have on-shored a lot of manufacturing back, from cutting edge semi to ship building, using their Korean and Taiwanese partners. The EU has not and remains much more venerable, with its only strategy being wishful thinking, hopes and dreams that the status quo and favorable alliances last forever, which history proves they don't.
>Lastly, a factor of 5 or so in computing power does not decide a war.
What about 5x in everything else, not just compute?
The eu doesnt plan to go to war, so the strategy revolves around that. They just need to be thorny enough to make attacks painful and then sit behind their nuclear Shield and allies
>The eu doesnt plan to go to war, so the strategy revolves around that.
As an EUropean I'll say that's where the big problem lies with the EU's fairytale idealist pacifist strategy: that its limp-dick approach being the key to endless peace and prosperity, it only (barely) works inside Europe with pacifist liberal equal-minded nations sharing a common identity and values, but doesn't actually work in real life outside the EU in a world where you're surrounded by hyper-nationalistic predatory thugs with imperialistic tendencies who want to see you take you down so that their empire can dominate you, and it ends up affecting us anyway leading to EU being forced to make concessions to these extortionist thugs (Trump, Putin, Erdogan, Aliyev, Modi, Xi etc) who know the EU will always sacrifice the financial welfare of its own citizens in order to pay up racketeering fees to avoid any and all direct confrontations, leading to a slow but constant decline in its global influence and economic prosperity.
That's why guerilla soldiers in conflict areas around the world typically kidnap EU tourists but not American ones, because they know EU nations are so weak they'll just pay them ransom, while they're afraid the US will send some military team or a drone strike and kill them.
The only guaranteed thing in real life, is that (someone else's) war always finds you anyway, whether you want to or not (see Trump's tariffs and his conflict with Iran how much that affected EU economy and consumers). So it might as well find you well prepared for it, which the EU sadly never is, because it holds on to an outdated mindset from the peaceful period of the post-USSR collapse, that was only a once in a lifetime exception in human history, in a world slowly returning to default factory settings, which is "might makes right".
So "better to be a warrior in a garden, than a gardner in a war."
>They just need to be thorny enough to make attacks painful
How did that work against Trump's tarif war? The truth is that you're right, but the EU isn't thorny enough, and that's the problem.
>and then sit behind their nuclear Shield and allies
And what happens when your allies don't feel like being your allies anymore and instead turn into economic adversaries working to undermine your economic prosperity to further their own in a zero sum game? How does your nuclear shield help then? Is the EU gonna nuke the US over an economic trade dispute? Who do you think has more nukes and better shields?
The main reason why guerillas kidnap EU citizens is that the US have a policy of not paying ransom. Which is game-theoretically smart, but gets really bad PR the first few times, and if I knew the people sacrificed the first few times, I might have a different opinion, too.
> I might have a different opinion, too.
That also encourages everyone else to go kidnap your citizens knowing you're a free walking ransome bag waiting to be seized, thereby making the problem worse and increasing the chances that you'd be in such a position.
Trucks and cars worked fine without them for decades
Because they did not use electric motors.
Now SiC MOSFETs are the most efficient devices for the power converters used in electric vehicles, where DC voltages of 800 V or 1000 V are used in the latest models, which require, respectively, 1200 V or 1400 V SiC MOSFETs.
GaN HEMTs (high-electron-mobility transistors) are the most efficient devices for the power supplies that are fed from mains electricity, where 650 V GaN devices are used for the universal power supplies, which must accept up to 240 V AC on input.
I'm curious about RF uses (esp phased-array radar / AESA). usually GaN is treated as the the best for power/efficiency reasons, but does SiC enter into the picture? not fast enough?
The GaN itself is grown on SiC (instead of Si) for high performance RF, so SiC production capability matters if you want a homegrown supply chain (but you won't use SiC in the signal path itself). Many fabs are on 100mm or 150mm wafers trying to jump to 200mm ASAP.
That's all motherhood and apple pie, but to the GP's remark, why is this supposedly "a huge deal"?
In my mind, it's a process improvement towards cost-effective economies of scale, not a fundamental change in the end product. The PR also mentions Infineon will be "further executing on its 200-millimeter SiC manufacturing roadmap."
I mean nice milestone accomplishment, but are SiC products a major supply chain bottleneck, or can the market reasonable expect unit costs to be chopped in half any time soon?
For now, SiC MOSFETs remain more expensive than silicon IGBTs, even if the difference is today much smaller than a few years ago.
So at any new design, one must ponder whether to use the better SiC MOSFETs or the cheaper silicon IGBTs.
Even if they are more expensive, SiC MOSFETs ensure economies elsewhere, i.e. they enable a higher energy efficiency, which reduces the operational costs for the end user, and also the use of cheaper devices in other parts of the system, which can compensate the higher cost of the SiC devices.
So now any price reduction of the SiC devices is welcome, expanding the range of products for which they will be chosen, thus their market share. The price reduction from using wafers with a double area can be very substantial, not only due to producing a double number of devices with a not much greater cost, but also by improving the yields for large-area devices, because usually the chips that are located close to the edge of the wafer have worse yields.
There already are many recent electric vehicles that use SiC MOSFETs, inclusively from Tesla and this is a market that grows quickly and SiC MOSFETs are a critical component for it, by enabling much better efficiency, size and weight of the electronics, in comparison with the old IGBTs.
I am pretty certain that the new EV chargers that can charge a vehicle in less than 10 minutes, or even in only 5 minutes, by working at voltages of up to 1 kV and currents of up to 1 kA, are made with silicon carbide MOSFET modules and it would have been difficult to make them in the past, when only IGBTs were available.
...Can't we just go back to thermionic vacuum tubes at this point? Those worked just fine at kilowatt powers and megahertz frequencies.
Vacuum tubes have huge conduction losses, i.e. the voltage drop on them is very high at high electric currents.
This is why for high-power switching they were quickly replaced by gas-filled tubes, long before WWII, e.g. with thyratrons (General Electric) and ignitrons (Westinghouse Electric).
Semiconductor devices have much lower conduction losses than even gas-filled tubes.
While conduction losses are much less a problem with semiconductor switches, many of the semiconductor switches have high switching losses, which limit their use to low switching frequencies.
Silicon carbide devices have lower switching losses than the older semiconductor devices, and gallium nitride have even lower switching losses than SiC devices, so both enable great increases in switching frequencies, which has many benefits.
As a bonus, in the future (when SiC crystals with a better quality will be available) silicon carbide devices will enable the production of high-voltage switches that require much fewer devices connected in series to reach working voltages of hundreds of kilovolt, or even of a few megavolt.
Nowadays, thermionic vacuum tubes or thyratrons remain sometime competitive in pulse switching applications, but only for very high voltages, i.e. above 50 kV. Even for such applications they have a too short lifetime, and there are better semiconductor alternatives, except that this market is too niche, so the big semiconductor device manufacturers do not bother to make such devices, which sometimes forces the use of obsolete vacuum tubes or thyratrons.
Only for the generation and amplification of electromagnetic waves in the microwave frequency range and at very high power, special vacuum tubes may remain preferable to semiconductor devices (e.g. the magnetrons of microwave ovens or the traveling-wave tubes still used in many satellites).
"Because they did not use electric motors."
Wooooo let's go back to 1827 for a moment and have a chat with Anyos Jedlik. Or maybe Andres Flocken in 1881. Let's see how they did electric vehicles.
Oh, lookie there, no SiC.
I mean sure, motor control existed before solid state switching devices, but it was also much lower efficiency and used much lower voltages. Power electronics absolutely are critical for the efficiencies and handling characteristics of modern electric propulsion.
A yes, and because the electric vehicles in 1881 were so convincing, we are all driving EVs today....wait
Point still stands - no vehicles electric or otherwise were using SiC until around the 1970s (and even then that was really niche applications) and widespread use wasn't until the 2010s.
Until then, we were predominantly using SiO2 MOSFETs and IGBTs.
You will note that timeline also maps quite well on when electric vehicles started to become viable in the market . Of course battery technology also had a lot to do with that, but efficient high voltage power electronics absolutely are an important part of the equation.
They existed.
I would not say electric cars "worked fine" until pretty recently.
The 2002 Electric Ford Explorer worked perfectly well for the USPS.
Power electronics are vital for modern diesel-electric locomotives.
That's a funny way to frame things. Do you think Infineon is forced to choose between those two markets?
Yes?! Infineon's fabs and new SiC wafers aren't built for the small feature sizes of CPUs and GPUs.
Making power FETs uses different process nodes than CPUs and GPUs. All semi fabs are highly specialized: some for power ICs, some for RF, some for DRAM, some for logic, some for NAND.
I thought Wolfspeed was already at 300mm at the start of this year?
I don't think they're claiming to be first in producing the substrate, just first to process it into functional power devices.
Wolfspeed/Cree does not have a fab and does not create wafers, let alone LEDs. They are purely a research/licensing company with a side business of selling luminaire fixtures.
Huh? Wolfspeed grows SiC crystals, makes substrates, fabs SiC wafers, and packages these.
Wolfspeed just barely started making anything. Their plant only broke ground recently in 2023, and they did not have anything fabrication-wise going when they first broke off from Cree in 2021.
I know this because I was there.
Wolfspeed didn't break off from Cree, the Cree name was sold off as they moved away from LEDs and became pure SiC/compound semiconductors.
The Marcy fab has been in HVM since mid 2023. Prior to that they had a large 150mm fab in North Carolina since at least 2016. Unless you are referring specifically to 300mm SiC?
you both agree then
How does this compare to GaN? Better or worse?
Different kind of applications. GaN has no reverse recovery at all and a way smaller gate source capacity and threshold voltage so you can crank up the frequency higher.
But GaN conducts heat worse and also anything that can withstand more than 650V becomes a huge die due to no vertical mosfet tech in GaN yet.
SiC is more or less like super Silicon, almost negligible reverse recovery, very fast switching, way smaller gate charge and a 1200V 1700V is economical and very performant.
For high power like 10s-100s of kW SiC is the best. Higher than that passive components become so big you don't need as high a frequency so you would go with Si IGBTs, lower than that GaN allows you to increase frequency by a lot and reduce passive components size AND in a lot of cases total losses.
So yeah, don't expect to see GaN in a car power train and SiC in a server power supply.
GaN is (much) better than SiC for certain FOMs, but worse in terms of thermal conduction and technology maturity. In addition, you can't yet make "vertical" GaN devices (there's a few flamed out startups that tried) which forces certain undesirable tradeoffs at high blocking voltages. So GaN's sweet spot is 100-650V devices, for 48V and mains voltage applications.
SiC is for high voltage, GaN is for high switching frequency.
There is an overlap between SiC and GaN at rated voltages between 600 V and 750 V, but at lower voltages than that only GaN devices are used and at higher voltages than that only SiC devices are used.
Even at 650 V where there are available both GaN and SiC devices, there is little direct competition between them. For such voltages, SiC devices are used for controlling greater powers, where also the power dissipated by the device is great enough to require a big package that enables efficient cooling, while the pure GaN devices (i.e. not the hybrid silicon-GaN cascode transistors) are available only in small surface mounted packages, which enable much higher switching frequencies, but at lower output powers, because it is more difficult to cool them well enough.
For now, the SiC devices are only unipolar devices, i.e. MOSFETs and Schottky diodes. But in the future, when SiC crystals with fewer defects will become available, there will also be SiC bipolar devices, i.e. thyristors and IGBTs. The bipolar devices will be available for much greater powers than the current devices, while also being much faster than the silicon bipolar devices. On GaN, bipolar devices are not possible, so GaN will always be used only for higher switching frequencies, but lower output powers than SiC.
Silicon carbide MOSFETs are the best devices for the SOTA electric vehicles and also for their chargers, which work at 800 V (where 1200 V SiC MOSFETs are used) or at 1000 V (where 1400 V SiC MOSFETs are used).
Diamond would be even better along several metrics, but we can't grow 300mm monocrystal boules yet.
The summary at the top of that page is:
Infineon marks innovation milestone with world’s first 300-millimeter power silicon carbide (SiC) wafer
PBusiness & Financial Press Nov 09, 2026
Infineon presents industry-first functional power devices processed on 300-millimeter silicon carbide wafers
Technological breakthrough underlines Infineon’s innovation strength and leadership in power semiconductors and wide-bandgap technologies
300-millimeter SiC process feasibility lays the foundation for secure, future-proof SiC supply to customers in automotive, renewable energy, industrial and AI data center applications
Infineon will present the first processed 300-millimeter SiC wafer to the public at the electronica trade show starting 10 November 2026 in Munich
I thought the Sic was implying that there was a typo or mistake in the title