TSMC Will Buy High NA After All: ASML's New 6x12-Inch Mask

0:00Last week, ASML got three of its biggest customers to publicly commit to the same roadmap item within 48 hours. TSMC was on Sunday, Samsung and Intel on Monday. All three signed on to move high NA EUV from today's 6-inch photo masks to a new 6x12-inch large format mask standard. And each company attached its own high NA milestone. Intel is already in volume production today with select layers. Samsung's targeting DRAM in 2028, and TSMC is targeting high-volume manufacturing in 2030. And this matters a lot for ASML. Half-size exposure field was the biggest technical argument against high-NA adoption, as we'll talk about in this episode.

0:43And so the high-NA ramp is what ASML's 2030 financial model depends on. And this week made that ramp more likely. And this matters for the entire ecosystem because the entire mask supply chain now has a public coordinated commitment to invest against. They know that their customers are going to buy it. So that's going to be today's episode. We're going to cover what ASML announced and more importantly, why it matters for ASML and for the supply chain. Hello, everyone, and welcome to another Semi-Doped episode. I'm Austin Lyons. I publish ChipStrat, which is strategy analysis across semiconductors AI and data center infrastructure.

1:21Institutional investors read it, and so do technologists and executives from the biggest companies to startup CEOs. So check it out at chipstrat.com. Hey, I'm Vic Shekhar. I run SemiExponent, a research and consulting firm focused on semiconductors and AI infrastructure. I also write Vic's newsletter on Substack, where I cover everything from optical interconnects, networking, memory, packaging, power semiconductors for a whole range of investors of various kinds, and engineers. you can find it at vicksnewsletter.com. All right, Vic. So, hey, what's going on over there, man? I feel like I haven't seen you in a while.

1:58Yeah, it's been a while, actually. I think we needed a little summer break off. We had a week off at the end of August and we're like, okay, let's just recover and regroup in September because, you know, stuff's always going on in this world. And if you don't take a break, it's not going to work out. So we had a little week off. But in that week, of course, we weren't quiet or anything. We were pretty much writing our sub-sacks and you know spending some time with the family but what ended up happening was like this week there have been at least like two or three new products that are all like personal ai agents that i've been messing around with a lot the first one i i messed around with was like grokbot and i was posting on x that i didn't really like it because it sucked a whole lot of tokens but then you know the whole idea was that it has this little like cpu uh and a little like vm machine computer right and then it goes and like operates your computer screen for you and you know handles your logins we we made a whole episode on grokbot and its implications on like cpus for ai but then later like you know several other things have come up of course the most notable one being the Astra 6 from OpenAI, ChatGPT 6 Astra, which is actually kind of amazing.

3:15So I signed, I've been a cloud user for the longest time. So I was like, okay, here you go. You know, OpenAI, take my hundred bucks. And I was like, let's try this thing. And I wanted to try the computer use thing because it's like pretty fascinating. You can like, it handles your browser and all that. It's awesome. So I just put in like, hey, I need to book travel tickets to, you know, OCP Global Summit in October. in the Bay Area where I'm going. And I was like, find me tickets. And so on the way back, I want to go to Japan where I'm planning to have my family over. And then, you know, I want to have a week there and then I want to come back, you know, and all that.

3:52So it planned all this stuff for me. And it was like, oh, here's your ticket fare. And then it's like, hey, do you want to fly instead of Tokyo? Do you want to try Osaka instead? So that your like travel itinerary could be from Tokyo to Osaka. You could do your vacation. Fly out of Osaka, right? I'm like, yeah, yeah, yeah, look up that ticket. So it went and like, I could see the house moving, right? Like, so it's looking at the thing, clicking buttons and typing in the airport codes. And I'm like, this is cool. That's awesome. And, you know, I like the magic of, hey, you told me to do this, but have you thought about that?

4:22I think that's where it's killer. It went one step further, too. It's like, they tried the Osaka thing and it's like, no, it's too expensive. Like, your original flight was better in and out of Tokyo, but you'll have to backtrack your holiday back. Okay. then it's like hey do you want me to try like booking only to Tokyo and I'll book a separate flight ticket from Osaka to you know wherever you have to be in and I was like oh cool let's yeah that's a good idea do that you know so it tried out all these scenarios for me it was awesome so I'm gonna play more with it you know what's a hundred bucks if it can help me plan a better ticket like I'll recover my hundred dollar subscription in like no time open it right totally to be sponsoring us like this is a pitch it's such a great pitch yes yes yes open air if you're listening um i use both claude and codex so my son who's been working on his own indie game all summer and he's working on getting it in the app store um and it's really great he has codex do all of it for him which is awesome because then he can focus on the storytelling and the story itself and it does all the programming um and then of course i use claude and do a lot of coding stuff.

5:26But I love your use case of like, try all these permutations of travel to save money, because I tried to do that. I know there might be better things out there. But it takes so freaking long, then I'm just like, screw it, whatever. I'll just book whatever, you know, it's like, I just get fed up. And one more thing I haven't even gotten down to is like telling it like, hey, I have points on this airline. I want you to select this or that, you know, prioritize this or that. So it can be like, oh, wait, if you select this, you will get points on this airline on this so much is possible and it's all like AI based, it's amazing actually so I really find a use for this stuff and I did another crazy thing which I shouldn't admit on camera but here goes, right?

6:06You YOLO AI all the time so there is this other tool called Instinct which I sent you it is like this little tool that has no information so if you go to instinct.co you see a webpage that's like a letter written to somebody like, hey, you know, we're building this. No information about the company or what it is. Literally no pricing even. And it's only an invite-based thing. And so I happened to chance upon an invite which I sent to you. So thank you to whoever's listening to this who gave me that invite. And so what I did was like, it's so dangerous. Do not try this at home, okay? Do not try this at home.

6:46So what I did was like, I gave it access to my stuff and email and all that. Because I'm like, I'm curious. like people are saying this is amazing this tool is amazing so if it's not really useful to me I want to know but for it to be useful to me I need to give it some access so I looked up beforehand can I revoke access if I give it and I looked yeah yes you can revoke access I'm like okay fine I'll revoke it if it's bad right so what I did is I gave this tool all this like you know details to my email and all that and it actually keeps a track and the way you use it is like you use it with your phone so you can text it on iMessage or like WhatsApp.

7:25And so it sent me a message saying, hey, you just got an email, by the way. You've got your press pass for the OCP Global Summit. It's been approved. Here's the code you can use to get access. And I'm like, oh, cool. This is cool. I should do it sometime. But then I was like, I was working on something else. And then I was like, hey, you're supposed to be a personal assistant, right? I'm like, okay, go ahead and sign me up with the code. See ya. I'm like, do it for me. You take care of it. And then it was like, it went off and I went back to doing my work. After some time, it sent me another message.

7:57And it's like, here you go. You know, I registered you. I put your name down. This is your affiliation I put down. You know, I put your meal preference as, you know, no meal preferences. And by the way, I ordered you a medium-sized t-shirt. How did you get my size right? Or did you just guess it? Or did you know I'm like Indian, average Indian size? They're like, how did you guess a medium shirt? by the way, I ordered you a shirt. That's amazing. That's pretty good. That's pretty good. So did you revoke access? Yes, I did. I saw my email come through with the QR code too. And then I'm like, oh, here's the QR code.

8:34I just click star on my Google Drive. I'm like, oh, I'm good to go now. And then I'm like, holy shit, this is amazing. But it's also too amazing for me right now. So I'm like, okay, that's it. I revoked all access to it. I deleted it. They allow you to delete your particulars. particulars, there's a way you can delete. On both sides, you can revoke access on the Google side and you can delete any hashed information that is stored on Instinct. They allow you to do that. So I cleaned everything up, but it's pretty good. And then I find out that Muse, you know, Meta's Muse Spark is like another personal AI agent that's come up, you know.

9:10So you can, I installed that, but I haven't tried it out yet. But I'm trying all these personal agents and I think it can be really useful, to be honest. Yeah, totally. Okay, lots of thoughts. So when you sent me Instinct, you sent me just like a very short text. It was late at night here. And it was just like, hey, check this out. And it's a link to a website. I click on it. It's like instinct.co. And then it comes up. It's white background with black text, just raw HTML. Looks like the very first time I made a website, it looked just like that, you know? No CSS or anything. I scroll around. All it says at the bottom is like San Francisco, California.

9:43It doesn't say anything about who it is, whatever. And I'm like, oh, Vic got hacked. Vic totally got it. This is a.co website. He said something very generic, like just like, check this out. It's got nothing here other than like sign up, you know? And I'm like, this could be someone in North Korea. This could be someone random, you know, who knows? And so then I like, well, I'm just going to click and go a little further, you know? So I click to sign up. And then it's like, you know, will you give us access to your like Google Drive? And it was like to write stuff, to read stuff and to delete stuff.

10:13And that's where I stopped. I was like, uh, nevermind. I'm just going to let Vic do this. And if, and if it's not a scam, it's one of us. Yeah, exactly. So thank you for YOLOing it. And, and, you know, maybe now I'll try it. And I think, you know, zooming out and I'm very excited about Muse. I've heard very good things. My, my son this morning, we saw it in the wall street journal and he was like, dad, you got to try it. And I was like, I know, I know. I just, it's, I can't keep up, dude. This stuff's happening every day, but I want to try it. And I think what this is, is this is the evolution of open claw, which was like the very early adopter tinkerer hacker.

10:45Like I want agents, I want them to do useful things. I want them to have memory, but you have to like do a lot of work. And now this is the abstraction to like more of a product experience where more people can just go in and they can just use it and they don't have to think about all that stuff under the hood. So obviously like growing up from, you know, homebrewed computers to like, this is a Dell computer that you order and it just shows up for you. So I think obviously, as we've said before, this does show, even if it's not GrokBot, even if it's not Instinct, Muse, I mean, think about the reach of Muse.

11:17Like Meta basically has Instagram. They've got Meta. They've got the whole family of apps. And so - WhatsApp is theirs too, by the way. WhatsApp. So literally, we could talk about hundreds of millions of people trying this, using hundreds of millions of cores on server CPUs that those people weren't using yesterday. and now they'll be using in the next handful of months. So obviously there's huge implications for server CPUs, for memory, and so on. I think just this product experience of agents is worth tracking. It's genuinely useful. If you tell my mom, hey, you just texted stuff, it'll do its job.

11:57That is the lowest barrier to entry for agentic AI. Literally everybody can text. My mom texts me. You know, like if he can do that, everybody can use AI if it's through a text message. Everybody's on WhatsApp too, you know, like it's amazing. I think it's come to a level of accessibility and usefulness. Actually, GrokBot was not, in my opinion, as amazing as what I saw ChatGPT6 do with its computer control. It was actually pretty amazing. I think we are only starting the era of a new era of like really accessible AI. So far, people didn't really know what to ask in a text box. I tell my family, why don't you use Claude to do this?

12:39They're like, what do I do? Oh, go to chatgpt.com or whatever. And then what? Oh, there's a text box. Okay, now what do I ask it? Yes, right. So many problems. Now it's getting to a point of usability that is very much for the common person, not the geek. Totally. I just thought of a use case for my wife to try it, which is we're trying to plan a family vacation with like my in-laws and with, you know, siblings. And so there's gonna be several families. And, you know, she, she like raised her hand and she's like, oh, this is gonna be for my, you know, parents anniversary. So let me schedule it.

13:12This will be fun. And then they won't have to worry about it. And once you start getting into that, you realize like, oh, everyone has calendars. Everyone has preferences. Every, every time you throw out, like, what about this location or this location? Someone was like, oh, how about that location? And it's something different. and it's like a never ending thing and you're never gonna make anyone happy. What if you could just have Muse or Grokbot or Instincts or someone do it? They can manage the calendars. They can throw out the ideas. And plus, obviously it's nothing personal. Like, you know, I think it's super annoying when someone comes back to like, okay, I see that you put time in here, but here's 10 other ideas, you know?

13:44But the infinitely patient AI can do it. And these are like real problems that I think real people would latch onto and be like, yes, please help me here. I will entrust some of my information to you. and even pay for it so that I don't have to deal with it. Yeah. And you know what we need to make all of this happen? Chips. And you know who makes them? TSMC. So that's a segue to the actual part of our episode today. I love it. I love it. And who does TSMC rely on? ASML. So let's talk ASML. They had news this week. News with TSMC, news with Samsung, and news with Intel. So we're going to talk all about it in the next 30 minutes or so.

14:27So first, you know, I'm going to zoom back out. And I think everyone knows ASML by now. But for anyone who's brand new to this space, obviously, ASML is the lithography company. And lithography is that step where you project the chip pattern onto the wafer, which allows you to etch or deposit or ultimately create the physical implementation of these transistors. and why asml matters is they're the only company on earth that sells euv lithography machines extreme ultraviolet we have a whole podcast that we've talked about before we'll link it to the show notes about lithography you should check it out um there are some competitors in the space nikon and canon still sell older duv tools but at the leading edge it's asml or nothing um and so So what does that business look like right now for ASML?

15:17Just in Q2, 2026, they reported in July, ASML's total net sales were 9 billion euros, roughly 10 billion or so dollars. Gross margin of 54%, which is pretty good when at the end of the day, they're making a manufactured heavy equipment machine, right? So to have gross margins of 54%, net income of around 3 billion euros, so 31 % net margin on the quarter. This is pretty crazy. Again, I think, you know, in my head, I go, oh, well, NVIDIA makes, you know, 70 % margins in these memory makers have 80 % margins. So that's like a lot less. But when you actually think about ASML as making equipment for manufacturing, like these margins are crazy, which of course they have earned because it's probably the most complex machinery in the world.

16:08So the question is, before we get into the news, one of the things you might have noticed in the news that I talked about a memory company, Samsung, and Logic companies, TSMC and Intel Foundry. And so one thing that's worth noting is actually ASML makes their money by selling systems and by servicing these systems. And actually, the servicing of the systems was 2.8 billion euros of that 9.3 billion euros. So almost 30 % of their revenue is actually recurring revenue. It's like subscription revenue. There's kind of an analogy here. Like if you think about Apple, obviously they sell devices, but then they also want to sell, you know, software and services and get some recurring revenue.

16:51So they're not just having this lumpy, like we sell devices during the holiday season kind of thing. ASML is trying to build a similar business profile where of course they sell these$400 million dollar machines, but then they also have this ongoing revenue of helping to maintain those machines. But of those system sales, actually 51 % go to logic companies and 49 % are for memory companies, at least back in this quarter. That actually was pretty surprising to me. I kind of just, you tend to think more about the bleeding edge EUV for logic and not as much for memory. But was that news to you? Yeah, to me it is because I'm always a little behind on catching up on the lithography and like wafer fab equipment space.

17:41But what I wanted to mention is when for people who are listening to this, when you say logic and memory, fundamentally, they are different process technologies, right? So when you say two nanometer transistor or three nanometer transistor by now we've explained before that this is by this point uh a way of saying it the dimensions aren't actually two and three nanometer it's like a gate all around where they stack multiple gates it's a different structure but the nomenclature has stuck right so that is like one side that is what you refer to as logic right and memory technologies are actually quite fundamentally different because the transistors there even the leading transistors there are more a 10 nanometer class node.

18:29So when you say exactly like EUV is required for memory technologies, it's not to make the transistor, is it? Right, right. It's really about memory is all about getting as many bits as densely packed as possible. So you can get as high of capacity as possible and really try to drive down the cost per bit. And part of that is making capacitors and making the transistor and capacitor structures as close as you can. And so a lot of it is about the sort of pitch, the lowest pitch that you can get to try to really kind of cram everything in. Right. So the memory nodes have basically this capacitor and a transistor.

19:17So what they call this is like a one transistor, one capacitor, or one T, one C. technology where you know the capacitor holds the charge if it is holding charge it's like a one and if it's not holding charge this is zero and you know transistor then turns on or off to charge it or recharge it or discharge it right so that's that's how memory works so creating this 1t 1c structure is where euv lithography plays a role in memory and it's amazing that it's like nearly half the utility is nearly half so euv shouldn't always you know be thought of as getting to the next greatest transistor node because wherever you read it in the news you will hear it as hey this is for a11 or a7 nodes which is like the angstrom era nodes which is basically when you say a7 it's basically a 0.7 nanometer transistor equivalent right so that is what you usually see in the news as to that's why we need EUV, but you actually need EUV to make memory as well.

20:22So that's the only thing I wanted to like clear up just to make sure that people understand the difference between these two regions. Yeah, yeah, that's helpful. And so to that point, when you look at the geographies that ASML sold to in this most recent quarter, 43 % of their system sales went to South Korea, 30 % to Taiwan, 14 % to China, and 9 % to US. So again, you can think, oh, South Korea, memory companies, Taiwan, obviously logic. China is an interesting one because at first I was like, what? They can still sell to China? But of course, we're talking about older DUV equipment that you can still sell to China.

21:03For people who haven't followed this closely, the ASML story was, if you go look at how much they sold to China, there was a period within the last few years where it really spiked almost up to like 50 % of their sales. And then now it has come back down. And that was obviously China seeing the export controls coming. And so buying up as much equipment as they could before the export controls came. And then United States, you know, obviously there are also obviously Intel Foundry is buying leading edge stuff, but there's lots of other companies that are buying older tools. You know, you can think of the global foundries of the world.

21:38So, you know, United States is small, but it's not, and that's not just Intel Foundry. I wonder if it's Micron for memory. Oh, yes. Yeah, true. True. Definitely. Okay. So let's move forward. So why was there so much news this week? Well, this week, all of these announcements were clearly timed and coordinated around the conference. SPIE Photo Mask Technology and EUV Lithography Conference, which was in Monterey, California. Nice place. I was like, dude, we should go next year. Monterey is cool and we could learn a lot here. And so SPIE is the Optics and Photonics Professional Society. And this particular event is the annual gathering of the photo mask world.

22:23So this is like the people who make and use masks, whether it's captive mask shops inside of Intel, TSMC, Samsung, merchant shops, names like Photronics, DNP, Toppin, which, you know, I don't know a ton about these companies. And so it'll be fun to dive into them more. You've got mask blank suppliers like Hoya, AGC, mask writer companies, inspection folks. Because not only do you have to make the mask, you have to pattern the mask, like draw your particular layouts. But then you have to inspect it to make sure that it is correct. And so you've got companies, Lasertech, KLA, Zeiss, all in this space.

23:04and so they all come to this conference and they are all in the supply chain of making the photo mask so I guess before we talk about what changed should we would you like to introduce the idea of like what is a photo mask what are some of the dimensions kind of like what's the current state of affairs yeah I'll tell you what I know and you correct me if I'm wrong because sure this is how I learn on the podcast live this is good so photo masks are essentially patterns that are on a different um you know substrate like thing other than the wafer itself where there are like patterns where the transistors should be or should not be so you can have like a positive photo mask or a negative photo mask which means that the the patterns on the photo mask can selectively block light from hitting the wafer underneath or they can selectively let light through so however you want to look at it a positive or negative and then depending on the materials used you know parts that are exposed or not exposed remain and the rest can be like etched away right so this is how basically features on the wafer are formed and fundamentally wafers have been 300 millimeters on CMOS for a long time.

24:27Those are 12-inch wafers. But the masks that pattern on these 12-inch wafers have never been 12 inches. They've been 6 inches. I think they've been like 6x6 squares or something like that. Then what happens is now, if you have to pattern a 12-inch wafer, you have to use the 6x6 in like different spots on the bigger diameter wafer and you have to kind of stitch them up together eventually. that takes time and it's like complicated to to do i think the benefit that you're going to tell me next and we'll hear more about this is that now we have gone from a six inch lithography mask to a 12 inch lithography mask so we can pat on the whole thing in like one go yes yes so high level you're right and we'll get into the details the mask is all about how you end up drawing where the transistors go.

25:23And the standard mask has always been a six-inch square, 152 millimeters by 152 millimeters. And it is basically like a quartz plate. And ultimately, so the question is, you know, how do you draw these tiny lines to allow the EUV light to either shine through or get blocked? And that's traditionally done with a tool called an electron beam, like an E-beam writer. And it's a very slow process to draw this. But ultimately, you write on the mask. And then, like Vic said, this ends up getting used in the semiconductor manufacturing process. Today, it's six by six inches, and that works fine for low NA EUV, which is what the industry is using today.

26:13Let me see if I have some notes on low NA EUV. So what happens is you have the six by six inch mask, but with low NA EUV, there is this sort of demagnification step where it gets shrunk down by 4X in both directions. So the final actual reticle size that you're patterning is 26 millimeters by 33 millimeters. And this gets you that 858 millimeters squared for the full reticle size, the kind of like postage stamp. And if you remember in the hopper era, it was like, oh yeah, every hopper is reticle sized. We cap it to that. And the reason is because you want to just make the mask, one mask, and then pattern it all over the wafer and get as many like hoppers as you could.

27:02And then if you remember in the Grace Blackwell era, they actually said, well, we need even more transistors and more compute. So we can't make that reticle size any bigger because there's this whole industry around the photo masks. So let's just take two. Let's create the Blackwell GPU, but we'll end up stitching two of them together to kind of create a GPU with two dyes. so that's the way the industry has worked so far now I want to just step back for one minute because I think there's an analogy here that will help people understand this if that was a little difficult to follow the reason I'm saying this is that lithography requires a certain amount of expertise to see how it actually works but let me see if I can throw a little light on it so when you're saying low NA I think we should just mention and NA stands here for numerical aperture.

28:00The larger the aperture, the better. The higher the numerical aperture, the better. Better in the sense that you can make smaller transistor dimensions. That's the one thing. So we are now going from low NA to high NA, which means we can make smaller devices and patterns. The other thing is this reduction in size that you mentioned, like 4x. right? The best way to think about this is think about a telescope like the ones that you, you know, you can like look, like the mariners use at sea, right? Like imagine pirates of the Caribbean, like looking out through the telescope. Now imagine that you flip it upside down, like you look through the wrong end, and now everything you see is smaller, right?

28:45So that's the same idea here. So you make a mask that's six inch by six inch, but then you're like wait I drew these large patterns on a six inch by six inch how can I pattern a transistor on this so the whole idea was this is actually a fun story that we wrote on this TIL section on our daily semi-dope newsletter as to how this whole thing came about so if you like the history behind semiconductors we write a lot of it on the daily semi-dope especially on like Saturdays so you should check that out a lot of good stories for your Saturday morning. But, you know, when you flip the microscope around, everything looks reduced.

29:21So you can use a bigger mask to imprint something like much smaller on a wafer because the light reduction is very important. This is very important feature. This is called projection optics. And projection optics only allows us to make patterns that are larger than the transistor and then reduce them to make like really small transistors. Otherwise, we can't make them. If you have to make the mask as small as the transistors themselves, how are we going to make that, right? Exactly, exactly. Yeah, so projection optics. That's what projection optics solves, exactly. Make it big and shrink it down using light and light tricks.

29:57Yes, so that's why we make these masks. Thing is that it still requires several steps to do this because it's not as big as the wafer, right? And so you can reduce it to make a Hopper era GPU that is like one reticle size, but then people were like, that's not enough. I want my GPUs to do more, But, you know, there are only two ways you can get more compute on a chip. Either you make the transistor smaller, you make the chip bigger. So they're like, okay, we can't do reducing the transistors yet. So we just stitched together two GPUs to act as one, like the Blackwell era GPU. And now you've got a bigger 2X reticle GPU that they kind of stitched together by shooting two of these pictures from the mask together.

30:36Exactly. And by the way, as we talked about advanced packaging, that's how you stitch together these GPUs. GPUs, so obviously advanced packaging, co-auth, e-mib, very important. Okay, so back to when you talked about low NA and how we want to move to high NA. So the numerical aperture. So the question is like, how do you make smaller transistors? And the simplest way is to decrease the wavelength of the light you're using, shrink the wavelength of the light you're using. but EUV lithography has 13.5 nanometer light and that's very specific as we've talked about before it has to do with tin and these droplets and you're hitting them twice with the laser to essentially pulverize them and they give off light at this certain 13.5 wavelength it's not so simple to just be like oh well find a different material that gives off 9 nanometer wavelength light like that's not so easy so the other dial that you can turn is okay we can't change the wavelength of the light but we can change the numerical aperture.

31:39And so if you have a bigger numerical aperture, you can have finer resolution and so you can ultimately resolve smaller transistors. But there's a big trade-off. When you go to a higher resolution, in this case going from low NA, which is 0.33 NA, to high NA, 0.55, what happens is you have to use something called anamorphic optics. um where this is kind of like uh like wide you could think of like wide screen film and it's like a trick that they use to make that work um but ultimately um the a wider aperture means that the light hits the mask over a wider range of angles and the euv master they're actually mirrors It's like this 40-liter, like, molybdium and something else reflective mirror.

32:33And so that multilayer mirror only reflects well within a narrow cone of angles. And so at 0.5 NA with the 4X projection optics, the steepest rays coming in would fall outside of that cone. And so it didn't reflect as well. So what ASML does is they make the demagnification step anamorphic, where ultimately what this means is they keep 4X in one direction, but they actually go to 8X with their projection optics in the other projection. And so this works, but ultimately what happens is the area that it projects on ends up being half as big. Instead of your final sort of reticle size projection, your exposure being 26 millimeters by 33 millimeters, you ultimately get 26 millimeters by 16 and a half millimeters.

33:25So that means with one exposure, you get half of the area ultimately. Okay, okay. I get it. So I'm learning this anamorphic optics, right? Okay, the term aside, I think the basic idea is that instead of the 4x reduction on x and y, you have 4x on one side and 8x on the other side, which means that you have now... Honey, I shrunk the kids. I don't know if you know that movie for a long time. I'm old enough. You're old enough to know that movie. Yeah, so you shrunk the kids too much on one direction, which means your transistors, your feature sizes get like much smaller along the 8x direction compared to the 4x direction.

34:10That automatically means that your size of the reticle has dropped by half in that direction where it was 8x. Where you shrunk more, the dimension drops by half, right? So that's what this whole thing is. Remind me what it has to do with the mirrors again. Um, oh, just the, the way that as you, uh, increase the numerical aperture, um, as the light comes in at like a steeper angle, it doesn't travel through the mirror sandwich as well. Okay. I see. So the numerical aperture being bigger means that the mirror can collect light from a wider angle. And therefore, uh, a higher numerical aperture in the mirrors means that this wide angle can now come in where it could not come in in the low na right right but for some reason with the four by four demagnification it that those that light coming in at the steep angles it like doesn't make it through all the way like i'm in my head i'm thinking of it like it's like fuzzy when it comes out yeah okay got it got it so the the mirror itself allows for more of the light collection okay got it so that's what a high na does and now we've gotten more magnification or or rather more reduction on one angle, and then the reticle size drops.

35:31Okay, now how do we get the reticle size back up? Yes, so you could just accept high NA as it is and use six by six inch masks, but your throughput is cut in half. If you used to make a Hopper GPU in one shot, now you're going to make half of a Hopper GPU in that shot. And then you're going to have to make another mask that's the other half of the Hopper GPU and have a second shot. So now the economics of the throughput of all these very expensive machines that you're buying, you can either cut the number of wafers that you're producing down or you can buy two$400 million machines, right? So now it's like, oh, it was$250 million for low NA EUV and now I gotta pay 400 million for high NA EUV because it's complicated.

36:18It's got even bigger mirrors and stuff. And you're telling me that my output goes by half. So actually I need to buy two of these. So I got to spend$800 million just to keep the same throughput. But it's not only that. It's not an economic problem. There's actually like a floor plan problem, which is you have to lay out half of your GPU on one mask, half of the GPU on the other, and then you have to like align it and stitch it together. And no like chip designer wants a seam running right down the middle of their GPU, right? Yeah. So this is a reason why some people are like, dude, high NA is never going to work.

36:53Who's going to want to do that? It seems more expensive. It makes it harder for the chip designer. Now, Intel is already using high NA in production because they found certain layers where they aren't as worried about this problem. Maybe the features that they were going to build were on a sub-reticle size die. Maybe it's okay if it's, like, smaller, you know? But other people are saying like, well, wait a minute, we need to recover that full reticle size. How do you recover the full reticle size? Ultimately, you go from a 6x6 mask, which was projected down 4x, 4x, to a 6x12 mask, which is going to be projected down 4x and then 8x.

37:36And you recover that reticle size, the full field of 26 millimeters by 33 millimeters with no stitching. So the masks did not go from 6x6 to 12x12. They only went in one dimension to 12. Exactly. Because only one of those dimensions on the high NA was changing in the projection optics. I learned something today. So that's the whole point because you have 8x reduction in one direction. But to make that equal so that you could make the reticle, a whole reticle shot in one go and not half the throughput, you can you now have to make a 12 inch mask on that dimension where the reduction was 8x the other dimension can remain same as the low na uh era of yes photography okay yes awesome oh by the way i wanted to say intel is already using this on their core ultra series 3 panther lake processors that is in production so they already have adopted this and it's kind of cool Like if you end up with one of those in your devices or something.

38:44But yeah, it's cool to think that like high NA is already in products and stuff. Yeah, yeah, it is wild. There's literally CPUs out there running right now as we speak that used high NA EUV. It's pretty cool. So, okay, now let's say everyone says, okay, okay, okay, six by 12. Doesn't seem so hard. Just make a six by 12 inch mask. Let's do this, you know. Okay, not so simple. This is where you have to get the whole supply chain coordinated around this. And this is why you really ultimately need the ASML's biggest customers to stand up and say, we're going to buy this, right? Because if ASML just says, okay, hey, I buy my six by six inch, like very high quality mask substrate from someone.

39:31Oh, hello, dear sir. Would you mind making me some six by 12 inch ones? And they're like, whoa, that's twice as wide. None of my tooling is built for 6 by 12 inch. And maybe that increases like, maybe there's like a yield concern because now the area is getting a lot bigger. And so they have to make this pure wafer, you know, in a lot bigger area. And so it's the mask substrate. It's the companies that are testing it now instead of building equipment that tests the 6 by 6 inch mask, they have to test the 6 by 12 inch mask. Anything that's automated that was used to carrying around 6x6 inches has to change to 6x12 inches.

40:10And so this is why it takes the whole industry and why those people in the supply chain ultimately need to see the end customers, TSMC, Samsung, Intel, say, yes, I will buy this. And they also need the customers to give them the timelines. You know, I want to move to this by, you know, 2030, 2031, 2020, whatever, so that they can all plan, invest in the R &D to make it happen and then plan accordingly. So now that you mentioned timeline, I think anybody listening to this would go to have that question immediately. So when is all this really going to happen? High NAE? I know some of it is happening now, but what's the plan for TSMC going forward?

40:51Because they weren't very big fans of high NAE as far as I could understand. And they were like, hey, let's just use low NA and push the limits of that as far as possible with other techniques like multi-patterning, which we won't get into here. We did speak about it earlier, though. But they were never going for high NA EUV. But now that they seem to have jumped on board, I'm assuming because they don't see any other way out. But then what's the timelines? Yeah, well, so TSMC says, hey, we are going, they announced that they're aiming for high NA high volume manufacturing by 2030. And by the way, the sort of the industry is together saying like, yes, let's target the 12 inch masks in 2031, pilot lines in 2031 and 12 inch high NA systems in full production in 2033.

41:44So the way I read that is the industry says, okay, it's going to take us five to seven years for everyone in the supply chain to make this six by 12 inch thing happen. And TSMC said, okay, great. I'm going to wait until you're 80 % of the way there before adopting high NA. And so, you know, what TSMC is going to do is they will adopt high NA just like Intel did in certain situations where the six by only having six by six inch masks is not a problem. and that will let them start to run it in pilot, run it in production, learn how the tool works, make sure the tool works. And then basically they'll have done all their learnings just in time for the 12 inch mask.

42:28So the way I read it, CSMC is waiting as long as possible. Also, it's because it does take that long to bring this monumental change. It's a monumental change when you're changing the mask size like that, I imagine. And it's like you say, the supply chain has to adapt to it over time. And not only that, all of the capacity needs to go into using those process nodes. Like, you know, it's probably going to be a future process node that uses all of this high-end features. And also the question of, you know, how many layers will you use EUV for or high-end EUV for? Are you going to use it for all the steps in a transistor or in the wafer?

43:09No, right? Probably not, right? like metal patterning and things like that i don't think you need uh euv for maybe there are only a fixed number of layers on which you lose euv for and initially those will start to be a low number of layers and then slowly they may increase it yes yes and so you make two interesting points there one which is like you could selectively use high na in places where you had to multi-pattern before um and which could be where intel is using it today hey if i had to multi-pattern with low na but I could do it in a single exposure with high NA, then even if my throughput is worse or I choose selectively in places where it's not worse, then it actually could be economical for me because it's ultimately not only less patterning steps, but then less deposition, less etch, so on and so forth.

44:00And then two, yes, you make the interesting point, which is like, ultimately, this is trying to intercept future process nodes. So although Intel is using it on 18A today, ultimately it would be about making 14a as high na ready as possible or even beyond right or exactly 10a or whatever exactly because we're talking you know five six seven years totally but this is good so i think the point is that although uh the news is very exciting to people like us uh about how this is a change in the way lithography is working and it's a big step for TSMC, Intel. I believe Samsung is also on high NA EUV.

44:45So they are also on this train. It's a big step forward, but the adoption is going to be over a long period of time. Although I think that memory, adoption in memory will be sooner than the adoption in logic. Say more. Why is that? I think that it's logic. Again, this is a little bit out of my depth for say, but I'm going to take a shot at it. I think memory structures are like a little bit like more simpler than logic transistors are. Like the transistor, one T, one C cell that we mentioned requires the transition, like this high aspect ratio capacitor and you want to pack as many of them as possible together.

45:28So the EUV has that problem to solve. But if you're going to use EUV to make a, you know, gate all around FET that has like four layers and each gate is like atoms thick and then you want to stack them. That's a much harder lithography problem, fundamentally because of the structure of the transistor is much more complex. So I think it'll hit DRAM faster. Yeah, that makes sense. The transistor is very complex in three dimensions. and these capacitors that they're trying to just cram together as closely as possible. It's ultimately like, like you said, just a big, long, tall, like rod or something.

46:07I think of like, just like digging a big, long well or something. Yeah, it's well. It is structurally very, like simpler. I shouldn't say very simple. To my friends at, you know, Micron and Samsung, whatever, where they're like, dude, this is not simple. Yeah, and if I'm wrong about the whole, you know, memory will use EUV before logic thing, leave it in the comments below because this is like we continuously learn through a lot of the people who listen to us because expertise is very vast in this industry and we only can cover this much so please feel free to let us know yeah totally totally we try to go very broad and we go as deep as we can but we can't go impossibly broad and impossibly deep right so that's where everyone else comes in now um but we are dick and i are trying to learn we're like the fastest learners right um so okay now to to maybe um make your point samsung's announcement was is that they're going to use high na euv in high volume dram production in 2028 so two years earlier than what was announced by tsmc and then intel came in and said we're already doing high volume uh high na today like we talked about in panther lake already and one of the new pieces of news that they announced was that they've processed 1 million plus wafers to date.

47:24Now, maybe logic does come first. Well, yeah. So our dear friend Lithos Graphene on X, really good. Check him out. Follow him. This is all like right in his wheelhouse. He pointed out like, okay, they probably haven't used a million wafers to make a million production parts, but sort of like they've burned through a million wafers in the practice of learning, testing qualifying production that kind of thing the point is still valid which is like intel is a million wafers deep in their learning process and if if uh tsmc is saying that they're going to adopt this later you know you can make the argument that intel is much further along in understanding and using high na in production so if you think that that's ultimately going to be a strategic benefit for them, then it's just worth noting that both companies have been very clear that this is the direction they're both going, and Intel is saying we're a lot further along.

48:24Actually, it's funny how history, in history, if you look back, the tables have kind of turned because Intel was kind of adamant against adopting DUV, deep ultraviolet, in the 10 nanometer era, which is what set them back a whole lot for the many, many years to come. but I think this time they're not making that mistake again. Yes, yes. Until late to EUV, it burned them, and now they're like... Oh, EUV, sorry. Yes, we're never going to let that happen again. So, okay, there was more news from ASML. We'll have to talk more about them. Actually, a fun fact, I'm trying to learn Dutch on Duolingo.

49:07Why? You used to listen to ASML earnings calls in Dutch? No, no, no. It's brain exercise. and I was like, what language should I learn? And then I was just, my wife's learning Danish. And so I was like, well, you already know Danish. So if we go to Denmark, go to Copenhagen, you can be the one to translate. And then I was like, why don't I learn Dutch? There's a lot of semiconductor companies there. And so quiz me sometime on my Netherlands. I don't believe you. I don't believe you. You're just doing it to get like ASML alpha at this point. Yeah, we'll talk in Dutch and you can tell me your secrets.

49:43and I won't repeat it in English. Yes. Okay. So yes, maybe someday we'll do a semi-doped tour of the Netherlands. But with that, folks, thank you for listening. Hey, Vic and I have newsletters. You should check them out, chipstrat.com, vicsnewsletters.com. I was actually talking to a CEO this week who was referencing Vic's recent posts. So that was pretty sweet. Pretty sweet. Check out semidoped.com. We've got a daily newsletter and leave comments. Send us your ideas and thanks for listening.