MicroVCSELs vs MicroLEDs, a MicroExplainer

0:00 Credo bought Hyperloom, a micro LED interconnect startup. Then at OCP this week, Credo's first active light cable turned out to use microvixels. Why buy one technology and fix the other for your first product? The bigger question is how we connect AI systems. Do we keep pushing each lane of light faster or spread the data across hundreds of slow ones?

0:21 And if we spread it out, LEDs are vixers. Austin and I are going to explain both and who's building them. Let's get into it.

0:35 Hello listeners and welcome to another episode of Semidoke. I'm Austin Lyons of Chipstrack. I write a newsletter for institutional investors and tech executives, and I also consult with semiconductor companies.

0:46 Hey, I'm Vic Shaker, and I write Vic's newsletter on Substack, and I'm the founder of Semi-Exponent, a boutic research firm on semis. How's it going, Austin?

0:55 Hey man, it's going well. Just got back from OCP.

0:59 Where are you at? I am still in San Jose in a friend's garage. I actually fly out to Japan today, uh, purely out of holiday reasons. This is not for semiconductors. I'm not hunting Japanese company Alpha. I'm not doing any of that stuff. We have great people already working on that. I'm literally going to go check out Japan for a week with my family on the way back because it's kind of on my way back to India.

1:24 And uh so we're like recording this before I hop on the flight. But yeah, this is uh post-OCP. It was an insane show. There were so many, there were so many people that they had to like close down the registration, and people were buying uh entrance tickets on the black market, and uh the the halls were overcrowded, and uh the exhibition halls were full.

1:51 Very exciting show.

1:52 Yes, yes, it was something. Um, and I had a ton of fun talking to people uh on the show floor, talking to people outside. A lot of people ran into both you and I since we were there together and said, Hey, we listened to the pod, so thank you people who listen. It was fun to see you. Um, I honestly didn't sit in that many talks.

2:11 I sat in some, but it was there's just so many people to talk to, so much to see. And also the talks were being put on YouTube. So I thought, yeah, I can watch them later when I get home at 2x, you know.

2:23 Yeah, if you go to the Open Compute Project YouTube website, you'll see all the talks there. Uh totally free to watch. I think it's amazing that they did that pretty much on the same day. It it avoided a lot of FOMO for the people who were there. And uh people like us decided let's talk to more people, let's visit the show floor because they have wheeled in so much equipment.

2:44 And if nobody sees that, it's kind of a tragedy. And so we spent a lot of time on the show floor talking to people, and that was so much fun. Now I have all the YouTube videos as there's everybody else, and so we can go and watch it um at our leisure, at our correct spacing, whatever, right? Totally, totally.

3:00 Which I do have um some time blocked next week to catch up on them because there was there were so many good talks. Uh, OCP did a good job, couldn't get through them all. But one big talk in the topic for today, um, we're gonna do a micro explainer on micro emitters. So micro LEDs or microvixels or both.

3:18 Lots of micro to talk through. Um, so that was really the talk of the show. Uh Bill Brennan, CEO of Creano, had uh keynote very early on. I think it was on the very first day, and um it was surprising. Uh so let's just dive into maybe let's dive into the Credo news, but then we'll zoom out and talk about microemitters, micro LEDs, microvixels, and everything.

3:45 But yeah, do you want to give us the news first and then we can zoom out?

3:49 The whole stir up happened on the conference show floor where everybody's like, Oh, did you catch that talk by Bill Brennan radio? And I was like, Oh, wait, I was watching this stuff over here. But luckily, you know, the YouTube videos helped. I watched it immediately as soon as it was posted. Uh so the whole thing was the background is that for the longest time we've been thinking that micro LEDs are going to be the way that cables are made uh with light, because copper is kind of reaching its limits and what it can do in terms of reach at a given speed.

4:22 Optics has its like pros and cons. I mean, it's expensive, uh, it's power hungry, all of that stuff. So there's always this middle ground uh where it's a battleground for what kind of technology has to go in. And for the longest time, a lot of people were working on micro LEDs, and uh these LEDs are pretty much like a light bulb in the ceiling, but you know, it's pretty much a small condensed version of it.

4:48 And uh there are many companies working on this. Avicena, uh Credo acquired a company called Hyperloom for 92 million uh with this technology. Microsoft was working with uh MediaTech, I believe, on what they call Project Mosaic. Uh and uh Marvel was working with this company called Mojo Vision to develop micro LEDs.

5:11 So all of these companies are like artificial reality uh VR, AR, VR companies who had these micro LED technologies uh already in production, and they were like, okay, let's pivot to data centers because there's so much money to be made. So there was all this talk. But then what happened at the talk at OCP is that everybody thought this is going to be micro LEDs, or at least I did, maybe everybody didn't.

5:34 But he got on stage and said, Yeah, we were doing micro LEDs, but we decided we'll just do microvixels for the sake of reach. And this uh sparked in everybody's mind the feeling that okay, microvixels as a technology is the future, and micro LEDs are kind of no more. Uh, that was the sentiment on the show floor.

5:56 But I think there's a lot more than that. Uh, I don't know how much of it we'll be able to get into. It's really a long discussion. I plan to write an article on that as well. Uh, but yeah, no, let's at least cover the basics of what these things are. Why are we talking about these microemitters? What are the differences between micro LEDs and microvixels, and how to think of this going forward?

6:20 Yes, yes, sounds good. I agree. Also, everyone was basically like, what? Are micro LEDs dead? You know, that's like when we saw people that came right from that, like, I guess it's microvixels now. But um, you know, maybe also focused on credo really quick, and then we'll get into that specific technologies.

6:36 You know, anyone who's been following Credo, obviously there's this whole conversation over the past year of, you know, is it electrical, is it optics? And Credo was the first to come out with the AEC, the active electrical cable, which was as we continue to talk at higher and higher speeds, um, the reach of copper shrinked.

6:54 And Credo said, hey, if we essentially put some digital processing on both ends and throw a little bit of power in there, we can extend the reach of copper. And uh, but still, there's as you look forward on the roadmap, um, the question is, you know, well, how long will AECs last? And and there's long-range uh communication over optics using very powerful lasers, for example, but this kind of like medium reach, there's a question in like the portfolio gap of passive copper, AECs, maybe something else, ALCs, active light coppers or something or active light cables, and then you know, your traditional longer reach uh optics.

7:32 And in this sort of intermediate area is where people were expecting Credo to come out and have these ALCs driven by micro LEDs. But then, yes, the talk was about like, hey, we learned a lot from micro LEDs, and actually microvixels are a good fit here. Um, so that's again the framing, but now let's get into it.

7:50 So maybe start us with micro LEDs. Um, and then of course I want to go into microvixels because people have probably heard a little bit of micro LEDs over the past, but not as much about microvixels. Although, you know, we have had we had a conversation with Al Yoon of PicoJoule, which was helpful, and people can listen to that as well.

8:09 Um, but yeah, let's dive into micro LEDs.

8:13 Sounds good. I want to set the basis for what these are called microemitters. And essentially, like you mentioned, there's a reach gap. But I just wanted to mention one other thing about why this reach gap is important now. Because obviously, lasers, indium phosphide-based lasers can go a long way. And I want to mention why this material matters, because uh indium phosphide is in short supply, and everybody like you know is scrambling to make this happen because optics is becoming a big thing in AI.

8:40 But yeah, indium phosphide optics can go, you know, thousands of kilometers if you have to, you know. But we don't need all of that for the data center. And the point is that why do you use a technology that is A in short supply, B, uh capable of a lot more when you don't need it? Can we use the right technology for the right application?

8:59 Copper beyond seven meters at you know two twenty four gig is about, you know, like already pushing it. PAM6, uh, which is a new modulation scheme, is being considered to extend the reach of copper. So the copper work is going on on one side, but it's becoming very difficult. Optics is like an overkill technology with indium phosphide.

9:20 So there is a reach of between, say, five meters and 50 meters, where there's a lot of applications you could do other things. And this is important because now it's becoming way more important to expand the number of GPUs in the scale-up domain. And it is becoming too hard to put too many GPUs in a rack.

9:39 So even at if uh you know, we we were at the keynote from Nvidia at OCP, and um uh Ian Buck showed the picture of like eight Oberon racks connected together in an NVL 576, and of course, that picture showed that it was co-packaged optics, hooking them all up together. So that's Nvidia solution, but everybody need not have to use co-packaged optics or near package optics.

10:00 Um you know, instead, there is a way to connect these things with other forms of light, right? And so 50 meters is enough to connect a lot of this stuff. And the question is, what is that technology? That brings us to micro emitters. So this is what the industry industry has been working on. And I'd say that although microvixels have been the microvixels, I don't want to differentiate too much, but here for the sake of the micro aspect of it, we'll just call it microvixels, have been around a very long time, right?

10:30 Like vixels have been used in data centers. That's one thing to remember about it, and we'll cover more about this. But micro LEDs, and we should enter into talking about micro LEDs now, is that it is a far newer technology for interconnects. It has not been deployed in data communication applications like long.

10:48 You know, LEDs have been in displays and all of that stuff, but for communications, it's not has it hasn't been a mainstay. Whereas Vixels have been a mainstay of communication systems for a long time before we went to single uh mode optics, right? Uh, but you know, Vixels have been around, they have been deployed, they they do work at scale.

11:07 With limitations, but they work.

11:10 This episode of Semi-Doped is presented by Crusoe. They build the power, the data centers, and the cloud behind AI. Did you know Crusoe can help you tune an open source model on your own data for just a few bucks? Frontier models are very good, but they don't know your company's terminology, your part numbers, or how you like your reports formatted.

11:31 So you keep stuffing instructions and context into the prompt of an expensive frontier model to get the job done. But for simple, repeatable work like reports, an open source model trained on your data can actually match a much bigger model. Think of a new grad with three months of on-the-job training.

11:48 You don't always need the PhD, and the open model is cheaper and faster to run. Fine tuning used to mean renting GPUs, working on orchestration and Kubernetes, writing training code, and babysitting the entire setup process. But Crusoe's serverless fine tuning removes all of that. Upload your data sets, pick an open model like GLM, Llama, or DeepSeek, and they handle the training.

12:13 You pay per million token used, not per GPU hour. And on the smaller models like Quen38B, $2 can take you a long way with millions of training tokens available. Once the training is done, you can then deploy them on Cruso Cloud in just one click or download the weights and run them anywhere. Your data only fine-tunes your model.

12:34 And it doesn't have to be work data either. You can try it on how you write emails. Pick a task, compare the tuned model to your current prompt, and see how it does. Crusoe's giving all new accounts $5 in credit to get started today. That's enough to start tinkering with fine-tuning right now. Links in the description.

12:52 So, micro LED. So let's talk about this new uh incumbent technology, the new technology that wants to you know displace optics. Like, you know. So essentially, what it is is that it's literally an a diode. Okay, it's if anybody who's done like a basic electronics course, it's just like a P P type semiconductor with an N-type semiconductor.

13:16 It is made out of gallium nitride and it uh emits light in approximately in the 420 nanometer range, which is actually blue light. And you can see it in these interconnects that, like you know, if you look at the place where the micro uh LEDs are, you can see blue light being emitted. So it is blue LEDs that are doing this.

13:37 You have red LEDs, you have green LEDs, but anyway. So a lot of the uh work around micro LEDs has been at 420 nanometers. And uh the thing with micro LED uh light itself is that it emits light with based on what is called um spontaneous emission, which means the moment you apply a little bit of voltage to this diode, it turns on.

14:01 I mean, like it's it doesn't have like a threshold voltage, like the minimum voltage after which it turns on anything. It's like a continuously varying thing, it's like the dimmer switch in your in your wall, right? The moment you start raising the dimmer switch, uh, you know, you don't have to like push it 75% of the way up to for it to even turn on.

14:19 Okay. This is an important analogy to remember because we need to use the same framing when we get to uh talking about pixels. But LEDs just turn on as you increase the light, you know, this just like you're used to at home. So what ends up happening is that this is a different form of light emission called spontaneous emission, where it's just like sending out photons randomly and they're not very correlated to each other.

14:43 This photon decides to leave at this moment in time, the next photon decides to leave whenever it wants in time. They're not like coordinated, they don't march together, they don't do anything like that. Just leave. And then what are the other things like a like an LED in your home, right? This light is very, very broad angle light, you know.

14:59 It's not like a pencil beam of light that you can use as a laser pointer, you know. That is a pencil beam of light. But LEDs are like dispersive, they're like a wide angle. Okay, and so the the third feature, feature or bug, I don't know, of micro LEDs is um that they don't have a very good line width.

15:18 So if you look at it on a spectrum around 420 nanometers, they are a pretty broad spectrum. And so this is important because as this broad spectrum uh light uh goes along an optical fiber, uh, it faces what is called chromatic dispersion. And I know this is a lot of terms, but we we have to cover these things.

15:40 Yeah, yeah. Let me jump in here. So um, if it's supposed to be centered around 420 nanometers, when you say um uh there's like a broad dispersion, you're saying some come out at 416 nanometer wavelength and some come out at 422 nanometer wavelength. So that's what you're talking about here. So the light coming out has different wavelengths.

15:59 Yeah, that's that's what line width is. So ideally, you want if you draw the wavelengths of a light source, you know, across wavelengths, you want it to be like a single line. You don't want like half a light bulb-shaped um you know uh dispersion of like wavelengths. You can't have everything from 400 to 440 or something like that, which is approximately the the bandwidth of the this this light source.

16:24 It's about 10%, I would say. So you have anything from 400 to 440. So this is what results in chromatic dispersion. And let's get into quickly what that means. Uh chromatic means you know, color, right? Uh dispersion means kind of how they scatter or something, right? Like dispersed. So what happens is if you send 400 nanometers down an optical fiber, it takes, I don't know, let's say one second.

16:53 Okay now, 440 nanometers should also take one second so that they all arrive together, right? All the wavelengths should arrive together. Only then will, you know, it look like the same thing has arrived at the other end. But it doesn't, you know, different wavelengths travel different speeds. So what happens is like all these different wavelengths arrive at different times.

17:13 And when that happens, you know, when you send a pulse, that pulse has basically spread out in time. Which means that you can't send pulses fast enough because now they start to overlap. Right? So when you can't send pulses fast enough because they start to overlap at the other end, that means your data rate limited, right?

17:31 You can't send data fast enough. You can't keep sending bits because they're going to start overlapping. So you can only send bits slow enough so that you can still account for overlap at the end of the cable. Gotcha, gotcha.

17:42 So if you have a very precise line width, it's a lot easier to make sense of the signal once it propagates down uh the waveguide or whatever. Um, but if your wavelengths are spread out, then they kind of smear and maybe overlap. And by the time it gets to the end of the waveguide, it's harder to tell.

18:04 Was this a one or a zero? And to your point, um, then you can't send signals quite as close together, which ultimately means you can't send information as fast enough, right? So I'm assuming that you're leading us to say that the problem with LEDs is that they can't be used fast enough. Maybe that's like the problem for data centers.

18:23 Uh yes. So it's a combination of uh speed and reach. So you see, chromatic dispersion limits uh how much spreading happens. The further you send the signal, the more the spreading. So if you want to limit spreading, one way to limit it is that you use it only for like, I don't know, five meters, for example, right?

18:44 And then the spreading is limited. So you can send a reasonably high data rate. Now, if you're going to keep it like for 100 meters, it's going to spread so much that your data rate, you can't send data fast enough. They're going to overlap, right? Yeah, you have function.

18:57 Is that spreading? Let me ask you really quick. Is the spreading a function a function of distance because the photons are just bumping into like the waveguide or something? Like why does it yeah?

19:09 It'll spread more the the longer you go because think about um, you know, like if two cars are racing, right? Like a drag race, for example. Um two cars like arrive, they go a short distance in a drag race, but the cars like are literally like I don't know, microseconds or milliseconds of each other in a drag race, right?

19:32 Yeah. But uh to think about like um uh a cross-country race that happens for 24-hour cross-country or something. Yeah, yeah, yeah. Um you have the cars coming at it much more different time intervals, you know. The leader might come in first, the the other, the next car might come in like 10 seconds later, one minute later, things like that.

19:53 Got it, got it. It's the same thing. Uh, or you could think of this as a person running on a track, like the the 100-meter dash has like extremely close timings, but you know, the the 400 meter will have different spacings of people, and the marathon will have even more spacing.

20:09 Yes, yep.

20:10 Got it. Thank you. Carry on. The other thing is that micro LEDs fundamentally you can't turn them on and off fast enough uh to send data fast enough. Okay, this is a fundamental problem of the device, and um right now we can only uh send about, I don't know, maybe three to five gigabit per second on a cable um of like I don't know, five meter reach, right?

20:36 I mean that the the LEDs can only like turn on and off so quickly that you can send like five gigabit per second. So it's limited in both directions, right? The LEDs can't go faster inherently, and they don't have reach. So, you know, both ways.

20:50 So if they can only go three or five gigabits per second, but we're talking about a single lane wanting to, you know, first it was 50 G, then 100 G and 200G, and then eventually we want to go to 400G. How do you get 224 gigabits per second if an individual LED can only send at like three gigabits per second?

21:11 So you have to bundle a lot of fibers together. So you send many, many parallel lanes. Uh, for example, if you bundle about 300 of them together, you get a one terabit per second lane. I've did I've written a very detailed post on all of this stuff when I visited Avi Sena's office. Uh, we looked at the cables, we looked at the the the data that's being sent.

21:30 So check that out on my Substack. I also have a uh full free article on Substack where I discuss this uh line width trade-off in micro LEDs and compare them to how lasers look, for example. That's completely free. So if anybody wants to check that out. Yeah, but essentially, too, if you have such a slow data rate, you have to uh now put you know multiple parallel highway lanes together.

21:54 And you have to put 300 of them, thousand a thousand of them, you know, if you want to do both directions or One direction or what speed you want. You know, you just have to double up the lens. And people don't really like that approach because it gets complicated to couple like maybe 300 uh emitters, micro LEDs into the cable.

22:15 I mean, it's doable, it's not like it's not doable. You have to put um a lens to focus because remember, I told you the light was like dispersive everywhere, like it just like fans out everywhere. So you've got to put a lens to focus it and make sure it goes into the fiber. Then you've got to have a lot of fibers.

22:31 It's not an impossible task. The question is, do we really need so many lanes? Like, what if we can do it without having so many lanes? What if we could use a technology that actually had reach? What if we use the technology that could modulate data at a rate higher than three or five Gbps? Like, you know, that that is what brings us to microvixels.

22:54 Okay, so what are microvixels? So Vixel is a is a way that people say it, but uh it's actually an acronym. It says VCSE L. Okay, so that's vertical cavity surface emitting laser. Okay, so the L is a laser. So remember that this is a laser. This is not an LED. Okay, so it's not surface emitting LED, no, it's surface emitting laser.

23:22 So so it's vertical cavity. So instead of a lot of the optics that you know in silicon photonics and all that stuff that people talk about, it actually emits from the side of the chip. So vertical cavity means that the cavity is vertical and the light is emitting vertically. And uh our podcast with Al uh Al explains very nicely uh how all of this works, and he has like a prop two in that episode.

23:50 So people should check that out definitely if you're listening to microvixels, because you will learn a lot from an expert on the subject. So that's definitely uh episode worth checking out. But let's talk about this technology. So vixels are made out of gallium arsenide. Okay, so it's not indium phosphide, it's not gallium nitrite, but it's gallium arsenide.

24:10 And um vixels are quite commonly used uh in like a lot of stuff, like like I think face ID is one example, and you know, those are like uh like basically like lidar, you know, light, uh, you know, like radar with light kind of thing. So but then they just shine light on your face and they kind of identify you.

24:29 That's how it works. So there's like a big supply chain for vixels because you know, we have put them in billions of phones already, which is another benefit, right? Like I told you, like this has been used already in data communications. Also, the supply chain is enormous, and on top of that, it uses gallium arsenide, which has been a mainstay technology in like cell phones.

24:49 Like every amplifier, uh transmit amplifier called a power amplifier in your phone is built out of gallium arsenide. Okay, so this technology is like widely available. The material uh supply chain is amazing. Uh, you can build them on um eight-inch wafers, uh, so it's not like like indium phosphide, you're limited to like three inch, and now people are struggling to get to six inch.

25:11 No, but this you can build in eight inch already. So gallium arsenide is like widely available. So vixels are widely deployed in both communications and in phones and in LiDAR. Uh so it's like, why wait? So can we use this technology instead? Like, why don't we use this technology, right?

25:26 Got it, got it. So, okay, on the one side, we started with you have traditional Indian phosphide lasers, and um, you know, there's a lot of struggle manufacturing those. Uh, and on top of it, they're very powerful. And if you're just sending light, even though it's high speed, you're just sending it, you know, short distance between GPU racks, let's say, uh it feels like it could be overkill.

25:48 And then on top of that, you know, maybe you're going down the route of co-packaged optics or near packaged optics. So there's concerns about um maintenance as opposed to just a traditional pluggable cable. Um, on the other end, you know, you said, oh, well, we could use all these micro LEDs, and but now we're gonna take like a 200 gig signal, we're gonna have to have some sort of gearbox to split it down into being able to send you know hundreds of three gigabit per second lanes.

26:16 And so there's a little bit of a little bit of trigger trick needed there and a little bit of complication. But then on top of it, now we have like hundreds of fibers that we have to align to these hundreds of LEDs, and they can't go quite as fast. And so, you know, you can start to see some manufacturing trade-offs there.

26:31 But then you're saying microvixels are kind of land in between. It's still the laser technology, um, and it probably has a better line width and it can communicate faster as we'll get into. Um, but I at the same time, that also unlocks the ability to not need to have so like hundreds of fibers, presumably.

26:53 Um, so is that where microvixel lands is somewhere kind of in the middle?

26:58 Yes. So microvixel is in the middle, it's between micro LEDs and Indian phosphide lasers. So the reach can go, I think the publicized reach at least now is uh 50 meters, right? Like so it can go 50 meters and it can do data rates of 50 gigabit per second easily. So it can do 10 times a micro LED like data rates, it can do easily.

27:20 It can probably do 100 gigabit per second, it can probably do 200 gigabit per second. Like people are talking about doing that stuff, although it's very difficult because it has inherent limitations on how it works, but it can do 50. And uh, if you look at uh a lot of the demonstrations, actually, that people there are other startups working on this too.

27:41 It's not just Credo or uh someone else, but yeah, there are lots of companies working on this, but you know, they all run it at like 25 or 26 something. Yep. Uh you don't even have to run it at this capability, and that has its benefits. Okay, so we maybe I don't go into that immediately, but it helps to run it slow.

27:58 Okay, and I'll contrast why this is later. But essentially, um you you can choose the wavelength of this light to be either 850 nanometer, which is the widely deployed wavelength in all of like communications. Like 850 nanometer pixels are like well known, well studied. Okay. Then recently there's been a lot of talk about uh 1060 nanometer pixels, which is what you can see from Lumentum Coherent.

28:23 All of these companies um are working on 1060. It has its benefits. We let's not get into the details of like what wavelength is better. That's gonna be for my substrap. But uh essentially you have a couple of wavelengths uh that are like higher in frequency, right? This is like infrared, this is not blue, 420 nanometers blue.

28:44 This is infrared. So here uh you can uh modulate faster, the reach is faster. So what that means is to get like a uh one terabit per second uh 1.60 uh cable, you'll have to put fewer fibers because you can run each one at at least five times as fast. Instead of five, you run it at 25, and then you can probably use five times lesser cables to get the same optical fibers to get the same speed.

29:11 So that's uh that's good in a lot of ways, right? Like, you know, it's simpler um and you have a lot of design flexibility, and it really solves the reach up to 50 meters. You don't get limited by five meters, ten meters. Because, like you mentioned in the beginning of the episode, active electrical cables already can get to us in about seven meters reach with some tricks and you know, all of that.

29:34 But uh what what so the question is like, why don't we just go for reach? And so that's Credo's argument. They're like, let's just go for reach. Like, I want to I want a technology that takes me to 50, not to 10. I know 7 to 10 isn't enough to build up our product line. So we abandon micro LEDs. We're just going to go to Vixels, and so we get 50-meter reach.

29:52 Yes. And I will say one thing that stuck out to me about what uh Bill Brennan said was they can still get the same reliability, even though these are um lasers and everyone has concerns about reliability. Credo can unlock the similar reliability as copper because they're already making an array of these microemitters.

30:14 And hey, if you're gonna use, let's say, you know, 50 of them, why not have like 55 of them in there and have some spares? So if one emitter were to fail, you could still have some extra K, uh, some extra fibers and some extra emitters and kind of on the fly, always make sure that uh you have enough working emitters to be able to send the information.

30:39 Yeah, this is a feature that is common to both micro LEDs and microvixels. If you have parallel lanes, why don't you put like a few spare lanes and you could always switch over to the spare lanes? So this is not really a feature uh that is specific to microvixels. Sure. It's wide but slow, and this is the whole concept of the micro emitters.

30:57 We are going wide, many lanes, but slow. We are not going, you know, 200 gigs per lane like an Indian Fosfide laser. Uh we are going slow, we're going 25, we're going 50, or we're going even five or three, like you know, micro LEDs. So whatever it is, this this is slow, but we are using many lanes. And we're using many lanes, you can always have spares.

31:15 So this is not a unique feature, but it it's it's nice. Yeah, it's a nice feature of wide and slow overall.

31:20 Totally. But and then I guess the difference with microvixels is you could increase this reliability from seven meters all the way to 50 meters.

31:30 Yeah, yeah, yeah, exactly. So and that's a good thing, right? Like you get like a whole new domain of interconnect that uh does not use indium phosphide but solves a lot of problems. So, futuristically speaking, it seems like that's a better product option to build. And the fundamental reason I have to bring back one analogy I used.

31:48 You know, I used the light, uh, you know, the dimmer switch analogy. Uh, the reason that the uh uh microvixels can actually do a lot more than micro LEDs is that in this scenario, when you increase the voltage to a pixel, it doesn't emit any light. I mean, it does very, very weakly, but you know, it doesn't really emit any light till you like bring it up to like at least halfway through, you know, and then bam, you've got an explosion of light.

32:12 All right. That is the the called uh stimulated emission of lasers, as opposed to spontaneous emission in micro LEDs, right? So stimulated emission means that all the photons now are in phase, like holding hands and marching, like in a military regimen, like forward, and the line width is narrow because all of these photons have a purpose.

32:33 Um they're just like marching, right? So you've got on the line width uh the spectrum of wavelengths, pixels are like narrower line widths, and they're not as wide as micro LEDs. And so the same chromatic dispersion argument uh applies here because now your narrower line width will take your signal further before it spreads out.

32:54 It still spreads. Chromatic dispersion happens to every form of light, every everything, right? But it happens uh it only at a longer distance because the line width is narrower. So it has like more room to spread before bits start to overlap. Gotcha, gotcha.

33:10 So you have stimulated emission, a tighter line width, which means you can send information further, and already you can send information faster than you could with LEDs. And yet, by the way, to your point, this is gallium arsenide, so an abundant supply. Um, and it's still a pluggable form factor. So, right, we're not talking about NPO or CPO.

33:32 This is just a cable that you plug in on both ends.

33:35 With Credo, yes, but there is plenty of talk about using microvixels for NPO, as we saw at the Astera booth. Uh, so those kinds of things are um in discussion, and there were some nice things we saw at uh the the show floor. But all of that is coming, yes. But even now I think you know we are pretty early in the whole vixel uh for interconnecting because even Credo's booth did not actually have a cable that was like showing electrical data on microvixels.

34:11 Uh, this they had some micro LED stuff, but so this the switchover I I feel is like very recent, and uh there's a lot of work to be done, and there's gonna be a lot more stuff. I don't think that we are done with micro LEDs or it's dead or anything like that, because that's why even Credo has formulated it as a microemitter, that's the broad technology.

34:34 There are still use cases for micro LEDs, you know. Um, and we the industry is still exploring them. But yeah, essentially they launched a work stream inside OCP called the short reach microemitter optics. Like, so it's SRMO work stream. So the the whole point of this is like a bunch of companies are getting together to talk about how to do short-reach microemitter optics.

34:58 So there's a lot of things, a lot of stuff that's gonna happen. This is not by in any way settled, but I think the industry has, at least for the near term, chosen a part of microvixels. It seems that way, at least.

35:12 Okay, okay. So microemitters is broad for it, could be microvixels, it could be micro LEDs, but you're saying it's gonna be in the near term, it's gonna be microvixels because of all the benefits we discussed. Um, it can reach you know 50 meters, it'll be pluggables first, is the pragmatic thing, but it the same technology can be used for near-packaged and for co-packaged, and we'll probably work our way there.

35:37 Um, I guess then lastly, what will we see micro LEDs anywhere? And if so, where will we see them?

35:45 Micro LEDs are still good for one fundamental reason that I think a lot of the discussion misses is that they are extremely uh heat insensitive. Gallium nitrite is what they call a high band gap semiconductor, which means that they the operation of the light source is very insensitive to temperature, and that's important in some applications.

36:11 I would think one of those applications is scale in, where you use optics within the tray. The reach is not going to be, I don't know, like meters, right? Like a tray, the server tray is only this bag, this big, like about a few feet. And uh within that is a very hot environment. It's very difficult to replace anything that goes bad there.

36:33 So you want a very high reliability source of light. I don't know the reliability numbers per se, but gallium nitride is in general a much better suited uh technology for something like that.

36:45 Gotcha. So it has a high band gap. So even if it's in a very harsh, hot environment, like chip to chip or die to die, where you're just in the server and it's just like kind of always hot. Um the heat isn't going to cause electrons to have all this extra energy and like jump the band gap and start to like cause light to be emitted, right?

37:07 There's the high band gap, it's like, oh, sure, there's extra heat, but um the molecules inside this particular diode are uh not, it doesn't cause the uh LED to turn on, so to speak.

37:21 Right, right. So that's the whole thing. So it's high band. I have a whole article on what band gap actually means on the substack as well. So go look at that if you want to understand what we're gonna do.

37:32 Put all of that in the show notes. Yeah.

37:34 Yeah, yeah. Link it uh, and then we can um, you know, because yeah, what you're explaining is right, if you have a high band gap, then heat cannot cause electrons to switch states and emit light. That's the broad explanation. But that is an advantage, and I don't think we are done with this technology yet.

37:52 I think we're still going to continue looking at it. That's my that's my broad take, although a lot of uh excitement is currently there around micropixels for good reason, as we all as we discussed here, right?

38:05 Got it. All right, man. Well, I think this was a good micro explainer on micro emitters from micro LEDs to micro pixels. Uh and with that, uh, any closing thoughts or should we just call it here?

38:18 Uh no, this is good enough for this episode because there's like a lot more to come. Uh, we spoke to several exciting uh builders in this space, uh, and we would rather have them come on the podcast to talk about uh all these technologies. There are we spoke to people who are all willing to come on semidoped, thankfully, who are building uh the lights, light stuff, the chip stuff, and so we have so much exciting content planned after OCP.

38:47 Yes, love it. So we gave you the introduction here, and we'll bring the experts on and let them take it one level deeper. So, with that, people, thank you for listening to Semidoped. Uh, like it, subscribe it, share it with people. We appreciate all your comments, and we'll catch you next time.