PicoJool's Al Yuen: The Case for GaAs VCSELs in Scale-Up Interconnects

(upbeat music) Hello everyone. Today we have a special guest, Al, UN, CEO of PicoJuil. PicoJuil is an optical connectivity company and we'll get into all the interesting details, but first I want to introduce you guys to Al. So Al, tell us about you and your background. I know you've been in the industry for a long time. - Yes, so after grad school at UC Center Barbera, where a lot of the photonics folks have originated, I went into HP, HP Labs, where we focused on photonics, research, et cetera. And since then, around '99 I left HP and started my first company called Alvesta and we created the world's first 10 gigabit ethernet. At the time, 10 gigabit, which is 10 billion bits per second, we were actually trying to figure out applications at how people would use this in '99. We would make up things like people who wanted to stream video someday and all the rooms in their house. And so we had to actually, but today of course, we're doing 1,600 gigabit or 1.6 terabits. So since then, I've gone to various companies, I ran a division for coherent and then started some other solar companies and clean tech and then finally ended up at Lumentum. In my last gig and then about two years ago, through Playground Global, which is our funder, we started Peekle Jewel and so far, so good. And what we're doing right now is basically in the interconnect space and we'll talk more about that today. Awesome. Wow. What a great background. I love that you guys invented early 10 gigabit ethernet and then had to create ideas to sell people, to convince people that like, yes, this is useful, people will want to use it. That's awesome. Now remind me, you also helped invent the active optical cable. Is that true? Yeah, so very interesting story. So back in Alvesta, our first startup, again, another small startup at the time called Melanox, which of course now is inside NVIDIA and created this whole kind of hyper scale and the whole Infiniband, et cetera. So they approached this. So they had these very bulky copper cables, even back then. So 25 years ago. And they said, you know what, the copper cables are very bulky. They could only reach, you know, at that time, tens of meters. Now it's even shorter. But they said we were like an optical option, but we don't really want to commit to a full kind of optical solution. So could you put the optics inside the connector? And we said, why not? So basically we took that same transceiver that's typically on a board. And we literally embedded it right into the connector, as you see here. We then we thought, well, that's kind of, you know, not very efficient and clean. And so we embedded that whole connector inside. So the optics went inside the connector. And then this is the world's first demo of an active optical cable, meaning electrical to electrical. So electrical comes in, electrical goes out, but inside the electrical to optical transition. So electrons come in, photons carry the information, and then electrons go back to the point B. So that's how the whole active optical cable concept came through Melanox. And since then, for the last 25 years, the AOC has been the standard workhorse and many, many data centers today. Amazing. So having invented that and then watched it sort of become just widely adopted and probably produced, you know, I don't know, in the millions of cables or something, how has that impacted the way you think about like what's possible as an entrepreneur and in this space? - I like to think of ourselves more as engineers, right? So there's a difference between, I would say, scientists or researchers, or what people call R&D, right, research and development, and engineering or product development. And so I always kind of tell people I'm more of an engineer and engineers solve problems. And they want to basically create products that are that kind of basically meet specifications, right? So if you need a Prius, you certainly don't design a Ferrari, for example, right? So that's a little bit overkill. And so therefore the product fits the need to spec costs, reliability in our case, reach, or at the number of amount of power, et cetera. So in answering your question, I feel like what we do is really look at trying to solve the specific problem. And today, right now, copper has shrunk to about three meters, four meters reach at 200 gigabits per second per lane. And therefore they can't get the information off each rack, right, so racks and racks of these GPUs, CPUs, lots of compute power. So these racks are getting very hot because they're packing more and more GPUs per rack because they can't exit the rack. And so the exit the rack, you need an optical solution. And so there's many technologies. And of course, Vixels based Active Optical Cable is one of them. And that's what we're basically focused on. Solving this problem of low-cost, highly reliable, again, going way back to our roots, which is replacing a copper cable with an active optical cable. So the problem really hasn't changed is just that the speed and the aggregate bandwidth is now 1600 times more than the old one gigabit ethernet. So it's exciting. It's been a long journey, but we're still at it. And then we see a long future for Vixels based technology as we go forward. Nice, nice. OK, so you have this history as thinking like an engineer of just like, what is the problem that's right ahead of us in the industry and not saying, like, we need to invent new physics, but just like, how can we be thoughtful about, you know, oh, same cable form factor, for example, in the AOC case, electrons and electrons out, but we could use optics here and could we put the transceiver, you know, on the end of the cable. So, you know, thinking very pragmatically. And now, you know, fast forward 25 years or so. And I guess in 2024, you started pico-jewel. And it sounds like you're going to tackle the problem of, again, of communicating lots of information. And this time, it's, you know, GPUs that are talking to each other, like, rack to rack, where it's such high bandwidth, copper is shrinking to three meters or so. And so you guys, again, are thinking about how do we use, essentially, active optical cables, but this time with Vixels. So tell us, like, why Vixels and what about the problem made you want to start pico-jewel to actually start a company and, like, get in the game here. - Great question. So, Vixels, for one thing, has been around since '96, right? So it's a technology that's been in product in the field, in data centers since 1996, starting with the first gigabit ethernet. So, you know, I'm more of a historian today. So there's a one gigabit ethernet that, you know, HP created and at the time Honeywell, looks at a very, very early companies that did one gigabit. And so from the one gigabit all the way to today, again, 1,600 gigabit, or what they call 1.6 terabits, has all been Vixel-based. And going back to what I was saying, the practicality of these solutions has to meet capacity, demand, cost, all of those, right? You need to check all of those lists. If you're trying to get into a hyper-scaler today, you need to basically meet all that checklist. You can't say, oh, I meet everything, but it's three times the cost of your perturget. Or everything is great, but I can't get the reach. So they want all of that met. And so right now, the only thing is copper, right? And that's mainly driven because of the cost. Cost for copper, of course, is very, very minimal. Even active copper, where you have some signal integrity, signal processing, et cetera, built in like an AEC, active electrical copper. The cost is still quite minimal compared to other technologies that are more elegant, longer reach, et cetera. So getting back to your question, what's very important in a data center, if you have different fiber optic communication, I think everybody hears it and goes, oh, isn't that over the ocean? They have these trans-elantic sub-CE type of cables and said, absolutely. And so there's many different flavors of optical communication. So when you say that, it doesn't cover all of them, right? So it's a very big umbrella. And what we're talking about today for data centers is a very short reach. Today they call it scale up. So it's basically a row, typically 25, 30 meters, right? Or in English terms, 75 feet. OK, so it's a very short, you know, within your house, basically from one end to the other, and potentially even shorter. So in that very short reach, what happens in communication is the shorter the reach, typically the higher the volume. So if you think of it, it is-- you have a lot in the data center where you're trying to connect all of these different GPUs, CPUs, or ASICs together. And so there's millions of interconnections, miles of fiber inside a data center, which may be a football field length today. But when you leave that data center, and you go off between cities, between buildings, between countries, you have fewer fiber, but you need longer distance. So getting back to the data center now, and long kind of answered you a question is, You have to have millions per month capacity to meet the volume demand. So that's one of the boxes where you got to check off all the boxes. So if you're able to demonstrate one or two racks or go to a show and demonstrate just technology, again, technology, science, R&D shows the capability. That's called a demonstration. But in order to ship in volume millions per month, you need this whole ecosystem. You need people who make the connector. You need people who make the chanceevers. You need people who make the sockets. So everything has to be millions per month. And any one of those build-in material parts that go into these chanceevers, if it's missing, then you can't really ship in millions per month. And that's what's happening with a lot of these new technologies that are fantastic, by the way, from a demonstration, from a capability going into the future. But VIXIL's domination is that the last 25 years, we've been shipping in the millions per month. So therefore, there is no invention of a technology or capacity or foundries in order to build these up. We already have that. So now we design the latest VIXIL. So today it's 200 gigabit, which we just announced. And then we put it into the whole ecosystem and they package together. And then while we can build millions per month without waiting for machines to be or actually even buildings to be built up and then new machines and then new process to develop, all of that has existed for the last 25 years to get service this very short-reach data center application. Okay, so yeah, let me reflect it back to you. So the problem that you're trying to solve is scale up with optical interconnects. But the key insight that you have, kind of with your pragmatic hat on, is how can we use technology and a supply chain that already exists and can already ship millions of cables, components, parts per month. And so that's one reason why VIXIL's is so attractive is because they're not a new technology. The supply chain is not new. So even though as we hear about all these other interesting things, because I think anyone paying attention, you know, listens and they say, "Oh, you know, the Broadcoms are coherent, so Luminthums I hear about Silicon photonics. I hear about EMLs. You know, I hear people talk about micro LEDs, but you're saying like, "Hey, don't rule out VIXILs even though they're not new. They're still, there's advantages to VIXILs." But tell us, like, if VIXILs have been around for so long, like, why aren't other folks trying to take VIXILs to the 200 gig per lane times 8 lanes, 1.6 T and beyond? Yeah, so like you said, there's multiple technologies, silicon for tonics, EML, even the micro LEDs that are coming, all vying for this 200 gigabit type of electrical signal coming in. So from the ASIC GPU CPU, you have 200 G per lane of electrical signal, right? That's coming in. Now, when you go to the electrical to photonic or E to O type of transition, then it can go to different lanes through, again, a different IC that may be a gearbox, they call it. And so you don't necessarily have to run at 200 G for a straight through, which we can, right? So that's the most elegant is if you can go straight through without having to go through it intermediary gearbox, then it saves power, costs, and the latency, which is the delay going through there. So 200 gigabit would be ideal, but what's changed in hyperscale is different from Ethernet. So Ethernet information is sent in what we call packets. So whatever information, oh, where should I go in Italy? I'm going on vacation to that. So that information is divvied up by your search engine and sent in packets, and then that comes back to you. And we're not very cognizant if there's an error drop or there is some delay because those are in the hundreds of nanoseconds or milliseconds into us. It's like, you know, 1,000th of a second. We're just not going to delay. But for hyperscale systems, which are literally made up of thousands of GPUs, acting as one brain, if you will, right? One supercomputer, one high performance kind of cluster of computers. Then that latency is super critical because the GPU notices anything in the tens of nanoseconds or hundreds of nanoseconds. So therefore, the typical Ethernet bit error rate, which is the measurement of how much errors you're getting per second. And so if that needs to drop from 10 to the minus 6 to below 10 to the minus 10, or what we call error free today, because any errors then slows down the whole training inference, all the AI infrastructure that's needed. So that has changed and allowed these very, very, you know, typically the higher cost single mode. Oh, I'll mention that. Single mode being longer distance, longer reach, very high performance for long distance. Now has come into the data center into the hyperscale systems because of this requirement for very low bit error rate, very high performance computing and connectivity. And now VIXILs has to raise its bar from 10 to the minus 6 typical Ethernet to the last 25 years to 10 to the minus 10, 10 to the minus 12. And we've done that. Right. So now we've pushed our VIXILs to 200 gigabit, even though you can use it 100 gigabit and 50 gigabit, and RZ, you can still leverage that to very low bit error rate. So the answer is that things have changed for hyperscalers to require error free. And that's allowed all these very high end single mode solutions now to compete directly with VIXILs because of that additional specification that's new to hyperscale AI systems. Okay. Okay. So because the capacity for sustaining errors is much, much lower, we don't want all these GPUs to just be waiting. That sort of changes the game from like the cloud sass day to now when everything's acting as one big computer. And so this bit error rate has been much, much lower. And so VIXILs, which are short reach, have to either, like essentially come down to meet that or what you're saying is these other technologies that are longer reach higher power. Those already were closer in the demanded, like the necessary bit error rate. And so people are saying, oh, why don't we take those technologies and bring them into short reach? But surely that has like power and cost trade offs of taking something that could talk at low error rate, like over a long distance and trying to bring it in. And so you guys must be taking a different tact and saying like, no, no, no, like let's just make VIXILs error free essentially. And presumably that's like a cost or a power trade off that you'd rather go with, or is it back to the manufacturing supply chain capacity? Yeah. So again, it's a quite complicated matrix of items you have to check off. Right. So those who were shipping Silicon for tonics and EML for long reach or DFBs, etc. Single mode, very high performance devices have been around for, again, similar time, 25 years. And they've been used for long reach because they have that super high performance, very few fibers. And then what we would call WDM, wavelength division multiplexing. So fibers are very expensive and you're going hundreds of miles, hundreds of kilometers. And so you want to use very few fibers, but then pass more information through more wavelengths, more colors in that same fiber. For again, short reach when you're talking about tens of meters, you're not as kind of locked into the cost of the fiber. It comes down because you're only 10 meters versus 10 you know, kilometers or extending. So therefore, the capacity for volume supply chain of the very, very high performance low bidder rate would be a natural, right. They come in hyperscalers, want low bidder rate. Let's go with the far away, right. Let's go with the super high speed, you know, high performance single mode. But they've been used to building in the maybe 100k, 100,000 kind of volumes because you don't need as many of those between cities, between countries. And all of a sudden they come into the data center, even though their performance is excellent, cost is a little bit higher because of single mode nature of packaging. However, just the infrastructure to build millions for month now is 10, 20, 50x would exist. So that's brick and mortar. That's basically saying, I only have one pizza oven and I've been used to, you know, a small clientele that I can make maybe about 20 an hour. So I'm going to go, is it, I like to place an order for 5,000 pizzas and I need them in an hour. So you're going, I need, I need basically 100 pizza ovens. That's the exact same problem that the very high performance single mode long reach traditional having coming into. So Silicon Phytonics, EML, excellent, excellent technology, very, very good bit array, all of those. But then the infrastructure needs to be built up. And that's what you're seeing, right. You're seeing a lot of announcements with people holding shovels and saying, we're investing in the next, you know, supply chain, the buildings, etc. And that's great, right. It's certainly bringing more manufacturing, not only in the world, but back to the US. And those are all great for the industry as an old. But Vixels have been around for 25 years and shipping in the millions. So we're not building, we're not putting shovel to ground, et cetera. We're just changing the actual VIXAL performance, the chip, and then getting leveraging the existing infrastructure. And so what we call for VIXAL capacity, it's unconstrained. Right. So there's constrained, meaning they're sold out. We're sold out through next year. If you're going to place an order, it's going to be have to be eight months, 18 months lead time, a year and a half from now, we can get it to you. For us, unconstrained just means we have a certain lead time that's basically only limited by our cycle time of building through the factory. So for us, you know, a VIXAL run, and then a packaging run may be for eight, 12 weeks, but that's limited just the fact that we have to build it out and ship it. But it's not limited by the constraint of the supply and ecosystem of machines or pizza ovens. We have plenty of pizza ovens, place the order, we'll get you your order in the cycle time that we commit to. Gotcha. Okay, that's that's very interesting and a great sort of competitive advantage there for you. So on the constrained side, is that what we hear about for listeners, like when they hear about like Indian phosphide being a bottleneck and just like, or is it particular like where in the supply chain is it constrained? And then for for you, like what's different about VIXALs that does make it unconstrained? Yeah, so for Indian phosphide, a lot of the silicon photonics and EMLs, et cetera, is based on this material, Indian phosphide for VIXALs, our technology has always been gallium arsenide. I don't know for the listeners, that may be okay, one three five compound versus the other, what's the difference? Indian phosphide is material constraint from the very beginning, like you can't even get a base substrate, right? Even before you process it, just the substrates for Indian phosphide are limited before you get it all done, and basically made into a product, whether TML, silicon, photonics or VIXALs, the Indian phosphide substrate, bare material is already limited, gallium arsenide, unconstrained. So we start with that. And now you go through the fabs, and then of course fabrication, et cetera, depends on foundries, they're usually very large companies that do foundries, companies like Peacled Jewel and others, we don't own large clean room factories that make these, we design the VIXAL, we design the individual kind of VIXAL chip device, and then we use foundries to manufacture, right? So those foundries are available, but they can't get enough Indian phosphide starting material to do that. Now, after that, again, once you get to the chip level, you dice up, you go, okay, I've got the laser, I'm ready to go now to get to from the chip to a plugable device, right, to an actual optical engine, if you will, there's a ton of stuff happens, right? Now you've got laser drivers, you got boards, and then for single mode, you have to have all the machines that align that particular silicon photonics, EML to a very, very small core single mode fiber. So those machines have to be readily available, et cetera. And so VIXALs, again, has that millions per month type of volume infrastructure doesn't need to be built up. And then we get back to from the very beginning, Indian phosphide material constraint, and then you have to build it into lasers, and then finally, you have to package it into transceivers. And at all along that supply chain, it's not used to building millions per month. So all of that has to be built up like hardware machines, alignment machines, testers, et cetera, to get there. Again, VIXALs has all of that infrastructure existing already. I see. Yeah, that makes a ton of sense. I love your props, by the way. I liked the little VIXAL that you held up. So tell us more. So you, yeah, that's awesome. Maybe tell us what we're looking at. And then walk us through exactly what you designed and then kind of where it gets handed off and built and packaged. And yeah, where your responsibilities end. Yeah. So basically, again, background is super important. Right. So there we call kind of this tree of knowledge of VIXALs. Right. So you have this line from Honeywell through Finisar. And then Finisar goes into two six and then goes into coherent. So this is coherent line. And then for us was obviously HP went into Avago, went into Bracom. And there's this kind of HP Bracom line. And then finally, there's this, I would say Pico like E2O and I'm throwing some old names from 25, 30 years ago. They go into JDSU, which is another big name in the 2000.com time. And then JDSU spins off Luminum and VI. And so we're in the Luminum arm. So all of these three major arms have a lot of VIXAL knowledge. And so the strength of Pico jewel is we've tapped into and we have people, designers from all three of these branches. So imagine that all the know how, again, not patented know how about recipes. And you know, I used the example, you can hand three different chefs a recipe for making soufflé. And most likely you're going to get three different types because it's really difficult to get a perfect soufflé, if you will. It's not just getting some, oh, you get the egg, you crack the egg, you beat the egg. So it's a lot more to them. The same thing is in the VIXAL. So answer your question, what do we do? So we take all that know how and then we design the epi layers. So you see all these little kind of lines. So these are epi layers that are designing this vertical cavity. So it's not an edgy meeting. So edgy meeting basically is a flat chip. And then the light comes out of the edge, surface submitting, right, VIXAL's vertical cavity surface submitting comes out of the surface. And so we design all of the kind of internal cavity of the laser, all the doping's and not getting into too much detail, but all the process. So after we design, we hand it over to a epi foundry that grows this material, they give us back a epi wafer that's unprocessed. And then we've been working with wind, some I can doctor in Taiwan, that's our foundry. So once that epi wafer is ready, we hand it over to wind. And then they process it into their clean room process. And then they make the actual final VIXAL device. And then the beauty of the VIXAL is also compared to an edgy mirror is at the wafer level, you can start testing and probing each one of these 100% tested what they call known good die before you have to simulate it and dice it up into a raise, et cetera. So that advantage is huge because if you have to add the more work you add in before you quote yield the device, whether determine whether it's good or not, the better the cost, right? Because you always want to yield upstream, the more value you add, and then you yield downstream, you lose all that value added. So wafer level testing for VIXALs is really, really advantage of versus the other kind of edgy meeting type of technologies. Nice. Man, I'm learning so much history and so much sort of one VIXALs 101 here. That's awesome. So you work with wind, you design the VIXAL, you have an epi partner who helps with that wind is able to even do the testing at the wafer level because it's a VIXAL known good die. Ultimately, it gets built into the device. So they must be doing the package, like some level of packaging for you to win. Right. So wind is only on the wafer processing. So they come to the wafer and then it's stifed up into individual VIXALs and these individual VIXALs, we work with our partners to build into either active optical cables, right, or chanceevers, et cetera. And there again, another foundry, if you will, but this is a packaging company. However, with that said, companies like TSMC, right, Taiwan semiconductor now is going into co packaging, meaning after they make their silicon wafer, they'll start packaging the optics directly on top of their wafer, which is co package. So there's another whole field that's growing called CPO, et cetera, where the traditional semiconductor foundries that are kind of wafer processing. Now are stacking up different technologies together. And they call it 3D kind of wafer level plan of packaging. So that's all emerging for us. We only do the wafer at when and then that wind VIXAL goes to our module integrators, partners, and then they'll build it up into the active optical cables or chanceevers. Gotcha. That's helpful. So then take us back to your roadmap. I know you mentioned a 50G version, 100G, 200G. Can you tell us more about, and I know you also mentioned a recent launch. So tell us more about your roadmap, what you launch, what you announced. As I mentioned early on, 200 gigabit per lane is kind of the benchmark, right? So bar you have to clear and EMLs, Silicon for Tonics have all done that. And now VIXALs have reached that, particularly just announced our 200 gigabit will start sampling next quarter. And so that's for a very simple chanceever where you have eight channels of 200 gigabit coming in. The aggregate or the combined bandwidth of that, eight channels of eight by 200 is 1600 gigabits or 1.6 terabits. So So that's the standard ramping today. There's 800 gigabit chanceevers as well. That's also shipping. That's 8 by 100. And then the next generation or today's generation is 8 by 200. So that's the 200 gigabit pixel that we announced. However, there's many, many different flavors of that because of the specification. So aggregate bandwidth 1.6T, check. But there's different ways to get there if you want very low power or very low bid error rate. So running 200 gig, I liken it to a Ferrari. It can give you the 200 miles per hour, but it's a very high end, relatively expensive because you need certain signal integrity, signal processing, all of that that adds to that. So now I say, I want a very low cost low power, but I still want low bid error rate. And then what people have done is let's slow it down. Let's use the 200G performance and the VIXL, but then actually run it at 100 gigabit. So now you have an excellent, excellent VIXL that gives you a lot more performance. And if you use it at half the speed, you really get much better bid error rates or the signal to noise ratio or relatively intensity noise drops as well. So that's 100G. So but you need more lanes. So to get to 1600, you need 16 lanes of 100. And then recently, something came out called micro VIXLs. And that's going even slower, down to 50G. And then they call it NRZ. So instead of pan four, which has four levels, 0, 1, 2, 3, now we go back to the original NRZ, which is 0 and 1. So now you have the use of the entire 0 to 1 signal to noise, which again reduces your bid error rate. But you need more channels. So you need 32 channels at 50G to get to 1.6T. But all three, we are shipping, and all three are in demand by customers, depending on whether they want what they call fast and narrow, by 8 by 200. Or they want a LPO, linear drive, no DSP, low power. And that would be a 16 by 100G. And then finally, if they want really, really low bid error rate, very, very low power, then they go to the 32 by 50G NRZ, which they call slow and wide. We tend to call it fast and wide and then faster and narrow. But when you're in the high speed interconnection, we try not to use slow in any of our marketing. That's good. That's good. OK, interesting. So this is really cool. So you're saying, OK, there's many different ways to get to 1.6T. You could have 8 times 200, which would use pan four, require a lot of DSP and power. But it's definitely possible. Or you could do 16 times 100 or 32 times 50. And you need less DSP for each of those. The fast and wide 32 times 50 has much less-- because it's NRZ, so less DSP and stuff. Yeah, this is all very fastening. So will that approach still hold once you move to 3.2T? Is it going to be that combination of different possibilities? Great question, because people say, in any time you have a technology, they always say, what's the rope map ahead? What's the future? Is this the end of the road? OK, 1.6T, we get it. VIXALS can do it. But is there a 3.2T? Is there a 6.4T? Is there a 12.80? I mean, Andy Bechtelshin, Nertorius. He's created an XPO that's literally going to give you 12.80T and a big plugable today. So they're thinking way ahead. They're planning way ahead. Because no one has ever told us in the last 30 years, oh, whoa, whoa, we have way too much bandwidth. So we have to have that bandwidth. And exactly like you said, what's the future? So number one, we can use something called bi-dye, bi-directional. So meaning we can just add another wavelength, not the complexity of a WDM where you have 8 or 16 wavelengths, like single mode that would do. But we basically just add another wavelength to our existing one. So two wavelengths passing them in both directions, so bi-directional. That doubles the bandwidth without changing anything else except for you just add another laser at a different wavelength. And you can leverage the entire ecosystem. So from 1.6 to 3.2, we can add another wavelength. The other ways to do it, of course, is to increase double the speed. So can we do 100G NRZ? That's in the works. So we're developing 100G NRZ. Today is 50G NRZ. But we're developing 100G NRZ, leveraging our 200 gigabit Vixel running at 100G NRZ, et cetera. And then in the future, we can go to more channels. So the beauty of Vixel is against surface emitting. Means for edge emitters, you can only have a one-dimensional rate. So you can have a one-by-four, one-by-eight, one-by-12. But it just makes you a long bar, if you will. But for surface emitting, we can have a two-dimensional rate. Meaning I could do two-by-four, two-by-12, two-by-16. And then essentially, a couple of the light very elegantly with a optical fiber bundle to do that. And then in that case, I'm kind of unlimited, if you will. So I can go up to 64 channels today in a four-by-16 connector that's the size of-- let me just show you-- so a four-by-16 fiber is this size. Here's my finger. [LAUGHTER] And that has 64 channels in there. If I run them at 200, that gets me to 12.80. So in essence, the technology of today, without having to go to 400G per lane, which we're also looking at, but at 200G, with more channels, with more colors, like another color for by-die, you double, you triple, by size, et cetera. So that roadmap to 12.80, we believe, is very solid, very clear, without even having to invent any new technology to get there. And then with new technology, it just gets easier if you can do a 400G per lane, et cetera. Sure. Fascinating. So it's just the same 200G VIXL kind of over and over. It's just-- if you want to put it in an array, and you get more of those, or are you having to invent a new VIXL to try to get it the 100G version to run at NRZ? So we're just starting to test. We believe that the 200G VIXL has the capability to go to 100G NRZ. So it's not a new VIXL. It's just basically using a different coding with respect to the signal coming in, running at NRZ instead of Pam for. So that's just to take advantage of the 0 to 1, using up the whole signal to noise ratio as one bit, as opposed to four bits. So that's the difference. Gotcha. Yeah, this goes back to your engineering pragmatic mindset of just taking a Lego block and figuring out different ways to place it or different ways to use it to sort of unlock this whole roadmap and the future roadmap. That's pretty cool. So OK, so if a hyper-scaler-- you talked about unconstrained Gaia Mars Knight and working with Wynn, and they're used to making this stuff. They've got all the pizza ovens they need. So if a big hyper-scaler comes to Pico Jewel and says, we want a million of your pizzas, what does that look like? Like, how does that actually happen? Right, so if they want a million vixels, then we would give them typically an eight week, 10 week, kind of lead time. That's kind of the basic. Obviously, we can accelerate that and kind of push and have engineering carry certain wafers. But typically, eight to 10 weeks on the vixels side. So you get a wafer or you get individually diced vixels. If you want transceivers, then that eight weeks tags on a certain number of weeks to package it all into the chanceeaver. So but those are, again, constrained strictly by the process of packaging, not about ordering equipment or having to build capacity, et cetera. That ecosystem works, and it just leverages the typical cycle time we call it of building out. So typical cycle time, eight weeks for the vixels device, and then potentially four or six weeks for the additional module and did after that. And so you're looking at anywhere from 12 weeks to 16 weeks to get to the full module starting from a epic reactor, growing epic and going all the way through. And so we'll continue to drive that lower, as far as lead time or cycle time. But also yields-- I didn't really mention too much-- yields are a way to say, if I make a vixel, how many known good die can I get out of this wafer, or this particular process, and the higher the yield up to, obviously, 100%, the fewer waifers I have to run through, the higher the capacity of this factory, if you will. Right? So if every wafer goes through and I get 100%, then I need fewer waifers, and therefore I need less capacity for the demand. But obviously getting to 100% is very hard. but Vixels have been perfecting the that process for many years, for many decades, and now we're leveraging all of that. It's not new, it's not something that has to be established. It's not based on new technology. So that's very important, since we've, when has been doing it for about 10 years, since we transferred that for consumer electronics application back in 2016, they have a capacity of up to 1,000 of these wafers per week, right? So that capability, and then the other thing is there's about 240,000 vixels on each wafer, right? Because they're really tiny. So they're very, very tiny. And so that adds up to a million yielded, maybe 10 wafers. So the capacity is huge for data. So I think we have no worries for us. Once we get to the 200G and the product specs are met with the customer reliability, qualifications then we just ramp readily with win, and they're ready to go. Nice, amazing. It's quite compelling. I think normally when people hear, oh, there's a startup that's trying to compete in a space that has these huge incumbents, the question is, well, how's this startup going to compete? How are they going to get to market? How are they going to find customers? How are they going to build up supply chain? All these things. But what I hear you saying is that you're taking industry veterans that have some process know how across-- probably similar ways of thinking as competitors, just you've been in the game for a long time, and then tapping into an existing supply chain. And also, at the end of the day, it's not like you have to go win 50 customers. But there's probably a handful of big customers that would be really make a difference for pico-jewel at the end of the day if they said, yes, we'll take some years. And then-- but the most important point, I think of all of this as the unconstrained gallium arsenide, being able to make 10 wafers with 240,000 on it, whatever you yield, we're talking millions of vixels very quickly. Because I think as we see across all of semiconductors, whether it's memory or CPUs or AI accelerators or whatever, there's obviously there's just so much demand in such constrained supply that truly, obviously, you want to compete on cost and on engineering performance. But I do feel like there is just a little bit of like, if it's good enough, and it's in production, and you can install it into my data center, like, game on. So it feels like you guys have a strategy that will allow you to deliver shipped vixels as soon as you possibly can. Which-- Yeah, so the model has been around for decades and the Silicon, right? I'm in Palo Alto and Silicon Valley. Most companies, most chip companies, designing integrated circuits or CPUs, GPU, they don't have their own foundry. A lot of people use Intel or even AMD uses TSMC, etc. So AMD is this huge chip company. They don't have their own foundries today, right? And so TSMC, Intel, global foundry, I mean, many foundries are the, quote, factory floor, the clean rooms of all these startups. So just because of our small startup size, doesn't mean we can't ship in the millions for a month and compete directly with the very large presence of other optical suppliers and companies, archipelters, et cetera. So that's the beauty, right? So we can stay very lean. Our motto is, stay small. And that has to do with many things. Our name is Peacle Jewel, right? That's low, very, very low power. And that's one of the things that's driving. And then we're also very well experienced small team, but we can have huge benefit by leveraging wind, some I conducted, by working with our supply chain. And they've got the factories, they've got the clean rooms, and then we can ramp very quickly by providing our designs and our unique kind of specialty and then partnering with these large companies that are already shipping. And they would just drop ship, right? So we don't need a big large company in order to ship in the millions per month. That's amazing. Yeah, definitely punching above your weight, that is awesome. So when is your high volume ramp targeted for, if I recall, maybe the press release said something, but-- Yes, press release said we're starting to sample next quarter. And that's in different flavors, right? So we've got customers for the 50G NRZ, 100G LPO, and also the 200G. So all of those will begin to sample in the process of from our device to actual product shipment or revenue for our case is a period what they call qualification, right? Qualification or reliability testing. So everything has to kind of not only meet spec at 0 hour, but it has to be predicted to last 10 years or a number of years, et cetera, through what they call accelerated aging. They test it out at higher temperature, higher bias conditions, and then they kind of estimate back at the normal operating condition, can it last 10 years in the field, et cetera. And that period can, for very small companies, like tier two, kind of customers, be as short as three months, to go through that. But for tier one, because they obviously have much more to lose if they have an issue with their connection, could take more than six months to get those up and running. So ramping at win is available today. However, we have to go through this qualification cycle with our customers before they give the orders. Everything is approved, meets their specifications and qualification, and then we start ramping. So we also said that we'll most likely start ramping in early 2027 next year. - Okay, okay, got it. Thank you for the education here. So yes, sampling, and then the qualification process isn't in ramping. And it's probably, yeah, as you've been saying throughout the whole thing, you're not concerned about the ramping. Obviously, you had to build the product and let customers kick the tires. And once they say, let's go, then it's off to the races. Well, we've covered so much. This has been so amazing. I've learned a lot. I know the listeners will have learned a lot. I guess is there anything else, any last things about Pico Jewel or anything that we didn't talk about that you were hoping to cover? I guess one thing that's very interesting to me, right? As you can see, the image that you're seeing is a very experienced, elderly startup person. And one thing that point out is that very few people went into hardware and photonics in our space, because young people over the last literally when the dot com came out, it was 25 years ago. So those who were coming out of the workforce were more in the application side, what we call the software side, right? So people wanted to go into computer science, et cetera. And so what we found is the aging experience, hardware folks and photonics needs to transfer all of this knowledge. So that's one of my passions is to kind of bringing on the next generation, the generation after that for the photonics, because we don't, you know, just like Vixels and other technology. We see many, many decades. And as I said earlier, no one's saying, oh, way too much bandwidth. So we see bandwidth increasing and bandwidth demand increasing with robotics and autonomous vehicles and what have you. Everything is going to be a bit in connectivity kind of constrained. And so we want to basically spend time to educate, train. And so we're trying to hire, you know, hardware engineers and then train young folks, maybe without the experience to kind of be the VIXIL designers, be the chanceeaver designers of the future. And so that's really exciting for us, right? Because we've got all this knowledge, you know, 10, 20, 30, 40 years. And it's really kind of feels wonderful to have the hardware excitement again, not only in the markets et cetera, but the investment communities, Silicon Valley booming with photonics and hardware. So, you know, we don't take it for granted. It's a great opportunity. And we definitely want to basically take young kind of entrepreneurs, young engineers, folks that are interested in this space, you know, along for the right. And then they take it from there. I love it. I love it. Very inspiring. It's very cool. And, you know, that it's never been a better time, probably, for interconnects and photonics and optics type folks. And I love that, you know, you industry veterans want to bring up the next generation and give them the opportunity to learn from folks like you and to kind of revitalize rebuild, make sure we have like a reinvigorated workforce so that, you know, for my generation and the generation of my children, that they can keep having more and more data moved around faster and faster. That's right. That's right. Yeah. So really enjoy our conversation, awesome. Thank you. Awesome. Cool. Thank you, I appreciate it.