The Six Five On The Road at the launch of Lattice Semiconductor’s new Lattice Avant mid-range FPGA platform. Patrick Moorhead and Daniel Newman sit down with CEO Jim Anderson, SVP of R&D Steve Douglass, and CSMO Esam Elashmawi. Their discussion covers:
- Announcement of mid-range FPGA platform, Lattice Avant
- Lattice’s investment in customer experience
- Lattice’s competitive edge/product differentiation
- Investor outlook after this announcement
Be sure to subscribe to The Six Five Webcast so you never miss an episode.
You can watch the full video here:
You can listen to the conversation here:
Disclaimer: The Six Five On the Road is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded, and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors, and we do not ask that you treat us as such.
Patrick Moorhead: Hi, this is Pat Moorhead, and we are live with the Six Five at the Lattice Semiconductor Avant Launch event in Silicon Valley at the Computer History Museum. It is rocking here. We love good tech and we love even more talking to the most influential people in the industry. Awesome stuff. Dan, how are you?
Daniel Newman: Doing well, always good to be here in the Silicon Valley, especially when we’re talking about silicon. And today was a big day for Lattice Semiconductor. It’s great to have him here on the Six Five on the road. Pat, we’ve got a lineup of the stars here from the company. You want to go and introduce him to the show? Yeah,
Patrick Moorhead: Absolutely. So Jim, Steve, Esam, welcome to the show and congratulations on the big event. Y’all know I love product launches and in my role I get to do about a 150 a year.
Jim Anderson: So do we.
Patrick Moorhead: But congrats.
Jim Anderson: I’m jealous.
Patrick Moorhead: No, it is. Oh, we can sign you up. Just we’ll talk after the show. No, but seriously, congratulations on a big launch. It’s been years in the making and it is fun. I admit as an analyst, seeing the front end and the planning and where you’ve taken the company, Jim and your team, and now we are here today, but I know you like it’s not time to even, let’s pat ourselves on the back for five minutes and move on to make this thing even bigger a success.
Jim Anderson: For sure. We are definitely even working on the next big thing. But yeah, today is definitely a big day for us, a major product announcement for us. And that’s the launch of Lattice Avant. And what that does for the company is it doubles our addressable market. So if you look over the history of the company and Lattice will be 40 years old next year. We’re all a little bit older than 40-
Steve Douglass: Some more than others.
Patrick Moorhead: I don’t know how many in some more than others crowd.
Jim Anderson: But Lattice, if you look back over that 40-year history, this is hands down the biggest product expansion in the company’s history. So we’re super excited about that, but more excited about what that means for our customers because for our customers, what that means is we’re bringing just a tremendous amount of innovation and product differentiation to our customers. Lattice has historically always been really strong in small FPGAs and now we’re taking all of that strength and we’re bringing it to mid-range FPGAs and our customers are super excited about that. Very excited about the competitive differentiation that we’re bringing.
Daniel Newman: We start out the morning, we got a chance to get a little sneak preview. Can’t say too much about that because we didn’t want to get out ahead of everyone, but it was really interesting.
Patrick Moorhead: Well, maybe we did. You asked us not to.
Daniel Newman: We did. We wanted to. We didn’t do it. But it’s always good. I like getting around these kinds of launches because frankly we just need to see innovation continue to be pushed and Lattice over the last several quarters. I think both of us have written our pieces about market outperform, consistently growing, being in the right market spaces. I love that you’ve had such low exposure to the consumer because of course as this kind of market is rotated over, some of those bets you made in communications and in automotive and in the edge in 5G have really started to pay off big. And so congrats on the announcement. By the way, that moment when you were sitting there, I think you may have even had your shoes off, Jim, and you were just meditating, like an hour before the launch.
Jim Anderson: You’re not supposed to say that, Daniel.
Patrick Moorhead: He’s a real people’s person.
Daniel Newman: Like I said to you, I said, you look like you’re in the zone and you’re ready to go. And you were taking that one minute and then you came up on stage, you had that perfect moment, holding the chip. And I even tweeted, I’m like, that’s that CEO moment where-
Jim Anderson: That was fun. Yeah, it’s tiny.
Patrick Moorhead: The challenge though is you have a chip that small.
Jim Anderson: I think I’ve got it here.
Patrick Moorhead: It’s like, get this picture. Well, where is it?
Daniel Newman: I zoomed in.
Jim Anderson: Well then, I got to make sure that I… Yeah, this is the chip that is the platform. But yeah, it’s also small, but I had to make sure I had the writing up on the right direction too.
Daniel Newman: But yeah, you’re like-
Jim Anderson: Yeah, very fun to show that.
Patrick Moorhead: So, Jim, it obviously takes a village and a team and you’re the coach of the team, but there’s a lot of deep tech that goes into that. And I’m curious, Steve, you can double click and listen, I know you love all of your technology babies, but maybe talk about some of the highlights of the innovation, the investments that you’ve made in terms of people and IP to get to Avant.
Steve Douglass: So, Avant has really been an exciting program for us and it really has been a focal point for adding new innovations. And as you mentioned, adding new people and having talent to be able to scale our technology to that next level. When you look at the types of solutions that we are providing and the problems we’re solving with mid-range compared to the small end of the market, there really is a lot more new technologies that you need to have. And it really scales faster when you can have your core team leverage and catapult from bringing in talent that is expert in these other areas.
Steve Douglass: And when you look at the types of innovations our customers needed for their next gen products, it wasn’t just in the hardware, it was in the software as well. And so we had a lot of innovations on both fronts. Our hallmark innovation has always been power efficiency. We’ve always been really good at cramming in a lot of technology and a small footprint with very low power. But it was also very important as we scaled up to the mid-range, we significantly increased the amount of bandwidth that they have to get data in and off the chip and scale the compute capabilities as well. So those were some really key focus areas for us when we were scaling our technology to the next level-
Patrick Moorhead: What did you do on the AI side? I mean was it a 6x increase in performance of AI or maybe it was even more, correct me if I’m wrong.
Jim Anderson: 30x.
Steve Douglass: Well, we have 30x compared to our prior generation.
Patrick Moorhead: Oh my gosh, okay, there we go.
Daniel Newman: So, six times five.
Patrick Moorhead: Exactly. No, thank you very much. No, but how did you implement that? Was this classic new design more LUTs, smaller form factor or was there some secret sauce in there that enabled you to do it? You can’t do 30X with scaling.
Steve Douglass: No, it doesn’t just happen if I go into a new technology.
Patrick Moorhead: So, if you can share a little bit of that secret sauce, it’d be great.
Steve Douglass: So, it was really a blend of factors and first having more logic sales, that’s a big factor. So we’re five times bigger, so that was a decent chunk of it, but the fabrics running 50% faster than the prior generation too. So that gives you another kicker on top of that. And then we further increased and scaled amount of DSP and embedded memory that we have in the core with all that logic. And each of the DSPs can be further decomposed into smaller compute elements. And we have dedicated logic for implementing some of those AI inferencing functions. It’s all about matrix multiply, guys. And if you’ve got the math engine that can compute those things efficiently, you’re going to get a lot more scale and performance out of that.
Patrick Moorhead: Do customers have to toss out their software investment in Nexus to be able to take advantage of what you’re bringing the table?
Steve Douglass: Absolutely not. We are going to have the exact same development tool flows supporting Avant that are already supporting Nexus today. So our Radiant software, our Propel embedded tools, they’re all leveraging Avant architecture.
Daniel Newman: So, I want to spend just a second, maybe Jim, maybe Steve, but software seemed to be a highlight of the show. Obviously when you talk to semis, we talk chips, people tend to always want to drive to the hardware, to the physical, the wafer. And it seems like a lot of your innovations coming out of leading and leading into software, Jim.
Jim Anderson: Yeah, we definitely try to emphasize that today because it’s been a very big part of our strategy. So software for us, the key part of our strategy is about using software to make it very, very easy for customers to adopt our Lattice solutions and get to market quickly because we want customers to be able to easily take advantage of all the great innovation that Steve and the team are putting into the silicon. We want that to be very easy. And so from day one of the avant development, all of our software thinking has been how do we make it really easy for customers to adopt Avant.
Patrick Moorhead: Customers have a lot of choices out there. I mean, they can go all the way from a controller to a GPU to an NPU to an APU, to an ASIC and everything in between. And I know DSP fits in there somewhere, but how do customers decide which direction they go? It is pretty amazing. I look at some of the capabilities that you put up today, particularly in the AI part. I’ve seen at other companies who basically doing different types of chips than you are. Maybe, Esam, we could start with you on that one.
Esam Elashmawi: Yeah. And before that, I just want to comment on what Steve talked about, the innovation from the Lattice development team and just the consistent pace of execution. And when we started this, we spent time with over a hundred customers and you can imagine… I go visit the customers, I come back, Steve will ask me, so what do you learn from the customers? I’m like, Hey Steve, you just got to make it more capable, significantly lower power, much smaller, and it has to leverage all of our tools.
Patrick Moorhead: Did you have to talk-
Steve Douglass: That’s when I said let me talk to the customer, I’ll get the real scoop.
Patrick Moorhead: Did you actually have to talk to customers to get that?
Daniel Newman: Does that feel like the Office Space moment? I take the information from the customer and I bring it down.
Patrick Moorhead: Then why can’t we just directly talk to the customer.
Daniel Newman: The engineers can’t talk to the customer.
Esam Elashmawi: But tremendous work by our development teams to bring out the Avant platform. And to your question about how to decide what to use, there’s a lot of factors, but they’re engineers, they want technology. They want a solution that makes it easier for them to get to market. They want to innovate on their products, they want their end products to be differentiated. And having a flexible programmable FPGA with these innovative capabilities around power efficiency, meeting their performance and application needs, and having that tiny physical size that Jim just showed you that makes a big difference for them. It helps them innovate, get to market faster, and differentiate on their own products.
Patrick Moorhead: It seemed like the industry used to be a lot simpler and in semiconductors it was performance, power, and area. PPA. That was it, baby. But really we’ve gotten to more of the experience, which is how do I make the connection between the customer and even the customer’s customer. And experience no longer is a bag, just a bag of parts, going to suffice. So I’m curious Esam, how are you enabling new experiences? We’ve kind of tiptoed around that a little bit, but I want to hear from the guy who spends a lot of time with customers enabling this and how are you different? I mean at the end, how are you different in enabling their killer experiences?
Esam Elashmawi: And there’s a few things that we do. Number one, customer intimacy is really, really important. But like you talked about you have to understand the customer’s customer as well. And one of the things that we’ve done within Lattice is actually we’ve brought in people that worked from our customers that know the end product. I’ll give you a good example. When you look at our compute market where we do server and client type applications, we actually have folks within our marketing team, we call them marketects that actually came from our end customers. So they know the challenges, they know what they’re trying to do, they can take… And we’ve got all the FPGA experience within Lattice, but that helps close the bridge because now you’ve got open conversations within the company that we can go speak their language. So now we’re using the same terminology as they’re using or that their end customers are using. And that really helps, not just from the customer intimacy and building the trust, but it helps us identify new ways to leverage this flexible programmable FPGA to build those new sets of applications.
Daniel Newman: Just building on that, Esam, one of the things that I know from doing so many events that we always look for are those customers. And you guys kind of broke out customers, IP partners, OEMs, and that slide was impressive. And I said, I wish you’d have shown it generation to generation. Because I think over the last few generations that slide went from very few to very full and getting them to show up and talk though we constantly are talking to customers, let us talk to your customers off the record, let us take them on the record. And that’s hard. You guys had a nice showing of customers today. What do you attribute the fact that they’re so willing to step up and be vocally supported?
Esam Elashmawi: And as you noticed as well, those customers… There were more videos, we just didn’t want to show them all because at length of time. But if you noticed the titles of those individuals, they were all decision makers and engineers and there are people really in the development cycle themselves. But it really goes back to again, the customer intimacy. The need to have and work with an FPGA supplier that’s willing to build a roadmap aligned with their needs. And that customer intimacy and us being able to work closer with the customers, I think that’s a testament of how we’ve made it easier for them to adopt our solutions, both hardware and software as well. And kudos to the sales team. I mean the sales team does an outstanding job in reaching out and building that intimacy.
Patrick Moorhead: Would you say customer proximity is your biggest differentiator as a company or is it the core tech? I’m going to make you decide here, Jim, is it…
Esam Elashmawi: We always pass on the tough questions.
Patrick Moorhead: Well, and I’m pretty sure I know the answer is going to be it’s amalgamation of all. But a lot of questions I get, Jim, from your customers or ecosystem and even investors are, why are you different? How are you able to do this? Because it’s not that often you see a lower left, up to the right on almost every metric, including things like, it means a lot to me, product velocity. It’s one thing to… Listen, and the financials are important, but there’s also partners, there’s stickiness, there’s them coming back to you for more and doing more with you. There’s the basket across product. There’s so many metrics and all those look really good to me.
Jim Anderson: I think of course you have to have a great product roadmap and that’s certainly what we’ve been building. But I think culture is important to customers as well. I’ve had a lot of customers talk to us, talk to me about, hey, the culture of Lattice, when we interact with Lattice, it’s different and we appreciate that.
Patrick Moorhead: How so?
Jim Anderson: From when I first joined Lattice, I noticed this, it’s a very collaborative, customer focused culture. Every Lattice employee loves supporting and interacting with customers. And when I first joined the company, I said, hey, I think this is a real fundamental differentiator for the company. And I think that that’s not the case across all companies, but I think that that can be a real differentiator for us. And it’s a basic DNA that’s within Lattice, which I think has been cultivated over our 40-year history. And so when I joined, I really wanted to encourage that. Like Let’s absolutely be even better about that. Let’s do even more around that. And so I think yes, of course we have to build great products for our customers and we have to innovate for them. But I think taking it to the next level with a customer is having a culture that really embraces the customer and wants to bring them in and form a relationship that’s not just about one design win or two design win, but it’s a multi-generational-
Patrick Moorhead: It’s more strategic.
Jim Anderson: Absolutely. Okay. Absolutely.
Daniel Newman: Okay. So Jim, the ground truth as Pat and I always like to say when we’re on our earnings podcast is always what you guys come out with when you share results. But the investor that are here, several were here, saw their badges walking around, and so while it wasn’t your investor day, that’ll be later, I believe in ’23. The investors tend to be the, what you say, the arbiter of how good a product is. And they basically… It’s how the company performs in the market over a long time. We’ll just say that. Maybe we’ll agree, maybe we won’t. But it tends to be companies that perform well.
Patrick Moorhead: If anybody’s listens to the podcast, Daniel and I look at it a little bit differently, but-
Daniel Newman: We like to talk about it. We call it the ground truth. That we really do. So you know said you’re going to double the SAM. That’s obviously in itself a really big opportunity for investors. But how do you think the investor community should be looking at this announcement? Is there anything more to it than just that double the SAM opportunity?
Jim Anderson: I think, it’s for investors, the really simple way to look at it, which I think is the right way to look at it, is the products that we announced today, the platform that we announced today creates a totally new revenue stream for the company. A totally greenfield brand new revenue stream that is completely additive to the company’s existing revenue stream. We’re certainly focused on continuing to grow in the part of the market that Lattice has always been traditionally strong in, which is small FPGAs. So we’re certainly focused on continuing to grow there, but Avant as it enters production and ramps with our customers, that creates a completely additive revenue stream that we believe will accelerate revenue growth out in time for the company. And as an investor, that’s kind of music to my ears. Additive greenfield revenue growth on top of the existing business. And it’s a very straightforward extension of our existing business. When we look at the target customers of Avant, 90% of those target customers are already customers today. So it’s a natural extension of our existing business. I think it’s about revenue acceleration. That’s what’s exciting for the investors.
Daniel Newman: Do you have a sense of revenue expansion? You said 90% of them are out there. Do you have a sense of how quickly they’ll adopt? I mean, I saw the chart, so a lot of them are saying yes.
Jim Anderson: Definitely, but we got to leave a little bit for the investor day next year. Daniel, come on. I mean-
Daniel Newman: Leading the witness, sir.
Patrick Moorhead: Jim, I think final question on the investor part. Is there any play between kind of lowest power and the mid-range, meaning if you’re in mid-range and it’s good experience, does it help you in the current business at all?
Jim Anderson: Definitely. I think when you look at, especially industrial and automotive customers, but I think this is also true with communications customers as well. But the majority of the types of FPGAs that an industrial and automotive customer uses is actually mostly mid-range and small FPGAs. That’s the majority of what they use. And so now with this product line expansion, Lattice is able to service almost all of the FPGA needs of the vast majority of our customer base. And so that certainly helps us.
Patrick Moorhead: Okay.
Daniel Newman: Well, Jim, Steve, Esam, I want to thank you all for joining us here on the Six Five big day for you all at the Lattice Avant launch. And we really appreciate y’all joining us here and we look forward to having you on our show again sometime soon.
Jim Anderson: All right. Great.
Steve Douglass: Thanks for having us, guys.
Esam Elashmawi: Appreciate it.
Daniel Newman: All right everybody, thanks for tuning in here to the Six Five on the road at the Lattice Avant launch, at the Computer History Museum, in Mountain View, in the Silicon Valley. For Patrick and myself, hit that subscribe button if you enjoyed what you heard. We would love to have you for all our episodes, but we got to go. So we’ll see you later. Bye-bye now.