#129 – An Interview with Brett Fox and Dr Jeroen Fonderie - Device Doubling Decretum

Chris Gammell: This is the Amp Hour Podcast, recorded January 21st, 2013. Episode 129, with guests Brett Fox and Jerome Fondry. Device doubling accretum.

Dave Jones: Welcome to the Amp Hour. I'm Dave Jones from the EEV blog.

Chris Gammell: And I'm Chris Gammell of Chris Gammell's Analog Life.

Brett Fox: And I'm Brett Fox, CEO of Touchstone Semiconductor. And Jerome Fondry.

Chris Gammell: Hello guys. Welcome to the Amp Hour.

Dave Jones: Is this our first dual guest episode?

Chris Gammell: Co-located, yeah. I think so. First dual guest episode.

Dave Jones: There you go. Thank you very much for joining us, guys. It's not often we get a CEO and a VP of engineering on the show. So thanks for giving up your valuable time. I'm sure it takes a lot of effort to run a semiconductor company.

Brett Fox: Yeah, we're having a lot of fun with it and we're happy to be here.

Dave Jones: Awesome. And we've got a whole bunch of viewer questions, audience questions, which we'll get on to. But yeah, tell us about Touchstone Semiconductor. How did it start? Where did you guys come from? What spurred you to go off and run a rebel group and form your own semiconductor company?

Brett Fox: Well, we all came from Maxim, actually. Most of the founders and myself came from Maxim or Linear Technology. And the idea actually is one that started with me and had been in my head for maybe 10, 12 years, since about 2000 when I left Maxim. And the timing wasn't right then to start a company like this. But eventually, later on in the decade, it became more and more right because the availability of anything was starting to open up. And, you know, just for virtual things, plus some things that were going on in my life personally, where I was afforded the ability and kind of the umbrella to start a company like this because I got involved with some venture capital people who wanted to start a company like this. So therefore, I had the chance to get going on this. And one thing led to another. And your own and I and another individual who was right next door to us, Adolfo Garcia, starting touchdown.

Dave Jones: Right. So you said that sort of almost like the venture capital company wanted you to start. So they were fishing, were they? They were fishing for somebody to take their money and form a semiconductor company.

Brett Fox: They don't like you taking their money. If you raise money, we could spend a day on just raising money. But what they do in some cases is they bring on somebody like myself as what's called an entrepreneur in residence. And what they do is they'll pay you and you'll get an office. You'll get all their contacts inside of venture capital and out with the concept being that you're going to start a company that's interesting to them. They may or may not fund it at the end of the day. And in essence, they were interested in doing an analog company a la Maxim. And I happen to know the managing partner of this venture capital firm. So when I left my previous employer, Mike Krell, and I was looking around for something to do, I looked him up and said, hey, I'm available. And he used to come up here and spend a day with us. One thing led to another. And I ended up being an EIR with them. And then that led directly to Touchdown.

Dave Jones: What is that acronym?

Brett Fox: EIR is Entrepreneur in Residence.

Dave Jones: Oh, yeah, Entrepreneur in Residence. Right. This is interesting. This is the first time I've sort of heard of this because we've had others on the show before that have raised venture capital. And it's like extracting teeth. They do it the other way around. They have their great idea. They're so passionate about it. And they try and, you know, they visit, you know, hundreds of different venture capital firms and give their pitch and get rejected by everyone. Yeah. Yeah, exactly.

Brett Fox: So that's a different experience. Well, we ended up having to do – well, we had both experiences. So we had the EIR experience, which started us. And it turned out the VC firm I was working with didn't end up funding us. Oh, wow. And that ends up actually being – yeah. So it ends up actually being a negative because of various things. Because it's like, why didn't they fund you? And there are many different reasons that VC companies don't fund people. And it turned out in this particular instance that had nothing to do with us. But it ends up being kind of a black mark. So the other war stories that you just heard about, you know, hundreds of VC firms and everything else, we went through all that. We were raising our money in early 2008. And if you remember the economic climate in 2008 with Lehman Brothers and all those things.

Dave Jones: Oh, no.

Brett Fox: That directly went into what happened with us. So we ended up finding our first investor, Gil Kogan of Opus Capital, within a month of starting to raise funding. And we're thinking, great. We've got one. And usually you find one. You find the second one, the snap of the fingers. Right. And it didn't work out that way for us. And you need to because we were raising – at the time, we were trying to raise $11 million. And we ended up raising $12. And as the year went on, it got worse. So we had to retrench and essentially keep funding the company out of our own pockets. And then in 2009, we started raising money again. And then late in 2009, we found Coastal Ventures with Pierre LeMond, who's kind of a living legend here in the valley. And Pierre ended up being the second part of our syndicate. And then in early 2010, the money went in the bank, and we were up and running. Wow.

Dave Jones: So it's almost as if you end up – you either go one way or the other. You end up with like a black mark against your name, and you're just poisoned. The word gets around to every venture capital firm pretty quick. Or once you get somebody interested, then you get a good word passed around all the venture capital funding. And then everyone wants to jump in. Is that accurate?

Brett Fox: Kind of, sort of. But you have to remember, there's another thing going on. And I don't know whether you guys have touched on it in previous shows, but semiconductor investing is not in vogue. Right. It's something that's been going –

Dave Jones: Why is that? It's expensive, man.

Brett Fox: Well, I have my own theory on it. And my own theory is if you go back to the mid-'90s when there were a lot of companies that were kind of one product wonder companies where essentially they were funded. And for a minimal amount of money, the VCs were able to flip the companies and sell them for hundreds of millions of dollars. And these companies were no more than a test chip or a chip that just wiggled. And that model perpetuated beyond 2000.

Chris Gammell: Ah. Wow.

Brett Fox: And after 2000, it started not working anymore. So there's an old thing about throw the baby out with the bathwater. And that's kind of what happened with semiconductor. So if you look at the two guys that fund us, Gil Kogan was an original investor in Maxim. So when we walked into Gil's office – Gil and I had a mutual friend in common that introduced us. When we walked into his office, halfway through our initial pitch, he looks at me and he says, you know, I was an original investor in Maxim. Wow. And boom. You know, it was three meetings later and we're done. You know, so it's very easy. It's kind of a love at first sight thing with VCs. So he fell in love with us. And Pierre, if you look at Pierre's background, you know, Pierre was a co-founder of National Semiconductor, was involved in the funding of linear technology, was chairman of the board of Cypress, Microchip, Vitesse, many other companies. So unbelievable background. So when we walked into Pierre's office and met with him, it was three meetings and we were done. All it took was a year and a half in between and, you know. Right.

Dave Jones: But still, that was it. Eating out of pizza boxes. Yeah, exactly. Right. Exactly.

Brett Fox: So when you find the right guy, it's absolutely love at first sight. And it's very simple. And they fall in love. And all the things you hear about, vigorous they are and everything else fall out of the window because they love you. And they say, wow, this is great. And they ignore all the other things that possibly can go wrong because all of these guys know that many, many things can go wrong in starting a company.

Chris Gammell: Right. And usually with venture, it's often, you know, it's based on the people, right? I mean, you mentioned that you were the entrepreneur in residence. That's often, that's the same kind of model that, like on the software side that Techstars does or Y Combinator where they choose the people instead of the final idea because they trust that the final idea will be good because of the person or people in the case of groups.

Brett Fox: Not being a VC, there are different funding models that they use. One is the one that you mentioned, Chris. The other one is to look at specific areas of interest and fund various companies that are going after that specific area of interest. So they have thought processes that they use. It's a fascinating area. It really is.

Chris Gammell: Was there consideration in the fact that it was fabless? I mean, maybe our listeners don't know with Touchstone that you guys are a fabless semiconductor design company, correct?

Brett Fox: Yes, that's correct. It's impossible, by the way, to start a fabbed semiconductor company today.

Dave Jones: I was about to say, yeah, you couldn't, could you? No, you can't. It would be – Okay. In fact – It would cost $2 billion right up front. You could start very, very tiny.

Chris Gammell: Yeah. You could get like a secondhand Asher and Etcher and, you know – Right. So you could hire Jerry Ellsworth and she could be there. Yeah, exactly.

Brett Fox: And at the end of that day, you'd still be $300 million in the hole or whatever it is. That's true. Yeah. It won't work. So the only way to go is fabulous. And one of the things that really changed since I originally had the idea and today is really the emergence of world-class analog technology with companies like TSMC. And that's a game changer. When you look at – and Yerone, maybe you can talk more about this. But when you look at what we can do with foundries versus what a Maxim can do or a linear technology can do internally, almost all cases, there's no difference or little difference.

SPEAKER_00: Yeah. Foundries have come a long way. I mean they'd be just purely for digital design. But those days are long gone. They have a very wide variety of process options to choose from both in geometry as well as all the process features at each of the technology node. And they even do SIGI and, you know, the most obscure things that you could think of. Yeah. They have technologies to support it. And analog is a big part of that. Started with power.

Dave Jones: And what was the reason for that change?

SPEAKER_00: Well, because they think and found that they can make money there, right? It's very obvious. That's why they do it. Make money, yes.

Dave Jones: Duh. Oh, I thought it was some marvelous advance in technology or something. No, no. They just decided, no, there's money. Yeah, yeah. I can smell it. No.

SPEAKER_00: For them, the beauty of analog is that analog is behind the curve for the migration to tighter and tighter nodes, right? So we don't need to be in 22 nanometers for most of our analog functions. And it's a great way for them to keep utilizing their fabs, the ones that run equipment from five years ago. So it's an interesting preposition for them.

Dave Jones: So it was basically they decided, well, look, all the processors are all moving to these new whiz-bang, you know, ultra-small, you know, geometries. So we've got all this equipment sitting around. What can we do with it? And they decided, well, analog can still use those. So they pushed it.

SPEAKER_00: I'm sure that's part of it. And so that's how they started. And they started promoting it. And then many of the companies that would be their customers looked at it and said, well, that's enough. But you got there. Because this is, let's say, 10 years ago. And there were no good linear capacitors. There was no support for bipolar transistors. There was no matching data. All the things that analog design engineers really rely on was all missing. But they put that in place over the years. And now they're excellent. And no thin films as well.

Dave Jones: Have the CAD tools changed at all? Have they improved? Yes. Out of sight. Is that another reason why that you can, you know, easily start up a, well, I'm saying easily in quote marks, you know?

SPEAKER_00: That's definitely part of it. The part of the support that a foundry needs to give you is to have a PDK, a process design kit, of that technology that you can work with and that has all these analog features. And they do.

Dave Jones: Ah, so it's almost building block type. You can just, right. Would you only use those building blocks they supply? Or would you go to them and say, hey, guys, we've got this weird idea for a chip. We need you to, we need to figure out how to make this work.

SPEAKER_00: Depending on your size, they may do that. If they have to put in their R&D hours, then they're going to, of course, look at it. Of course, yeah.

Chris Gammell: They charge you for it.

SPEAKER_00: But if you can do that yourself, they may be very well willing to keep supporting it. And there's actually a bunch of companies out there that use that model.

Chris Gammell: Yeah.

Dave Jones: So would you say that the majority of your chips use just like off-the-shelf blocks from these, you know, starter kits, so to speak?

Brett Fox: I would say off-the-shelf transistors. It's not like we can take an op-amp from here.

SPEAKER_00: No, yeah.

Dave Jones: Oh, no. Okay. Yeah, of course.

SPEAKER_00: No, but so far, I mean, we're small. So we don't have a big R&D arm to support the device development, as in go figure out a new transistor type or anything, right? So we're using the basic building blocks that they provide. But then our skill sets, our added value is the circuit design that we do with those problems.

Brett Fox: Even though we did have one design where one of our older design engineers actually really wanted to change the shape and size of various transistors to get what he wanted, which was kind of interesting. It worked. It worked for a silicon, so I give him credit.

Chris Gammell: All right. Those old engineers will get you every time. Yeah, exactly. Yeah, the graybeards.

Brett Fox: You know what, Chris? Those older engineers are worth weight in gold. They're a dying breed, sadly. Oh, yeah. But there are people out there, if you go back to the guys who started out in the late 60s, early 70s, who really understand the device physics in a way that the younger generation of designers doesn't. And they're used to really working directly with the fab and the foundry. And they are absolutely priceless to have around. I can't state that strongly enough.

SPEAKER_00: Yeah, certainly true.

Chris Gammell: It's like having the old guys that were writing machine code and assembly for microcontrollers, right?

Dave Jones: So, Brett, what was the reason you wanted to start it? You said that you had the idea back in, what, the 90s? Late 90s, early 2000. Why were you unsatisfied with what Maxim were doing? Something that they were doing wrong? You thought you could do better?

Brett Fox: You know, I don't know whether Maxim would. Well, you obviously think you can do something better, you know, or else you don't do it. But, you know, it was at a time, personally, where I was seeing the end of the line for me in terms of what I could achieve at Maxim. I'd done a lot. It was a great company back in the day. And I have nothing but good things to say about my experience there. But, you know, for me, it just kind of felt like, okay, I have a choice to make. Like, I can either stay there and keep doing what I'm doing or try and do something different. And this is really what I wanted to do. And I kind of saw a possibility to get there.

Dave Jones: So, what's your take on the whole Maxim? I mean, Maxim had a reputation probably back as, yeah, probably in the 90s of producing, what, 600 new chips a year or something ridiculous like that. And you could never get them. You know, they were just, you know, hopeless. They got such a bad reputation that they had to spend a long time trying to fix that. What's your experience with that?

Brett Fox: Well, I've lived both sides of it. I remember being there and dealing with it. And it's horrible when you're running a business and you can't supply your customers. You know, it's not a fun thing to deal with at all. And then on the other side of it, competing against them, it's interesting the angst that's built up in the customer community. If you look at part of the reason why I know there's a question about, you know, second sourcing and why did we do second sourcing. You know, part of the reason we did second sourcing is because it serves a need out there. And customers are unhappy with Maxim. They're happy with the fact they can't get their product. And we're providing an alternative and a service. Sure. You know, to customers by doing that.

Dave Jones: So what was the cause of Maxim's demise in that area in terms of getting that bad rep? Were they just biting off more than they could chew? Did they not have enough fab capacity? Too many, just, you know, designing way too many parts?

Brett Fox: And what was this? It was more of a philosophical thing. The philosophy of the company was more around the lines of, you know what, we don't believe in any customer's forecast. Therefore, if you're showing upside, we're not going to build it until we see the whites of their eyes. That's the best way I can describe it.

Dave Jones: So they build it to order, basically. That was their philosophy.

Brett Fox: Not even build it to order. It was more until it's on the books and until you desperately need it. We believe we can. There was an old saying at Maxim about robbing Peter to pay Paul. And that was kind of part of what people did there, you know, to kind of keep things going. And I think they've gotten somewhat better. I don't know whether they've solved all their problems. But I think they've at least tried to address them would probably be the best way I could say. Not that I want to be a balance piece for the competition. No, yeah, exactly.

Dave Jones: No, yeah, I remember about five years ago they tried to address it seriously. They tried, you know, the CEO came out and tried to fix things. And, yeah, I don't think that reputation's gone away. No. You know, because there's a lot of engineers out there who simply will not touch Maxim still because they've been burned too many times. You're designing their part and then you find you can't get it, you know, and you're screwed. And that makes you look bad as a design engineer. So, oh, nasty stuff. So how does a company like that produce, like, in the order of, like, 600 new chips a year? Well, it's the same.

Brett Fox: You know, we're doing, you know, essentially it's the same thing we're doing in a smaller scale. You hire good designers. You have metrics and goals. And the thing about our business, which is kind of interesting, is it's typically one engineer development effort. It's very different than digital, where a digital design may be 20, 30, 40, 50, 100 people working on it all at once or more. You know, our stuff is typically one engineer per development. And from that, from one, you get many. So from one development effort, we may develop five, six, seven parts from that one development effort.

Dave Jones: What sort of variation can you get on, say, if you design a basic op amp, what variations can you do in terms of, can you do, like, different power grades or what?

SPEAKER_00: Yeah, yeah. So are you talking about off one design or are you talking about what you have on design? Off one single design, yeah.

Dave Jones: As Brett just said, yeah, you put the effort into producing one design and then you can get half a dozen variations of it. What sort of variations can you get?

SPEAKER_00: Well, with amplifiers, it's one of the things that you mentioned is definitely there. You can vary the amount of power that you've been in. And with that, the speed characteristics of the amplifier will change. You can change component methods that are used, right? You can make it unity gain stable or 10x gain stable. You can, of course, do things like put duals and quads on the die. But that's typically a different, the same design effort, even though it's a different layout, right? But it's still, you leverage the design effort that's been put in already. And that's amplifiers. And amplifiers are probably not even the best example for doing this. There's other things where you can go in many different ways.

Dave Jones: So what is your biggest product? What is your biggest area at the moment? Is it amplifiers? Is it references? Or what is it?

SPEAKER_00: Our most successful are the ones where we most put most of our effort.

Dave Jones: Yeah, your most successful products. I would. Have you found it? Have you found it's amplifiers? People desperately need, you know, second source maximum amplifiers?

Brett Fox: Well, I would say at a high level, we're seeing interest across the board for what we're doing. Everything we have is selling. Everything we have has customers' interest. There's no question about it. What we have seen, which has been a real positive surprise for us, is our assumption at the beginning was it was going to be very difficult for us to sell our proprietary parts at the beginning. That was another reason we started off doing second sources. But much to our surprise, we've seen a very willing audience for everything we've done on the proprietary side right from the beginning, which is really, really nice to see. We were expecting it.

Dave Jones: Well, I wouldn't have expected it either, because as a design engineer, I'm going to, well, who's touchstone, you know? Oh, I don't know. I haven't used their parts before. Are they going to be around in five years' time? You know, if I design their part in and it is a custom part, I can't second source that anywhere else. Well, that's a big risk on my part as a design engineer. So generally, I would only do that if, you know, I had a very niche need and your chip absolutely nailed it, you know, and there was nothing else, you know, on the market to do it. So, yeah, I find that quite surprising, as you guys did. So you seem to be building up more of a rep in that area of, you know, you can actually supply them in volume and, you know, people are getting a good vibe that you're still going to be around in five and ten years' time.

Brett Fox: Absolutely. Absolutely. In fact, the other thing that I was expecting, which didn't occur, was I was assuming at the beginning I would have to personally close a lot of the business, especially the big orders that we got, and I haven't had to close any of them, which is, again, fantastic, which means we can leverage, which is what you need to scale and grow a company.

Dave Jones: Right.

Brett Fox: So how many different parts have you got? We have now 62 parts in the marketplace. Today. 62 today.

Dave Jones: Yep.

Brett Fox: And we'll… Right.

Dave Jones: And that's after, what, two years in business?

Brett Fox: Yeah, two years in business. And by the end of this calendar year, we'll have well over 100 parts in the marketplace. So, yeah, we're growing by leaps and bounds.

Dave Jones: I just find that amazing that, you know, you can put one engineer on a, you know, and on a new chip and out pops a new chip and you can produce 50 or 100 in a couple of years. That's just… Well, there's more than one engineer behind us. It's just amazing. I thought, you know…

SPEAKER_00: There's more than one. Oh, wow. They're working hard.

Dave Jones: Oh, I'm sure you've got more than one engineer, but yeah.

Brett Fox: Yeah, we're very fortunate in that the team we have, I would say, pound for pound, if I was to use a boxing analogy, pound for pound, I think we have the best team in our industry. And I've been blessed to work with some really good people. And this team is the best I've ever worked with, bar none.

Dave Jones: Have you found it's hard to get new blood out there? Is there enough engineering talent coming out of the various units and things like that? I was going to take a stab at it.

SPEAKER_00: You can… Yeah, so I was going to say that, first of all, it's always… it's hard. It always has been hard. And it's supposed to be hard to some degree because we're looking for the top of the line people. As far as universities go, it's a little bit the same. I mean, there certainly has been a shift over the years from what was just electronics to much more digital design. And in that process, the analog side has been forgotten in some places. But there's still a whole range of universities where there's excellent analog schooling going on. They're both inside the U.S. as well as in Europe and outside of Europe and in Asia as well. So there's still plenty of people coming out. Now, the typical person that we would be interested in in this phase of our existence has already a couple of years of experience behind them. Because we need people, just like Brad said earlier, that run their own project, that run their own design, and that have done that. If you've looked at, I don't know, the folded cascode amplifier in a course of analog design in school, that's great, but that's only a start. We need people that have done that in some form of industrial environment a couple of years. That's where we're looking at now. But, yeah, there's still people out there.

Dave Jones: So it would take a graduate who's at least done some semiconductor fab theory and design stuff in their course. So it would take them at least a couple of three, four years on the job before they could sort of start actually designing their own chip and taking on the entire project?

SPEAKER_00: Yeah, that's correct. Before that, you can do blocks, and eventually, also, we're going to be looking for people like that a couple of years down the line. But at this point, we need people that are a little bit further in their career already.

Brett Fox: Yeah, and then just to add a little bit more color to that, the typical engineer for us has about 20 years of industry experience and roughly about 10 paths to a designer. So it's a pretty experienced group that we've got.

Dave Jones: Okay.

Brett Fox: It's not your typical 20-something startup type company. You know, it's a little different.

Dave Jones: Yeah, of course. Yeah, yeah, we've got a big more degree. So do you think it's an area that – yeah. Do you think it's an area that students should get into? Do you actually recommend – I do. Is there always going to be a future there? Is there a high-paying future guarantee?

Brett Fox: I think there absolutely is because one of the other fundamental things about analog is there are more opportunities than there are designers to solve all the opportunities. And regardless of whether it's IC design, level design, it doesn't matter. And analog is – we live in an analog world, right? Analog is not going away. So I think it's a – you know, if you want to be an engineer – Yeah, of course. If you learn analog, you can learn anything else. It's a fantastic area to go into. It's just fun, right?

SPEAKER_00: Right. I agree. No, but it's – without a doubt, I agree 100%. If you look over the last couple of years, during all these economic troubles that we've had, situations were harder for analog design engineers as well. But the employment situation was never for it, I would say. There were always – for good people, there have always been openings. So, yeah, I would recommend it.

Dave Jones: And you can earn the big bucks, I'm assuming?

Brett Fox: Yeah.

Dave Jones: Yeah.

Brett Fox: There's big money in it? Yeah? Yeah, there's nice money in it. I mean, but that's not the right reason to do it.

Dave Jones: You can always name your price, couldn't you? Well, yeah, of course. But, you know, I mean, these are young kids we're talking about here, right? They all be thinking about money sometimes.

Brett Fox: I hope not.

Dave Jones: Do you find that it's the passionate ones who are the good designers? Like, you know, the ones that have an interest in it outside of work? Or do you find that you can actually get those who just don't give a toss about this outside of work and that's just their nine-to-five-day job?

SPEAKER_00: Well, I'm sure the latter. I'm sure they exist. But the first one is a much bigger group. Many, many of the people that we have here are the type that you described for us that started off as a kid and were hobbying around and have just a passion for analog electronics. And, you know, that from one thing came another and somehow they got in school and still were doing the analog stuff and then got a job in there as well. And they just go for it. Their passion is to make it work and then to see it sell and end up in somewhere where somebody else uses it. That's what they go for.

Dave Jones: Do you think that you'll always be an analog company as such? Is there any, you know, oh, God, I wish we could do this chip, but it needs like a DSP core or something like that?

Brett Fox: Always being an analog company. I think if you look at kind of one way to look at us, even though we're not really competing head-on-head with them, is linear technology. Certainly would be a grown-up version of what we may become. You know, we'll see where we go. But more like them than we're the turn that Maxim has taken, I would say.

Dave Jones: By looking at your product portfolio here, you seem to sort of, you know, you're doing amplifiers, ADCs, low-power, you know, stuff, all that sort of thing. Is that your area of expertise? I mean, I don't see any RF stuff, for example.

SPEAKER_00: No, that's correct. But that's not because we do not have the expertise. We actually do have people that are very skilled RF designers as well. So it's just that we found that there is a niche in this very low-power area that we've had a lot of product ideas that we want to work on first. And, you know, at some point, I'm sure we're going to go into different directions as well. It's not a dogmatic choice that we only do in low-power, not at all. But there's things that need to be done that we think we're going to do.

Dave Jones: How do you come up with new or what chip you're going to work on next? Do you research the market and go, oh, that one's, you know, maximum aren't supplying that one, but we think there's a big need for it? And do you target it like that? Or do you just go, hey, we think this one's cool. Let's do a new, you know, let's do the world's lowest power, you know, voltage reference or something like that?

Brett Fox: Yes and yes. So we definitely are talking to customers, understanding what's going on. And we're definitely, we're creative guys too. So we have ideas and we have regular sessions with the designers and our apps guys to talk about what we think are the right things for us to go after. And those meetings are very dynamic. And from that, we kind of distill out the best ideas. Right now, we have about 10 times. Dynamic? Yeah. We have about 10 to, yeah, very dynamic actually. And it can get pretty fun in those meetings actually.

Dave Jones: As in lots of arguing.

Brett Fox: Oh, yeah. It's my favorite of all the meetings we have here. That's my favorite one. Because when you see people come up with really good ideas, ignoring the arguing, which also can be fun, by the way. Yes, it can. That's when you get, you can get sued by what's going on. So we have about 10x the ideas that we have designers in the company to solve them right now. So we're not idea limited at all, as it should be.

Chris Gammell: I bet. So you guys play in the low power space. I mean, you're not actually in any of the DC to DC kind of stuff that seems very unvogue because of, you know, just the need for LED lighting and everything else like that. Why aren't you in that space at all?

Brett Fox: Well, let's see what comes. Wait and see. We might see some DC to DC converters coming from Touchdown.

Dave Jones: Yeah. Whoa, whoa. Interesting. I would have thought that market would be absolutely flooded. I mean, there's so many DC to DC converter options. But as always, I can never find the ideal one. You're right, Dave. I'm doing my parametric search on DigiKey. So you know what, Dave? You're 100% right.

Brett Fox: That's crazy. Part of the reason we didn't go after DC to DC at the beginning is because so many people playing in that sandbox. So it's a big sandbox, but you have to kind of figure out where you want to play. And the bottom line is, for us, is if we can add value, we'll do it. If we can't add value, then we're going to move on to the next thing. So we found areas where we believe we've got some groundbreaking stuff where we can add value in that space. And that's what we're going after.

Dave Jones: You guys have just signed up with DigiKey, I believe, to distribute your parts.

Dave Jones: Is that an easy thing to do? Do you just approach them and go, hey, do you want to sell our parts? And they go, oh, we've already got 20 million manufacturers. No. Or is that their philosophy to absolutely have every single manufacturer?

Brett Fox: I think their philosophy is they want to have every single manufacturer that's selling. And they have to believe that you're going to be a viable player for them to do business with you. So that was a several-month courtship that went on between us and them to make that happen. It wasn't like send something into their website and then boom, you have a list and you sign up. That took a bit to make happen. But we love those guys and we love their model. And we know that almost from day one when we started selling our first part, customers were asking us, are you on DigiKey's line card? And it is absolutely –

Dave Jones: I'm surprised everyone would be asking you that. Yeah.

Brett Fox: And everybody was.

Dave Jones: If you're not on DigiKey, you don't exist.

Brett Fox: That's a very good statement. So now we're on DigiKey, so now I guess we exist.

Dave Jones: Yay.

Brett Fox: Yeah, exactly.

Dave Jones: Yay. I can find your parts now because as a design engineer, that's my first port of core, DigiKey or Mouser. Usually DigiKey because it works a bit better and they have a bit bigger range. But yeah, that's how it works. That's how you find your chips. And then if you get desperate and you can't find it there, well, then you might start searching out the individual manufacturers. But 95% of the time, bang, I'm going to find a suitable part on DigiKey, parametric search every time. So once you're in with them, do you get all your products automatically added or do you have to convince them for each new one?

Brett Fox: No, they're all on there. All 62 of them are on their website, all in stock. As I said, they're fabulous to work with and really happy with the relationship.

Dave Jones: So would they normally want to buy like a reel or two worth of each part?

Brett Fox: Yeah, absolutely. So they want to make sure because their whole shtick is, you know, they want to make sure that they can get parts from them to you anywhere in the world in 48 hours, I think, is their goal. So they don't want to have to call us and say, gee, we need a reel of this. And it defeats their whole model, I think, to do that.

Dave Jones: Of course. So are you on Mouser, others, or are you just targeting DigiKey at the moment?

Brett Fox: Well, we have DigiKey. In terms of the catalog, guys, it is only DigiKey, you know, at this point. And we don't have plans right now to add anybody else but DigiKey. We also have Future as a worldwide distributor. And then in Asia, we have World Peace, WPG, which is the largest distributor in Asia. They just took us on. And if you go to their website. Never heard of them. Yeah, World Peace is actually – they're called World Peace or WPG Group. They are the largest distributor in Asia. So that's a real sign of confidence that they've decided to do business with us as well.

Dave Jones: Are you allowed to say where your main market is? Is it Asia or are you selling mostly into Asia or are you selling into mostly the U.S. and European markets?

Brett Fox: Mostly at this point because of our distribution presence and our sales presence has been really North American-centric at this point. So it's primarily been North America, some Europe, and now we're starting to see growth in Asia as well.

Dave Jones: We've got a whole bunch of questions. Should we get on to them, Chris, unless you've got more?

Chris Gammell: I had one – because of your rollout schedule, and I don't know if any of the listeners asked this. How long is your qualification process? I mean I know that – so you said 62 parts. A lot of those seem like they're different flavors of a base-to-base design and then different voltages for references and stuff like that. But how long does that qualification take? Because I've never really done any of the chip design side of things. How fast can you roll stuff out?

SPEAKER_00: Yeah, once you have silicon out, there's a bunch of things that need to happen, right? Design engineering needs to take a look at it in the lab and see if it wiggles, if it does what they expect to do. And let's assume that it does. Then in parallel, test engineering is working on creating a program to do the production testing, which is a different world, of course. They need to get their things done. And then parts that have been tested with a finalized test program would need to go into some qualification. But that qualification doesn't last that – it can take up to a month, right? We do what most companies, most IC manufacturers do. We do a thousand-hour burn-in on our parts. And that's the longest pole in the tent. That's the thing that takes the longest. And besides that, there's things like easy testing, chip testing, and a bunch of other stuff. But the operational life test is the one that takes the longest. But in the grand scheme of things, it's almost never the limiting factor. It's getting all of our work done on time and get everything lined up for a proper product introduction and announcement to the outside world that takes most time.

Chris Gammell: Yeah, I'm sure it's a symphony of –

Dave Jones: I could imagine that you would have more people working on that side of things than actually designing the chips. You'd have more people in the test and qualification and production engineering side, no?

SPEAKER_00: No, that's not the case, no. Certainly right now, our company is very heavily – it's leaning towards design. That's where the – as it should be in this phase, right? We need to get products out. So that's where we are staffed the strongest.

Dave Jones: Where do you draw the line on what characterization testing you do? Because I'm sure there are some, like, really anal customers who want a curve for – a characteristic curve for absolutely everything. They want drift versus temperature versus voltage versus, you know, does somebody fart on the other side of the room?

Brett Fox: That's a weird chode if you guys ever –

SPEAKER_00: Yeah, well, I remember that one. I have to say we haven't come across many ridiculous requests yet. But – and it's – again, we want to be a company that is – Right, right, you will. Yeah, I'm sure we will. Yeah. We want to be a company that's easy to work with. And we try to accommodate from customers as well as we can. But, you know, typically, as an analog supplier, there is thousands of customers, potential customers out there. And we have to be able to communicate with all of them without physically talking to them. So it's key to get a lot of this information out through the typical operating characteristics in your data sheet or app nodes and things like that. That would be the preferred way of doing it. Okay.

Dave Jones: Because I can imagine that every customer would want – would, you know, like have a particular spec of that product that is important to them. And then another customer will have another part of the spec that's important to them and all the other specs don't really matter that much. Is that – I can imagine that's very difficult because you would have to try and put in as many characteristic curves for as many aspects of the product as possible.

SPEAKER_00: The latter is certainly true. If you talk to our apps engineers, I'm sure they would agree with you that there's a lot of these curves that need to be measured. But the first part, I'm not so sure. But people do care about the same features, right? They care about the speed or propagation delay or bandwidth on one side. Power. And then they care about how much power do I need to spend to get that and what's the accuracy – How much does it cost? Yeah. Well, obviously, that's the first one. How does it cost? How much does it cost? How much I'm not even tired. Yeah, these days.

Brett Fox: I think that goes back to your digi-key search that you were talking about. Yeah.

Chris Gammell: Right.

Dave Jones: Of course. You search for function and then price. Absolutely. You search for ascending price. Yeah. Is that something you're targeting? Are you targeting the low-cost space as well, unlike Maxim, who are pretty much a premium price supplier?

Brett Fox: I don't think we want to be – never considered to be a low-end manufacturer is good. I think we want to be considered a reasonably priced manufacturer for what we're giving people. We can't give it away because we've got to make money too. Of course. We don't want to be these guys selling LM324s or whatever for $0.08 or whatever we're going for these days. That's not our deal.

Chris Gammell: Yeah. Speaking of the cost, and this actually takes us into our first listener question. There's a lot of cost behind all that stuff, like you said. Yes. One of our listeners, Carmen, asked about if there's any plans to do really large app notes, like Jim Williams style. You said you were emulating – you hope to one day grow up to be similar to LT. I mean, is that going to be on the app side as well? Do you see that pushing more sales?

Brett Fox: Absolutely. First off, I don't think there'll ever be another Jim Williams. He was a very unique and talented individual. Having said that, you can certainly take bits and pieces, incorporate that into your own personality and the personality of your app. We have some pretty prolific writers who are brilliant engineers on the app side who are writing very nice app notes now. We'll continue to do that because it's both important for customers and it's important for us to brand to do that.

Dave Jones: Okay. Is it hard to find good apps people like that? I can imagine it would be. It is. It would be harder than finding actual designers.

Brett Fox: If you ask your own, he will tell you from his experience. It's kind of funny. He and I have this debate all the time about this. And his feeling is it's harder to find a good apps guy, a real good strategic apps guy than it is to find a really good designer. Interesting. Which tells you the import of finding good apps people.

SPEAKER_00: Yep. I certainly believe that.

Chris Gammell: How would you define a good apps person then? What do you look for specifically for any of our listeners that might want to be wondering?

SPEAKER_00: That's exactly the problem right there, right? How do you exactly define it? It's somebody that has experience doing board design. Somebody that has affinity with the way that the ICs are going to be used. But they also need to understand what it takes to actually design the IC. They may not have to be able to do it themselves, but they do need to understand enough of it. They know what is possible and what's not. And that's not something, you know, we talked about schooling earlier. There's PhDs that are working on the most advanced IC designs, right? There is no real school that you go to to become a very good apps engineer. You have to learn that in the field by working. School of hard knocks. It's difficult to find.

Dave Jones: And you've got to have a passion for that. You've got to have a passion for just building stuff up and building different things up, seeing what this chip is capable of, thinking outside the box and, you know, using this triple five timer for 20 million different applications. Yeah, that's absolutely correct. You know, that's really hard.

Brett Fox: So we want to, part of what we do, which was somewhat alluded to regarding characterizing our parts, is we road test every single part we do, you know, to the specifications and beyond, because we want to see where it's going to break, how it's going to work, and in different ways, because I think Dave, you were making the point about different specs are interesting to different customers. Also, you'll get customers pushing the limits in ways that they shouldn't. And you want to make sure that there's kind of soft failures, not hard failures for your parts when that happens.

Dave Jones: Are your specs typically conservative then?

Brett Fox: Yes. Yes, yes, they are.

Dave Jones: Like your absolute maximum specs conservative?

Dave Jones: You know, so if you go to 6.1 volts over the six, it's not going to pull up.

SPEAKER_00: I don't think there's a lot of companies out there that dare do that nowadays. It's a very dangerous game, right, because it fights you really badly. Yes, we are conservative, but that's no different if you go to companies like Maxim LTC. They're all doing that.

Dave Jones: Yeah, right. Yeah, you can just go, oh, yeah, I can drive 100 milliamps out of this pin. No worries. She'll be right.

Chris Gammell: But it's tough because, you know, specsmanship is just a tough thing in general. I mean, because there's even shades of gray with, you know, are you speccing typical, are you speccing maximum? You know, and then there's the art of reading that as the board level designer and trying to figure out, well, you know, Touchstone is known for always speccing maximums.

Chris Gammell: And then I can squeeze out a little more. Or they, you know, company expects typicals and I don't trust them as far as I can throw them, you know. So, like, it's just what you come to know about a company, you know. So, it's – who actually does that specification then? Like, is that just the group in general? Is that, like, the characterization group?

Brett Fox: It's essentially a contract between the designer, the product definer, and marketing. You know, so those three groups have to agree on what the specifications are going to be when we launch a product so that we make sure we're all in sync. And then if there are fundamental changes from those, which do happen, I mean, as much as we like to say that we know what we're doing and we're going forward in the right direction, you know, things do change and stuff does happen. And if there are fundamental changes where we can't achieve something, then we have to agree or else, you know, we're going to stop the project, which is a rarity. We haven't had that happen here yet at Touchstone. But it does happen and where you see people, you know, get in trouble is that they'll let projects go and then you have your oh shit moment, you know, where suddenly it's oh shit, it costs twice as much as it should. Oh shit, the specs are two times as worse as they should be or whatever it is. And then suddenly, guess what? You know, you don't have a product that's going to sell or you don't have a product that's going to make money and that doesn't do anybody any good.

Dave Jones: I wanted to ask a question about Chinese parts. Like you were talking about like an LM324 before, you know, and how, well, you probably wouldn't touch like an LM324 because they're churning them out for, you know, half a cent in China. Who's actually churning them out for that price? Are there actually these mysterious Chinese companies who, you know, churn out these jelly bean generic branded parts that you'd find in, you know, some little, you know, toy, you know, kids toy at the... Or what? Have you got an experience with that at all?

Brett Fox: No, I don't. You know, so I don't know of anybody doing that. But certainly there is a portion of any business, which is the jelly bean portion. And there's probably somebody out there making LM, there are enough guys probably making LM324s or 555 timers out there where there's somebody doing it for nothing. Or, you know, thinking they can, there's the other classic of we'll make it up in volume type thing. Yeah. You know, where somebody thinks they can make it up in volume. So maybe there's somebody out there doing it. That's a lot of fun. I don't keep track of that part of the business at all. Right.

Dave Jones: The old saying, how do you make money selling free stuff? Volume. Yeah, that's right. Yeah.

Chris Gammell: Yeah. All right. Well, speaking of, you know, other companies, one of our listeners, oh boy, this is a Reddit name, so I'm not sure what it is. Fat Angaboo was asking about other... Right, you're going to get that online, I guess. But some of the other analog startups, and some of these I've actually never heard of. So Linear Integrated Systems, VAT Corporation, and Solid State Microtechnology, which was later...

Dave Jones: VAT Corporation? VAT Corporation?

Chris Gammell: T-H-A-T, yeah, VAT.

Dave Jones: Seriously? That's their name?

Chris Gammell: I guess, yeah. I've never heard of that one. That one.

Dave Jones: It sounds like, you know, someone you'd get on the shopping network, you know? You know, a company selling the latest, Abdul, from VAT Corporation.

Brett Fox: I'm sure that's an acronym. That is probably an acronym for something. Right. I'm sure it is. I don't know what it stands for, but it's an acronym. Yeah.

SPEAKER_00: That, you know, for some of these guys, I had to look them up because I hadn't heard of them. But that is an audio-only company that took over the... They are not fabless. They do have their own fab. They took over a fab that was once Sidebackses a long time ago. But that's a four-inch line. So that works only if you work on small volumes, very high-margin parts. So it's a very, very different business model than what we have. They've been around for a long time. They've been doing this for 25 years. So whatever they do seems to work. But our goal is very different. We want to touch many more customers and grow substantially faster.

Dave Jones: No pun intended. Touchstone. Is that where the name came from? Can you explain the name? How did you come up with it?

Brett Fox: In a thesaurus. So if you look in a thesaurus, you type in Maxim. We need a name. Yeah. You need a name. So we go, hey, we're going to be Maxim-ish. So let's go type in Maxim into a thesaurus. And if you look at the various acronyms for Maxim, Touchstone's in there.

Chris Gammell: Wow.

Brett Fox: And it was the best one. And it kind of had a cool feeling to it. And you couldn't really peg it to one thing. So we liked it. That's the one we went with. Oh, that's terrific.

Chris Gammell: Man, we're in good company here, Dave. This is like how we come up with show names. Like our funky show title.

Dave Jones: I love it.

Chris Gammell: We'll have to type in Touchstone and see what comes up for that for the show title today. Wow. There you go. Oh, boy. That's great. Well, so to get back to the, you know, we were asking about the app engineers. And there's no school for that kind of thing. But, you know, someone else had asked about, you know, what you'd look for aside from the 20 years of experience, obviously. You know, is there anything else that you look for in the analog designer side? Or is it just mostly the experience and the patent portfolio?

Brett Fox: Well, there's a spectrum in there. I'll let your own answer. But there's a spectrum in there regarding that. Yeah, there's a spectrum in there in terms of the years of experience. So what he was talking about was the minimum that we would look for. But you've got kind of your thing in terms of what you guys look for on that end.

SPEAKER_00: Yeah, yeah. As I said before, at this point, people that have done their own design, brought out their own, taped out their own design, preferably at another company, and got some industry experience. Ah, okay. It doesn't have to be 20 years. That's certainly not. But a few years, yes. But I think there was a question also about how do I prepare myself for a career in analog? And definitely, you start out with, if you still figure that out before you choose your college or university, then go to a place where that's actually something that they care about. And you find that out by looking, is there anybody here that pushes things, that is moving the needle in terms of analog design? And that's where you want to be. Because from working with people like that, that's how you learn it. And then when you come out of school, then companies like us and companies like Linear and Maxim and TI, they're going to be interested in talking to you. And then once you have a couple of years' experience at a company like that, then once again, you build it up, right? So that's how it goes on.

Dave Jones: Sure. Speaking of patents, or patents, as you Yanks call them, is that important to you? Are patents important to you in terms of your product portfolio? Yes. So you're fans of the patent system? Is that good for getting more venture capital funding?

Brett Fox: We're not fans of the patent system at all, but we have to use it. Right. And unfortunately, in this day and age, it is so important because it's both defense and offense at the same time. So you need patents because there is an old saying, and the old saying goes like this. If you're a semiconductor company, at some point, guess what? You're going to get sued. Even if you don't intend to and you're not trying to infringe on somebody's patents, somebody's going to come after you and somebody's going to take you to court for something. So the more portfolio patents you have, the more you have to trade. And then on the other end, when your portfolio gets big enough and you get strong enough, guess what? You can become the hunter.

Dave Jones: You can stop being the big daddy and suing every bastard. Oh, great. A vicious circle, huh?

Brett Fox: It is. I mean, the patent system does in our country is horrific. Let's be honest about it. But we have to deal with it and that we're not going to change it.

Dave Jones: Would the goal for you guys to be every new design, somehow you'd be aiming for a patent on that?

SPEAKER_00: Yeah, that's probably a good way of putting it. It's not a goal in itself. The goal is to come out with a product that's going to sell. But as a rule of thumb, that would be about it. Yeah.

Dave Jones: And typically, what aspect of it would you patent? I mean, if it's an op-amp, clearly you can't patent an op-amp. Would you patent a couple of transistors in this configuration?

SPEAKER_00: Yeah, it could be as simple as that. It could be something that's very much on a transistor level where you say, okay, well, the way that I'm using the transistors in this configuration is new. And that gives me these advantages. But it could also be, and those are actually the better patents to the app. The stronger patents are the ones that are more on a system level. They're stronger because they're easier to detect if somebody else is doing it, right?

Dave Jones: Of course. I was going to say, well, that's the other thing, right? If you patent how a couple of transistors are configured, how do you know that somebody hasn't done that somewhere? I mean, just the sheer odds are that, yes, somebody's done it somewhere, you know, and published it in some obscure magazine 30 years ago, you know?

SPEAKER_00: Yeah, that could be. And then that opens up a whole other can of worms because if you find that out after the fact, it's all good. But it's also from a protective view. Only if your performance becomes so much better than what you could do without that way of doing it, otherwise you would never find out if anybody else is doing it, right?

Dave Jones: Got it. Yeah, of course. We've got another question from Jose Kieran. I think that's correct. And he asks, what development tools do you use for IC design?

SPEAKER_00: We're using Cadence, almost exclusively. We have some other EDA vendors whose tools we use as well. But the majority of the suite is Cadence, both front and back. So both for schematic entry and simulation as well as for layout.

Dave Jones: Is there a reason for that? Were you familiar with those tools or was they simply the best option by far?

SPEAKER_00: Well, most familiar with the bigger name. So with the tools from Cadence, tools from Mentor and some of the other players. But Cadence is, I think it's fair to say that they're still the touchstone of the EDA world, right? They're the standard. And we enjoy working with them. So it was a natural choice.

Brett Fox: And there's actually one other aspect to Cadence too. When you're recruiting, it's interesting. But designers ask you the same question about what's your tool set. And because Cadence is so well known in the analog world, it's actually a recruiting tool to be using Cadence as well.

Dave Jones: Of course. Yep. Because there's only really two major players that you would consider using, Cadence and Mentor, right? Is that probably true when you're doing a startup semiconductor company?

SPEAKER_00: Yeah, that is correct.

Dave Jones: And how much do tools like that cost? It's probably, you know, you probably can't say, but it's millions, right?

Brett Fox: Well, we can't say. What I will say is in terms of capital expenses, it's one of our highest capital expenses that we have.

Dave Jones: That's what I always thought.

Brett Fox: Yeah, it's very expensive.

Dave Jones: Well, I mean, what expenses would you have? You would have wages and you'd have CAD tools, right? That's probably it.

Brett Fox: Mass cost. Mass cost.

Dave Jones: Oh, well, of course. Mass cost, of course. Wages.

Brett Fox: Yeah. Which fortunately for us, you know, we're not in deep submicrons, so we can actually afford to do what we're doing. Yeah, we've got a manufacturer. But that's, well, different story. Yeah.

Dave Jones: What's the price difference between, you know, like the masks that you're doing and masks for, you know, high-end, really low feature size? Millions. Oh, yeah. It's in the next. Millions.

Brett Fox: It's in the millions of dollars difference. Yeah. We can get ours for relatively cheap. And it's all relative.

Dave Jones: As in what? Under 100K or? It depends on. It depends on. 100K?

Brett Fox: Somewhere in that range, I would say. It just depends upon what we're doing and who we're using. But, you know, if you're looking at doing something deep submicron, that's well over a million dollars to do that. Wow. And that's the other reason why you don't see a huge amount, especially of digital companies coming out in the startup world, because you just can't afford to do it. It's just too damn costly to get there.

Dave Jones: Yeah. How confident are you in your tools that when you push that button to produce that mask that it's going to work? And the other question is, the flip side of that is, how often has it gone wrong? Look directly at your own right now.

SPEAKER_00: Because that's the question they ask me every time that we have to sign a PO to purchase another set of these masks, right? But, yeah, so with the tools nowadays, you can catch almost all mistakes. Not everything, but almost everything. So it's very unusual for us, and that's not just over here at Satchelon, but it's unusual to get your parts back and see that they're just dead, that they don't do anything. Right. It's –

Dave Jones: But it has happened, has it? We had an interesting – Has it happened at least once?

Brett Fox: I'll tell you a story about that, actually. So the first part that we ever did was the 1001, which was designed by your own at night and on the weekends. And our foundry – we got it back. We were so excited. He and I were in a conference room. We could see the truck coming with the wafers. So he goes and runs into the lab. I go run after him. And he's like, I've never had this happen before. Yeah. Nothing's functioning. And I was saying, boy, maybe something's wrong. And it turned out, after a couple days of investigation, that our foundry vendor had inverted a mask layer. Yeah. And caused the whole thing not to work. So it cost us – Do you get a refund for that? Well, what happens is they fix it at their cost. At their expense, yeah. The killer, by the way, guys, the killer is the time. Oh, yeah. Because the one thing that you're fighting and starting a company is time more than anything else. So when you lose that time because of a mistake, be it yours or somebody else, it's not like in software where you can just re-spin the software and out of pops the day after. When we have a mistake, it's several months. So it was brutal for us when that happened. It really, really hurt.

Chris Gammell: You guys make board-level problems look even easy compared to that, right? Yeah. Yeah.

Dave Jones: Board-level is easy. You can spin a new board in 24 hours if you want. You just pay extra money and you get a new board in 24 hours. How did you find that that mask was inverted? What steps did you go to when you finally realized? Did you have to look at the dye under a microscope or what? How did you –

SPEAKER_00: It's all of the above. You measure. You look at the dye. You look back at databases and masks that were generated and eventually you found it. As difficult as problems are to find on an IC, in the end, you always find it. This one wasn't very difficult, actually. Yeah.

Dave Jones: Yeah. Well, you had to. Was that the last thing that you would have thought of? Oh, it's not.

SPEAKER_00: Certainly it wasn't the first thing that I thought of.

Speaker ?: I was sure that was.

SPEAKER_00: Right. You think like, oh, maybe it'll oscillate or maybe it's right to the geotypical problems. Yeah, yeah, of course. But this wasn't one of the ones that we didn't mind. But yeah.

Dave Jones: Because if you power it up and just nothing happens, that's the worst kind of fault to troubleshoot, isn't it? Well, sometimes. It can be. No.

Chris Gammell: Wow. Wow.

Dave Jones: Wow. On your first ever part, did that really, did that make you think twice about getting into this business? Like, oh, is this how it's going to be? No, no, no.

SPEAKER_00: Not really. No. You guys have experience, so that helps. Yeah, yeah. We've seen.

Brett Fox: Yeah, when you've been around as long as we have, you've seen it. So it wasn't what we expected, but you know what? It's life and you just go deal with it and you move on.

Chris Gammell: You know, as a fabless company, how hard is it to grab the ear of the actual fab? Because I would think that, especially when you're starting out, it would be tough.

Brett Fox: It's a great question, Chris. This is where having good investors really helps. So, you know, when it comes to the people that we have around the table, both in our investors and our board of directors, many of us have personal relationships with the senior people at these foundries. You know, so we're fortunate enough because we have those relationships that we can get the ear of the foundry. And, you know, the thing that you're doing is one of the other things that, you know, if you're doing what we're doing, you're always constantly, you know, selling your company to somebody. And what I mean by that, you're telling your story on just about a daily basis, just like we are to you guys, you know, to somebody. And, you know, part of the reason, you know, if I work with a TSMC, for example, you know, TSMC is very picky about who they do business with. And if they want to do business with you directly, they want to be convinced that it's worth their time because what they'll tell you is on any startup, they're going to lose money. So it's an investment for them and their time. So they wanted to believe in us. And, you know, with the relationships that we had, you know, we ended up telling our story to Rick Cassidy, who's the president of TSMC North America. And Rick's been a big supporter of us almost from the get-go, you know, because he's an ex-national guy, believes in the analog model and believes in us. And it's been a very good relationship with him. It's been a good relationship with our other foundry vendors as well. Huh.

Chris Gammell: And so do you guys spread it around? I mean, I don't know how much you're allowed to say about that kind of thing, but are there... We'll say a little bit.

Brett Fox: Yeah.

SPEAKER_00: So we do. Yeah. Yeah. We do. We have several foundries that we work with. Okay. That's smart.

Dave Jones: And why? What's the reason for that? Just, you know, covering all your bases, make sure you've got second sources in case they're busy at the moment?

SPEAKER_00: It's partially to get features that one doesn't have. You go to another. And it's also to cover your bases. Okay. It's all up to you.

Chris Gammell: Right. That's very smart.

Dave Jones: So when you design a chip, you would have a particular foundry in mind? Yeah, yeah. Definitely. You'd be using their toolkit? Yes, definitely. Is that...

Chris Gammell: Yeah. Have you seen any of these found... We've had a couple of questions about the more out there, not geometries, but topologies, I guess, with silicon carbide and graphene and everything else. Have you guys started looking into that side of things? Or is that kind of...

SPEAKER_00: Even if we would want to do it, we can't, like you said, it would need to end up with the foundry first before we can do anything with it. And graphene is definitely still in the... IBM did a lot of work on it a couple of years ago. I think MIT did some things. Yeah. And it's all very promising. It's still...

Brett Fox: It's a very small market. There was an article, interestingly enough, today about graphene, I think, in EE Times Europe. And I think the market size for it is maybe $125 million today. So if you look at the overall semiconductor market, it's a gnat. It's not anything. It's maybe someday, but not yet.

Chris Gammell: Yeah. Well, you mentioned Sigi before. Very good mobility.

SPEAKER_00: So that's good.

Chris Gammell: Yeah.

Dave Jones: We've had an ongoing discussion in quote marks on this show for the last two years about something Chris brought up. And way back in one of our early episodes, he made the claim that within, what was it, Chris? Five years, we would have chip printing machines on our desktop. Printing machines. I mean, I wanted...

SPEAKER_00: The 3D printing.

Dave Jones: Like, as in, you know... Yeah, no. Well, as in, you could print your own chips. Oh. Right? There's a... Yeah. Ah, Chris has gone offline, so he can't defend himself here. This is great. Yeah. Chris claimed that, you know, all of these... Yeah, we would have printable electronics, right? So that you wouldn't be buying chips from the component manufacturers like you guys anymore. All of the designs would be online. Everyone would have their own little chip printing machine. And they'd, you know, you'd print out your new chip. And Chris said, oh, yeah, it'll happen in five years. Here he goes. He's back. We're having mumble problems. And I called complete bullshit on the idea. I said it wouldn't even happen in 20 years. Do you guys... Have you guys investigated the printable electronics area? I think I'm more in the...

Brett Fox: I think I hate to say it, Chris, but I'm more in the bullshit camp on this one. Yeah. So, sorry about that. Yes. See, now this actually has some weight behind it. That's the problem.

Chris Gammell: Yeah. Well, when a CEO of a semiconductor company tells you that, that hurts. But you know what? I'm going to hold out hope. I'm going to hold out hope. And I'll...

SPEAKER_00: Because it's just too complex. You know, it's going to be very much for small things that are not too complicated. Because geometries have to be, by the very nature of it, have to be very much scaled up again. Right, right. So, it's... Yep. The quality can never be what you can do if you have what you can do with silicon nowadays.

Dave Jones: And how can you beat it? I mean, the component availability these days, you go on DigiKey, how many hundreds of thousands of parts are there for every conceivable application? And they cost like a dollar. I mean, there's... You know, well, you know, average... I don't know. The average chip cost might be a couple of bucks or something. But, like, how can you beat that? You can't. There's no point having a machine that will print chips yourself. Because they will be uncharacterized. They'll be untested. They'll be un... You know, all that sort of jazz. So, really, you can't beat being able to buy a 50-cent microcontroller, for example. Yeah.

Brett Fox: And then you also have the issue of how you're going to package them and how you're going to cup. Yeah.

Dave Jones: Exactly. Scissors. Yeah. Scissors. Speaking of packaging...

Dave Jones: Speaking of packaging... So, I win that one. That's just... I just want to confirm that, that I win. Chris, do you formally give up?

SPEAKER_00: He's offline again.

Dave Jones: Oh, he's offline again. He can't defend himself here. By definition, Dave, he's given up. Right. He's quit the entire radio show business in protest. All right. Speaking of chip packages, what... Do you have favorite packages? Do you find that your customers want the more... Like, the really ultra-small packages? Or do you... Do you always make a thing like, oh, we're going to make it available in an SO8 or something like that?

Brett Fox: No, I think... Because we believe that's important. I think the general rule is smaller is better. Everything's shrinking. Everything's smaller, so...

Dave Jones: Oh, no. Don't tell me that. Don't tell me that. Yeah. It's depressing.

Chris Gammell: Yeah. Maybe it's a good time that they can bring up their... You guys have a lot of breakout boards available. And demo kits, right? I mean, that's a good program you guys do.

Brett Fox: Yeah, absolutely. So we try and make everything we do available with demo boards. And we're pretty... Relatively pretty easy to get them from. You know, so we're trying to be very liberal with that. I think we think it's an enabler for our customers. So, you know, people ask for them. Generally, we'll give them away. And we're certainly giving away free samples every chance we get. You know, plus there are vendors who have... In fact, we have an app note on our website about this. Vendors who have available from SO to DIP type things where you can do... So you make it a little bit easier to evaluate, you know, parts. Yep.

Dave Jones: So do you find that those free samples are really sucking engineers? Yeah. Is that a... Are we that easy?

Brett Fox: I hate to say it, but, you know, go back to your Maxim experience. You know, it's one of the... Yeah, exactly. ...is one of the positive things that we learned from Maxim. Is Maxim did this work really well for them. Now, what we're doing that's different than Maxim is our parts are actually available. But, you know... You know, but you give people something...

Dave Jones: Wham! Take that.

Brett Fox: Yeah, exactly. Wham! Maxim, guys. You know, take that. But it is a good, simple thing to do that makes it easier for people to do business with you. And that's what we're essentially trying to do.

Dave Jones: So when you bring out a new chip, would you go, right, we're going to manufacture... How many would you manufacture up front? Even though, like, I assume you'd have no initial orders when you do a chip. You just think, oh, we'll put it out there. Would you manufacture, like, 100 reels? Or what is the sort of minimum you would order from a family?

Brett Fox: Well, you've got multiple wafers that you have to use. So about the minimum run that you can do is 12 wafers, depending upon the diaper wafer that you get. You know, especially in this day and age where most of the founders are 8-inch wafers, you're getting a hell of a lot of parts, you know, initially. And then you make use, in today's world, liberally of dye bank. So from dye bank, you can build out in 4 to 5 or 6 weeks, depending upon, you know, your back end.

Dave Jones: What's this dye bank?

Brett Fox: Well, dye bank is, if you can imagine, you're holding either in wafer form or dye form your product. You know, so it's prepackaged. Oh, okay, right. It's prepackaged. And then you package it up. Sometimes, you know, especially with us, you know, we may do some personalization of the part, you know, post-package. Yeah. You know, there's various trimming that we're doing. Yeah, we're doing various trimming types of things that we might be doing. So it gives us a lot of flexibility, and we can get parts to a customer very, very quickly. It also, which is a very pragmatic thing, is it limits our, I'm trying to think of the right word for it, but it limits our downsides, so to speak. Where if things don't go right, we're not sitting there building up a lot of stuff, and the cost goes way up. In dye form, it's cheap. Yeah. In packaged form, it suddenly gets a lot more expensive.

SPEAKER_00: Once it's put in a package, it's not just that, right? Then you also test it. So now it's the fully loaded cost.

Brett Fox: Of course.

SPEAKER_00: Whereas if you keep it in as a weight, much of the cost hasn't gone in yet. Yeah.

Dave Jones: So what would the cost breakdown be compared to, like, the wafer, the packaging, and the testing? Because they're the three major... In percentages, you mean.

Brett Fox: In today's world, this is, I think, one of the other big things that you can think about, because it does vary from part to part, by the way. So in general terms, the definite cost is more in the packaging than it is the dye today. It has changed significantly from when I first got into this business. You know, and I think about Maxon back in the day, we were on five-inch geometries, if you can believe it, with four-inch wafers. Yeah, you know, so the cost was much more in the dye than anywhere else. But it's changed a lot.

Dave Jones: Are there any questions that you saw on here that you wanted to answer, guys?

SPEAKER_00: I don't know. The examples of other companies that were out there were all companies that had been around for 25 years. They're not really exactly the same type of companies that we were.

Brett Fox: The only other comment I would make, I don't know whether it came up in one of the questions or not, is that I think we are the last or one of the last semiconductor companies that you're really going to see that's venture-backed, that's going to build to scale. It is, I hate to say the word dying, but I think it's... Really? Really? It's just because, well, because if you go back to the story I told at the beginning of how hard it was for us to get our funding, and we were a pretty experienced group with a pretty good reputation. It's just, it's very, very hard, and it's very difficult for various reasons, and we could do another amp hour show on that. But it just seems to me from the experience that we've been through that maybe there'll be another one that comes along, but I don't think there's going to be a lot more. I don't expect there to be a big resurgence in semiconductor companies starting, even if we're successful. I don't expect that to be what happens. I think for various reasons, the heyday of the semiconductor startup is probably passing, which I think is in one sense good for us because it means there's going to be limited competition. Taking my touchstone hat off, I think it's bad for our industry because that means there's going to be less innovation, which is sad to hear.

Chris Gammell: So where do new companies come from then? Would that just be from spin-outs and stuff?

Brett Fox: I don't think you're going to see very many new companies. I think that's part of the corollary. So I stock with what we're going on. I think you'll see the odd one or two come out, but I don't think it's going to be the large group that it was years ago, and that's disappointing. I hope I'm wrong, quite honestly. I truly do.

Chris Gammell: So no China upstarts?

Brett Fox: Well, it's interesting. There certainly are more in Asia than there are here. You could look at China. You could look at India. They certainly have tried the analog model over there. It hasn't gone great. In fact, there was a trend in the early part of the last decade of the Chinese analog startup, in theory, because the cost of engineering is so low. And those things looked great on paper, but the reality was they didn't work because the engineers couldn't get anything out. So it just didn't end up happening. So the few companies that ended up doing that route didn't succeed. The one company that does stand out in my mind that kind of has somewhat of a model like that is Monolithic Systems, if you're familiar with them. They're actually an American company based here, but they have a large design center over in Asia, if I'm not mistaken. So they may be – but they've kind of stalled because their model is kind of narrow. But they have made it, and they are public, which says something.

Dave Jones: Do you see any more consolidation in the market?

Brett Fox: Yes, almost by definition.

Speaker ?: Right.

Brett Fox: Almost by definition, that's going to continue happening. There's no question about it.

Dave Jones: Even major ones like TI buying National?

Brett Fox: We'll see. You have to be a soothsayer to have said you're going to see that one come. Yeah. Who knows the real reasons why it happened or what. And it's not clear yet whether that's going to be successful for Texas Instruments as well. We'll see.

Dave Jones: Of course.

Brett Fox: Typically, the other thing of our industry is typically mergers don't work very well. A lot of it is you have a community of designers that's used to working a certain way, and you merge that with another community of designers, which is used to working another way. And typically, it's like oil and water. They just don't mix. Yeah. Culture clashes. Yeah. Culture clashes. And we'll see what happens. You've got a Texas culture with a Silicon Valley culture. They are different. Those companies were major rivals for many, many years. Yeah. That's going to take some time to sort out probably.

Chris Gammell: I agree.

Dave Jones: What would you guys do if you were approached for a buyout? You know, we're not for sale.

Brett Fox: Not for sale. Yes. That's what we're wanting to hear. But the reality is I say that laughing because, you know, if the price is right, then, of course, we're going to listen to it. But the reality is at this stage of our life is anybody who wants to buy us is going to try and buy us not for the right price. And it's just the way the world is. It's a buyer's market, not a seller's market. You know, so for us, you know, there are a couple of things about us. We didn't do this. You know, we were talking earlier about making money and all of that. And we certainly went into this with the goal, obviously, of winning. But if we wanted just to make money, all of us would have stayed at maximum or linear. That wasn't why we did this. Oh, okay. We did this.

Dave Jones: Right, as in personally wanted to make money.

Brett Fox: Yeah, we did this because we believed in what we're doing. We thought this would be something that was, you know, fun. I don't want to use the word glibly, but we thought we would really enjoy this and have fun with it and make it a very successful standalone entry, which was part of our goal. And it's part of why we have a diversified business. We're not just in one area. We wanted, we were thinking of the future when we started the company. We were thinking of the long term when we started the company. And we didn't want to be a one-trick pony, which is why we're doing it the way we're doing it.

Dave Jones: There you go, kids. That's the lesson for today. If you want to start a semiconductor startup for money, it's a mugs game. That's right. Oh, boy. I've got another question. We were talking about the die before. Do you guys, like, if you've got leftover space on a die, do you put, like, a little picture of an animal in there or you sign your names on there? Is that allowed? Is fun allowed at Touchstone Semiconductor? Farm animals, right?

SPEAKER_00: We do the farming. Yeah, yeah, yeah. I've seen a couple of examples like that. Yes, yes.

Brett Fox: But there is something we do. No, no, no, not here. But we do do something fundamental, though. We do, maybe not between the die, but we will do test chips.

Chris Gammell: Yeah, yeah.

Brett Fox: What we will do is carve out a few die if we've got something going. In fact, we just did that on a development effort. Because if you want to test something before you see silicon, you can take a block and you can put it on another die intended for something else. Oh, right. And then you can get a first look at something. And it does give you a step up. Because as great as the simulation tools are today, there's nothing like seeing silicon.

Chris Gammell: Just like having a board in your hand. Exactly right.

Dave Jones: What would be typical differences that you wouldn't see in a simulation that you would get on die? Is it oscillation or is it excess power consumption?

SPEAKER_00: Yeah, it could be all of the above. But it typically has to do with parasitic coupling that wasn't perfectly modeled when you did it.

Dave Jones: As in from transistor to transistor? Yes. Like you get incapacitive coupling or something?

SPEAKER_00: Yeah, from transistor to transistor through substrates or wells or what have you. And they can play tricks on you.

Dave Jones: Is that just due to the process tolerances? Manufacturing tolerances on the die?

SPEAKER_00: No, not necessarily. What is that? No, it's...

Dave Jones: Oh, they just haven't got the model accurate enough?

SPEAKER_00: Yeah. If you imagine it, it's difficult to grasp in a model. Of course. We use tools that are called layout parasitic extraction to try to grab all these features and put it into an electrical model. But there's a limit to what you can do there.

Dave Jones: I think we're almost up on our questions.

Chris Gammell: All right.

Dave Jones: Have we gone through them all? Have we covered them all?

Chris Gammell: Well, we have one more. Chris? One more. Just kind of a general future predictions. I mean, where do you guys see the sensor market going? I mean, you guys play in the sensor market a lot. But where do you guys see tech going in the next five to ten years?

Brett Fox: Technology as a whole. I think, you know, there's the classic. It's going to be smaller. It's going to be lower power. And it's going to be cheaper. I mean, that's the general thing, I would say, that we're going to see happen. I think you're going to see, you know, even for us, even though we're not a, you know, it's funny. There's all this talk about integration these days. You know, we're going to be integrating more and more as well as all of our competitors because the feature sizes are shrinking. You'll see, you know, from a company like ours, which we truly hope for, the beautiful thing about our world is that it's not a static world. So there's going to be new functions that nobody ever thought of or dreamed of. And hopefully we're going to come up with our fair share of them. They're going to lead to new market opportunities and new things that, you know, board-level designers like yourself can do. And that's where, from our perspective, we get so excited about the future because we know and we can see some things. We've got some things in development which, you know, nobody's ever done before. And those are the ones that, you know, can lead to, not all the time, but can lead to real groundbreaking things, lead to major shifts in the way people do things. So that's what we're excited about and that's kind of what we see.

Dave Jones: Do you guys have like a slogan or anything like that? We do.

Brett Fox: If you go to our website, it says we are the 21st century analog IC company. And the general concept of that is it's… Oh, trademark. Yeah, trademark. Trademark. Yeah.

Speaker ?: Look out.

Brett Fox: You know, the thought process was, you know, one, it separates us from all of our other competitors who were started in the 20th century. And it also points out that we are new, we are modern. And our whole thought process, which we've talked about a few times in our discussion today, is we want to be really regarded as being easy to deal with. You know, we're not trying to put barriers between ourselves and our customers. We're really trying to make it simple and easy. I know we're not going to be perfect, but, you know, having said that, we are really striving hard to make the process of working with Touchstone one that is enjoyable for our customers and for us.

Dave Jones: And I noticed that you sort of strategically added high performance in there.

Brett Fox: Yes.

Dave Jones: To make it even a longer slogan. Not just the 21st century analog company, it's high performance in quote marks. Why did you add that?

Brett Fox: Well, trademarks have a minimum word count to them. Oh, do they? I'm just kidding, Dave. It's because we are also a high performance analog company. So, you know, we wanted to make sure we got that in too.

Dave Jones: As in, as opposed to a jelly bean company. Yes.

Brett Fox: Yeah, it wouldn't be too good to say we are the 21st century jelly bean analog company. So we figured we'd go for the high performance angle. And, you know, and after we had a few beers, it sounded really good. So we went with it.

Dave Jones: And who designed the logo? Did you shop that out?

Brett Fox: No, you know, that actually a friend of ours did it for free. So, you know, hey, when you're a startup and get something for free and it looks good, it works. You know, so that's why we did it.

Dave Jones: Absolutely. Are you going to be one of those companies that bloody change their logo on the chip every year?

Brett Fox: Let me put it to you this way. We are not going to change our logo to green. And it's not going to have some silly little M in the middle of it, you know, like somebody else might have just done. No, we're not going to do that. We're not going to pay somebody God knows how much money to do that. No, I think we like our logo. Think of the consultants. Yeah, that's right. Think of the poor consulting business.

Chris Gammell: How are they going to feed their families? I'm not with touchdown. That's right. Awesome. That's what we like.

Dave Jones: Although your chips are too bloody small to put your logo on at least. You can't even put the bloody part. That's right. What's up with that?

Brett Fox: It's the way of the world.

Dave Jones: That's one thing that bugs me. Can you have like a list on your website, like a tab on the side that says, you know, decode the stupid number on the top of the chip? And then, you know. The little orphanage. Is there a – please, can you do that? I know we have it somewhere.

Brett Fox: Because – You know, it's funny that you mention that, Dave. I know we have it somewhere. I can't remember where it is, but I know we have it. And I kind of feel the same way.

Brett Fox: It should be on the front page. Somebody has to look at these damn things. I don't know who does it, but somebody does.

Dave Jones: Well, because I'm tearing down a product. You know, and I'm looking at these chips, and people want me to explain what the chip is. And I've just got to say, well, I don't know what that chip is. It makes me sound like a bloody idiot.

Chris Gammell: Yeah.

Dave Jones: Because it's just got some obscure number on it that I can't Google. Yeah. No. Well, thank you very much for joining us, guys. Thank you. It's been fun. Yeah, thank you very much. I think we're way over. We're like an hour and a half. Yeah. But we do like to go over with our guests. So people can ask me the questions.

Brett Fox: No, we're good. So thanks a lot. We're happy to do it again if you guys want.

Chris Gammell: Yeah. We look forward to all the new parts. It should be very exciting.

Dave Jones: And you guys have samples? Free samples?

Brett Fox: Free samples? Demo boards? Free samples, free demo boards, free apps assistance. Not free parts, though. And you're on DigiKey now. And we're on DigiKey. Right. Yep.

Dave Jones: Awesome. Great.

Chris Gammell: And on Twitter at Touchstone Semi and Touchstone Semi. That's right.

Brett Fox: On Twitter, we've got a Google Plus account now. So you can find us on Google Plus. We've got a Facebook page. So we are definitely soul media friendly.

Chris Gammell: That's awesome.

Dave Jones: Good work, guys. All right. Thank you very much for joining us. It's been awesome.

Brett Fox: All right. Take care, guys. We have fun. Outro Music