#281 – Crossovers and Call-ins

Chris Gammell: This is the Amp Hour Podcast. Recorded January 6th, 2016. Episode 281. Crossovers and call-ins. Welcome to the Amp Hour. I'm Chris Gammell of Contextual Electronics.

Alicia White: I'm Alicia White of Embedded.fm.

Chris Gammell: And I'm Chris White, also of Embedded.fm. And this is a crossover episode.

Alicia White: Dun, dun, dun.

Chris Gammell: I was going to say, cue the music. Yeah, you do that. It's like where the Fonz shows up on ALF and then they all go to Gilligan's Island, right? Someone had posted about the... Oh, it was Dave Vandenboud had posted about those. I didn't know there was... Are there really that many crossover episodes? I've seen some. They were big in the late 70s and the early 80s. Yeah. It was when you ran out of ideas completely. Yeah, right, right. Right before you jump over the shark, you do a bunch of crossover episodes. So that's coming, folks.

Alicia White: I'm so looking forward to them jumping over the shark.

Chris Gammell: I think we're the ones jumping.

Alicia White: Oh. Yeah. Whoops.

Chris Gammell: We're all jumpers at some point. So how are things going? I haven't listened to your episode this week, but I'll do that tomorrow. Well, it's only been up for a few hours, so I wouldn't expect you to get to it immediately.

Alicia White: I don't know. I've already gotten responses on some of the questions I asked. It's great.

Chris Gammell: Awesome.

Alicia White: This one is just me and Christopher talking to each other about random things. We talked about things you can put into your ear. Okay. You know? Like Q-tips? It was high technology.

Chris Gammell: Yeah. That's right. This is high. High. High. Q-tips at 30,000 feet. Cleaning your ears in an airplane. That's great. I can't wait to listen to that. If people haven't noticed, we are noticeably Aussie-less, but we will have at least, God, how many? Three. Two or three. Three? Three? Yeah, three Aussies who I'll be calling in. And of course, I mean calling in by, we'll call them on Skype, and through the magic of editing, it won't sound like it. Don't call us. We'll call you. Right. Exactly. Yeah. Yeah. So, what are we expecting here? Are we expecting more embedded, or are we expecting more non-embedded? I'm wondering if I'm going to be sitting here like, I don't know that. Ask those two.

Alicia White: I was sort of expecting card talk, where they ask us technical questions, and then we ask them about their relationships.

Chris Gammell: Oh, I like that. Actually, another one, too. I don't know. I was listening to the, I was listening to a recent, not so recent episode, and the trick I love is when they're stumped, they say, what color is the car? That's a very good question. Yeah, exactly. It's like, oh, oh, you're having trouble with your oscilloscope? What color probe are you using? What do you have this particular trace? Persistence set to. Yeah. Yeah. So, all the usual stall tactics, right?

Alicia White: I think since you did inform people that it would be us, that we're going to get a few more software questions, than normal. But if it goes too far software-wise, you know, I have a couple electrical questions outstanding.

Chris Gammell: You do? Oh, good. Are you still, is this the mice thing, or is there other things? We're way past that. Oh, yeah. Mice, you're gone. BB-8, something. No, that's gone, too.

Alicia White: BB-8's still around. I've been playing with those little tiny quadcopters, the Cheerson CS, Oh, yeah. Or CX-10s, that are just over the size of a quarter. And they have some really neat feedback control of their DC motors.

Chris Gammell: That's right. You did send me a sketch, and I tried in vain to guess what was going on there, and then it was quickly rectified. So that's, but yeah, it's neat. It was using a little STM32, or what was on there?

Alicia White: Yeah, an STM32F0 line, their Cortex-M0 line.

Chris Gammell: Oh, yeah. That's what I use for contextual electronics, actually.

Alicia White: The way that they're doing feedback is sort of cool, but let's see if we get time for it. If we don't, then I'll put up my best guess in sketch format, and all the electrical engineers can tell me just how wrong I am.

Chris Gammell: If you hear gurgling noises, followed by us dropping off, that's because we've drowned.

Alicia White: Yeah, there's a bit of a storm.

Chris Gammell: Oh, yeah, right. In your new abode. People don't know, Chris and Alicia moved to a place by the water, right? More water than we had anticipated. Right, well, yeah, especially now. Yeah. Oh, we should probably, so if someone hasn't listened to embedded.fm before, and they haven't heard when you guys were on the show before, which was 255 or 256. Can you give the quick rundown of embedded.fm?

Alicia White: Sure. Embedded.fm is the show for people who love gadgets. You can tell I've said that a few too many times. But it is somewhat the companion show to the Amp Hour that we focus a little more on the software side of embedded systems, and we focus somewhat more on the engineering as a whole. We talk more to educators and people just really enthusiastic about how technology goes in their lives. Although we do have a few shows that get pretty deep into technology, like keywords in C and processors.

Chris Gammell: Oh, yeah, Kalman filters. Kalman filters. Oh, yeah. Yeah, I remember that one.

Alicia White: So, yeah, if you are on the software side, you should come try us out too. If you're on the EE side, you should only come because you want to laugh at us when we try to puzzle out, say, motor circuits.

Chris Gammell: No, it's good. No, I mean, it's important to have that kind of crossover knowledge, right? Even I think a lot of people that listen to the Amp Hour too are interested in electronics and might be coming into it, and I think that's really important that electronics people move into the firmware side too. I've been talking to people recently. It's like, name a project where you don't have a micro on it these days. I feel like you can always have some of those, but it's just such a draw to have a small micro on there, like that STM32 chip or something.

Alicia White: It gives you such flexibility, and now that they cost like dimes. Dimes? Yeah, I mean, you can get that one for like a quarter, and there are lots of other processors you can get for like under $0.10. It's just incredible. I mean, we're talking consumer grade, where you're ordering hundreds of thousands of them.

Chris Gammell: Oh, yeah, okay. I was confused, but I was like, I'm not paying a $0.10, $0.25. No, no.

Alicia White: You have to order a few more and not get them on the dev board.

Chris Gammell: Right, right, of course. Yeah. Yeah, they always sneak the better ones on there. Oh, well. Yeah, no, that's a good point, though. And, I mean, it's not like, especially when you already have 32 bits there, it's not like there's going to be an upsize in the future, too. So it's relatively stable on a lot of fronts, and there's a lot of cool things you could do.

Alicia White: Power efficient? Nothing like field updatable. Field updatable?

Chris Gammell: Well, and everything's becoming so much more modular. You know, you can build things with, you know, micro and a couple of I2C devices, and just not that much electronics going on there. So it's kind of, it's a strange new world. Did you guys see that new seed studio, the YO thing?

Alicia White: No.

Chris Gammell: No, it's like a, I mean, it's another ESP8266, but then they use the growth. Yeah. But the crazy thing is, like, the app that they have with it. So it's a little hooked, that hooks up to an app, and then, you know, because they're using a lot of their other modules and they have a pretty big library, it's just like tap and go kind of thing in terms of getting all the pins connected in software, and then you're just writing Python again. And so that's kind of the promise of, like, the IoT thing, but, you know, also the, I'm sure that there's other layers in there. But, you know, for getting started, it's pretty great. Yeah, hopefully that doesn't take off before I retire. It might, but I don't know. Like, I don't think you've both made your money on, like, people using apps to hook things together and write Python, right? No, but if it gets too easy, you know, it's just not worth it anymore. Yeah. Well, true. Yeah. But then you can manage people while they tap apps. No, no, no. We've done that. Been there, done that.

Alicia White: It's super fun. I'm always hoping people will need lower power. So, you know, I can always optimize and get rid of all of the fluff on top and, you know, like Python and interpreters and RTOSes.

Chris Gammell: Yeah. Yeah.

Alicia White: Yeah, they're 24 bits, so you have an 8-bit processor in there.

Chris Gammell: Well, the money's always in the corner cases, right? So that's where I think you guys will do okay. So we should kick in to start listening to questions and pretending we know what we're talking about, huh?

Alicia White: Sure. So our first guest, I wanted to make sure that people didn't miss the Aussie accent. So I have Steve Dalton as the first guest. Do you want to get him on the line?

Chris Gammell: Sure thing. Hello, Steve. Hi, Steve. Hello. Where are you calling from? I'm calling from the sunny Gold Coast in Australia. That's the top part? Is that right? Sorry. The bottom of the top part, if that makes any sense. The bottom? Okay. Yeah, cool, cool. Yeah, it's pretty nice here at the moment, looking out on my swimming pool. So what are you working on? What do you got a question about? Yeah, well, my question is actually around, it's not a technical one, but it's a bit of a pet topic for me because I help run a hackerspace here in Australia. I know a lot of hackerspace members that have kind of made this leap into the pro maker world, for want of a better word. But yeah, one of the things I've seen is a fair bit of negativity from people in traditional electronics and embedded software industry towards us. You know, all the usual arguments, we're hackers, we don't care about testing, industry standards, you know, regulations, which is kind of partially true. But also crazy things like, you know, oh, well, you don't use Altium, so you can't be any good or you haven't been trained to use the equipment properly. And on the flip side, we see people who just kind of see us as magicians. You know, they can whip up anything in a week. So really my question for you guys is, what do you think of pro makers? And do you think there's an intersection between pro makers and mainstream industry? Well, first, let's get an idea of what you mean by pro maker. I mean, I have some idea, but you probably have a specific idea. Yeah. So I guess one definition might be, you know, people that are making things to sell to other makers. But what I'm talking about more is people who've kind of, you know, they may not have a traditional electronics engineering background, but they've learned a lot of skills in hackerspaces. And then they've seen problems out there and then tried to build a business around it. So people who are making a business out of by using that maker skills.

Alicia White: There's actually a book called Maker Pro that is about the people you were talking about, the folks who are...

Chris Gammell: That was David Lang? Is that right?

Alicia White: It was a series. I mean, it was a bunch of essays from different people.

Chris Gammell: Oh, okay. Oh, yeah, I remember that one. That's right. Yeah.

Alicia White: Sophie Kravitz contributed to it.

Chris Gammell: Yeah. I think, I think though, the term maker is pretty loaded these days. Yeah. But, I mean, at the end of the day, it's, it's really people starting businesses and it's just that they start to edge into people's, you know, expertise areas and then they get a little uncomfortable about that. Oh, you're not doing it this way. You're not doing it that way. But I think what ultimately, I mean, like, so we were just talking about the, the dropping cost, not the dropping cost, the, the rock bottom prices of micros, what, you know, that enables you to do a ton of stuff. You could throw a micro at a, at a problem that you could have done with a, you know, a couple of transistors or a, you know, a 74 series logic chip or something, you know, and like, yes, those are all ways to do it, but sometimes the, the way that's fastest is going to be, and, and most accessible is going to be the one that gets it done. And then you hire people like Chris and Alicia to, to make it like super huge for, you know, if you wanted to go to full production type thing. Right. Yeah. I have mixed, mixed feelings about it, Steve, because, you know, people do have new things that they come up with and come with different perspectives. So somebody from a hacking or a making background, you know, they may have ways of doing things that traditionalists really haven't thought of. On the other hand, there's reasons for the traditional solutions being the traditional solutions. I mean, you, you mentioned testing protocols and things, and, you know, that's, that's going to be the difference sometimes between a business succeeding and failing or moving beyond, you know, five and 10 units to a hundred thousand units because you're getting sued. Yeah. Because you can't afford to make mistakes at those, at those levels. And so I can see some of the pushback there having some merit. On the other hand, there, there's probably a lot of it. That's, that's sort of cliqueism and protectionism where it's, Oh, I don't want, you know, people without a traditional education. That doesn't seem fair that I went to school for 18 years and, and you're just making something and look how well you're doing. I bet there's a little bit of that.

Alicia White: I can see both sides. I feel like I've been both.

Chris Gammell: Yeah.

Alicia White: I mean, I've definitely been a very traditional engineer college and then HP and then FAA and FDA certified products. So I definitely gone through the process of making a solid, super solid, robust system. Even to the point where mistakes would cost thousands or hundreds of thousands of dollars. And while that didn't come out of my salary, I felt very responsible for them. But then lately I've had a lot of projects where it's just build us a prototype. And I'm not enough of an electrical engineer to be able to do more than go to Adafruit or spark fun or whatever, and pull all these modules down and hook them together and then sort of feel out where I need capacitors and resistors. So in that case, I feel like I am being one of those maker pro people where I'm doing the making thing and I'm doing it professionally, but I admit that I'm only making a prototype. And if you want to go to production with this, I want you to find a nice double E who can vet this system.

Chris Gammell: we all do that too with software too. I mean, if we're doing something quick and dirty, we all, we all feel the things we're doing wrong as we do it. Yeah. Sometimes we're ignorant of it. Let's be honest. I speak from my, my own experience. I think, I think another thing that Elisa kind of brings up there though, too, is just the, the concept of ego, right? I mean, so if someone comes in and they're like, well, I know how to do this. I did this with an Arduino. That's a lot different than like, well, I made this thing like Elisa's doing. I made this thing. Let's talk about the realities. If you want to make a lot of them, this is, you know, we should probably talk to other people. And that's, it could be where some people have experienced, uh, that tension between, you know, traditional and maker pro type people. But I think it's, it goes both the ways because I've, I've known some crotchety old people who, who are traditionalist for the sake of being traditionalist and don't even try new things. And that's like, whatever you're going to. Okay. Enjoy. You know?

Alicia White: Well, I think, I think it was actually Neil Peart who said that the difference between an amateur and a professional is that an amateur stops when it works once and a professional stops when it never fails.

Chris Gammell: I can't believe, you know, a Neil Peart quote that I don't.

Alicia White: Uh, I didn't get the words right, but the drummer. Yes. Like he writes really good books. Okay.

Chris Gammell: Yeah. So Steve, what do you think? I mean, is that, I mean, is, is that, uh, close to what you were thinking or no? He's, he's not even there.

Alicia White: I think we'll call that a win then.

Chris Gammell: Yes. All right. An amateur hooked up this show. So blown away by, by our answer that he just had no words left.

Alicia White: He did awareness his networks. Welcome back.

Chris Gammell: Yeah, that's all right. Aussie, Aussie broadband. Ask Dave about it. Uh, yeah. Uh, did you, how much did you hear? Did you hear what we were talking about? I got most of that. Yeah. I just had a comment.

Chris Gammell: so yeah, I, I had a traditional electronic engineering background and I, you know, I did a graduate training scheme. I worked for three com in the UK, you know, before they went earlier and, you know, I really appreciate that time that I had and I got to do things, you know, and be taught things in a certain way. But, uh, you know, those, those opportunities don't exist anymore for a lot of people, especially here in Australia. We don't have the big companies that do that. Um, so I've had a lot of fun just in the maker world, making things and, you know, a lot of the things are one-offs or their, their prototypes, but actually building things very quickly and actually getting, you know, some immediate feedback, which I never got in that, um, you know, the big company world. So, um, I think there's a lot, um, there's a lot of satisfaction you can get from that, you know, that you sometimes miss out in the, in the big world, big, bad world. Yeah. The dopamine squirt of, Hey, I made a thing. Totally agree with that. Yeah. The first, the first six weeks at startups have always been the best part of my career. That's so fun. And I think we've got a ping pong table and a shitload of ideas. That's right. And a prototype that barely works. If you just touch it this way, it works. But if you touch it that way, it doesn't. One of the best things we, one of the best things we ever built was built in under a week. And it was a one-off for a trade show. And it was a giant, uh, monopoly board that controlled this huge screen. It was for, uh, you know, a tourism and trade show. And it was literally a single Arduino inside with some ethernet cables. And we made the buttons on our laser cutter and we did it in less than a week. And it, and it only worked, it was only needed once it had to last two days and then it was all ripped apart. But it was like one of the best things ever. Cause the screen it controlled was like massive. And it got on the TV cause it was like the best thing at the show. They actually did a little thing standing in front of it, pressing the button. So it was very simple, but you know, awesome. That's great. And, uh, and what's, what is, uh, what is the name of your program again? The one you, you work on? Sorry, what's that? Sorry, the company. What's the, what's the name of the program you work on? That works with the maker prototype folks. Yeah. So, uh, Gold Coast tech space, which is a hack space on the Gold Coast. But, uh, you know, we do cool all sorts of bits and pieces. We do a lot of stuff with kids and educating kids and moms and all sorts of stuff. So it's been fun. Very cool. And we should also point out that Steve is not a plant, but, uh, he has been on the embedded before. Uh, what episode was it?

Alicia White: Uh, 69. I believe it was dingo rabbit death match. 63. 63. And yeah, he told us a lot more about the Gold Coast, uh, tech space, which is pretty cool.

Chris Gammell: You can go back and listen to that one. That's great. Cool. Thanks guys. Thanks for calling in, man. Thanks a lot. Cheers. Uh, well, I'm glad you were able to get him back. Uh, we could just, we just like went the rest of the show and we're like, Oh yeah, we're going to talk to people.

Alicia White: We didn't actually have to answer the questions, right? That wasn't the goal here. Was it?

Chris Gammell: Come on. It is technically the amp hour, right? I mean, this is, so it's, we don't, we don't deal with answers too often. Uh, uh, who's up next? Uh,

Alicia White: Lee Wiggins. It's another Aussie. Another one. I know. Man. Everywhere. Yes.

Chris Gammell: But mostly Australia. That's right. Although, uh,

Alicia White: I do sort of fear that we should have had Lee talk to Steve instead of to us. Cause it seems like his question may be, well, let's get him on the line and find out what his question really is.

Chris Gammell: We can always call. We can always answer a different question. Yeah.

Alicia White: This is like car talk. We'll ask him about the relationship. If we don't like the question. Okay. What you do is you wait for him to answer. And then you say, hi, I'd like to order a pizza.

Chris Gammell: G'day, Chris. Hi, I'd like to order a pizza. Sorry. My fresh air. Delivery time. It'd take too long. Anyways. Uh, welcome. Lee, where are you calling from? Yeah. Thanks, man. Yeah. Lithgow, New South Wales. Just need to blue mountains. How's the audio. Is it all right? Yeah. Yeah. You've got a, you've got an accent coming through there. Yeah. That's cool. Yeah. It's kind of heavy. I'm going to say, sorry, can't help that. Comes, comes with the trade. So, uh, Lee, what kind of, how's it all going? Good. What kind of work do you do? I'm a, um, power line designer for Endeavor Energy.

Chris White: So, yeah, designing overhead systems, 11 KV, 22 KV, underground, anything that sort of comes my way. It's, yeah, doing a lot of steel mains replacement at the moment. So, it's a lot of rural stuff that's been good.

Alicia White: He's saying KV, right? Like, kilovolts. Like, kill in kilovolts. Yes.

Chris Gammell: Yeah.

Alicia White: I hope he's got an electrical question, because there's no way.

Chris Gammell: That's not that bad. Well, he's, here's, here's what I'm thinking. He's alive, and he's talking to us. So, he's pretty good at what he does. He didn't say how long he's been doing it. Yeah,

Chris White: haven't been got yet. I'd like to keep it that way too. Don't get me any second chances.

Chris Gammell: Yeah, man, that's nutty.

Chris White: Yeah, the high voltage, we've got a low voltage rescue kit that we use, that we have to put out near the poles when we're working low voltage live. But, the high voltage rescue kits are dustpan and brush.

Chris Gammell: So, that doesn't fill me with confidence. And low voltage is what? 600 volts, right? Yeah, 240 to 415, three phase. Jeez. Just street stuff. Yeah, just the boring, the ticklers, you know? Yeah, yeah, yeah. Gives you a mustache at night time, you know, that sort of stuff. Yeah, yeah, that sounds right. So, Lee, what kind of questions do you have? Well, I've been doing, obviously power is my thing. I've got advanced diploma in electrical engineering and doing my degree, but I'm currently doing the degree and full time and kids and family and all of that sort of stuff. So, I mean, it's mainly concentrated around power, but I love to do electronics as a hobby and have been doing it for a while.

Chris White: And a mate of mine, Toby up the mountains, actually put me onto Arduino quite some time ago and have been doing a few projects ever since.

Chris Gammell: And recently, I put up a dummy load, which has been really successful on Instructables. And it's got 60 odd thousand views now. And Dave did a really good review of the PCB, which I was really happy with. And it's just starting to, you know, go into the world of all the projects that I'm doing and starting to get a little bit beyond the Arduino processes as such. Not so much the IDE, but the processes that they've got. And I just wondered when the right time is to step out of the Arduino world and where to jump to from there.

Alicia White: Oh, I have so many, so many, so many suggestions on this.

Chris Gammell: I, I have one.

Alicia White: Will you go first then?

Chris Gammell: No, it's just, you know, I, I don't actually have one.

Alicia White: So, Arduino is great. And you said you're, you're okay with the IDE, right?

Chris White: Yeah. Like I'm, I've actually got a web development business that I run as well. So, I mean, coding and IDE and, and, and that side of the software side of things is not really a big issue for me as such.

Alicia White: Okay. Some people hate the IDE and they can remove the IDE and go to GCC and have a good time with an 8-bit processor because the Arduino is so ubiquitous, has so many things it can play with. But if you like the IDE, okay, but are looking for a processor that can do more underneath there, you have a ton of options. The first one I would recommend would be the embed system. That's from ARM. And so it's all the Cortex processors. Cortex M0s with their little low power, the Cortex M4Fs with their floating point and high, higher capabilities. And it's still online. It's still free. And it still has somewhat dubious debugging ability, which is one of the downsides in the Arduino IDE.

Chris Gammell: You hate debugging. We have a platform for you. Yeah. Debugging on Arduino sucks. Yeah. Debugging. Yes, it really is a nightmare. I have had a bit of a look at the ARM processors, but to be honest, it took exactly where to start and the ideas for developing them and physically how to program them. You know, I was finding it hard to get some information on because Arduino, there's so many great tutorials out there and there's so many marvelous things. And I've seen the F4 discovery boards. But it's even, you know, there's so many, I guess, tutorials on how to get started. But it's certainly not an easy thing. And I'm not so much concerned about me, but more of the people that are working with the projects that I develop, you know, being open source stuff. And I want it to be easy and doable for them as well.

Alicia White: And that's really why I say embed because it is free. It's online. So you're not worried about other people's operating systems. A lot of people are using it as the next step beyond Arduino. As for programming it, most, all of the boards that are marked embed ready or embed.org or have an embed in the name. They plug into your computer and act as a thumb drive. And so to program them, you simply compile. And sometimes you drag it over to the thumb drive and sometimes your browser will do that for you. But it's, it's sort of like the Arduino that you don't even really notice that it's putting code down there.

Chris Gammell: Yeah. Okay. And that very good.

Alicia White: It's really easy to use. And like the Arduino, there are a ton of other people doing this and it's mostly open source.

Chris Gammell: There's a lot of libraries out there for various types of peripherals and things. And, you know, they're more powerful than Arduino. So you can do things like, you know, Wi-Fi and networking, full networking stacks and things. So when it comes to, you know, programming the devices outside of the discovery boards and, and outside of the development platforms that they're in, say, I just wanted to grab an ARM processor. Let's say I picked an M4 that was, you know, suitable and to the needs. And I wanted to put that in my project. Will, how do I then program that? Is it something that you, like, you know, obviously I've got the ICSP with Atmel that I've been using, but is it similar for ARM?

Alicia White: Yes. There are a number of different programmers and they cost between 15 and $70. And if you get a dev board, I mean, some of them cost a lot more than that, but realistically, the home ones are 70 ish. Yeah. If you get a dev board, that's usually part of the board.

Chris Gammell: Yeah. Okay.

Alicia White: And sometimes you can even break that off and use that on other boards.

Chris Gammell: Yeah. I'm doing that right now. I have, for a client project, I have a dev board for a completely different chip and I use that as the programmer for something else. Yeah. Right. Yeah. Very good. Yeah. Okay. So it's not such a big step. It's, it's not such a big step. If you take the right step, if you try to go, uh, directly to, you know, standard production methods and pick a chip and build a board and pick flash to go with it, and then buy an IDE that, you know, that's way out there and much more complicated, but you don't need to go there. Yeah. Yeah. Just yet. What is, what is the end result of what you're, so this is still going to be kind of a hobby style projects or do you want to actually kind of maybe get towards making a device and deploy stuff into production like that? Well, at the moment I've got my dummy load that I've been working on and I'm sort of up to rev C at the moment and I'm hoping to do a Kickstarter on that, um, and do some improvements and the, and the rev B one's been very successful. I've been selling just the PCBs and a lot of people have been buying them and building the kids. And a lot of people have, you know, been posting that on the Instructables, which is awesome, which is what I love to see. But yes, ultimately I think I would like to go, you know, not go into business and, you know, give Agilent a run for their money, but I would like to just, you know, create this sort of stuff. And the next step, you know, I'll probably keep my dummy load with Artwinner because that's where it started. But my next step is to try and, you know, build a power supply or something similar to the Rigol. Is it 832? Something similar to that.

Speaker ?: Um,

Chris Gammell: but make it open source and hobby level and, you know, affordable for a lot of people. I'd say don't, don't underestimate the ability for your customers to tinker and create community around it too, because that, that'll have a lot of benefits just from like, oh, Hey, if you wanted to do this other thing or have this improvement, you know, that, that can have a lot of benefits where if you close it off and, you know, do a traditional like, oh, well there's one firmware version and you can update it. that makes it a little bit tougher. Yeah. Yeah. That's true. I think embed is still in that same family where it said, if you can, if you can plug in and then maybe even easier because then it's just like the board plugs in, you drop a new file into this file system, like Alicia was talking about. And then it's, it's, it possibly is even easier because you don't even install the Arduino IDE then. So. Yeah, true. That's true. Um, but it's, yeah, it's,

Chris White: it's something that I, I certainly want to keep open source and certainly want to keep hobby level. And, you know, the code for my dummy loads been on GitHub for ages and I've had three or four or even five actual commits that people have worked on and improved. And I really liked that and I would like to keep that going. So I'm glad to hear that, you know, a lot of people are on the embed system and,

Chris Gammell: and going for the, um,

Chris White: type processor and that it's not going to be a big jump for a lot of people. That's really reassuring.

Chris Gammell: Yeah. The only other step that I would, the, the one thing that I was going to suggest of course, this was my, uh, we made the same decision in contextual electronics and ultimately we said, well, we didn't want to go all the way to like what Chris was talking about, where you're buying a full IDE, like usually like a Kyle or something like that. And you're buying, you know, this really good software. Yep. So we actually kind of struggled through the, the eclipse and then using the arm tool chain and stuff like that. And open GDB and stuff like that, or whatever that's called, uh, OCD. Alicia. OCD. Thank you. Yeah. Uh, yeah. So, uh, you know, like that, that's, that is a struggle, but then there's, that's, that's to get to the, the troubleshooting level of side of things. So. Yeah. Right. Um, the other thing I've been looking at a little bit lately, just out of interest rather than anything else is, is DSPs and sort of where they sit in the, in the grand scheme of things. And whether I should be running my displays with a DSP or whether I should be doing it with an arm level processor or, um, from my understanding or very limited understanding of DSPs is that more sort of used in audio and video world of, of file conversion from audio to digital and not necessarily for driving displays, but do they sit in that sort of world? Should that be something I'd be considering to run a display with? No, I don't think so. I wouldn't either.

Alicia White: DSPs are really, really cool, but don't use them for what you've, you're better off using an arm for its general purposeness. And some of the arm controllers have LCD controllers as part of them.

Chris Gammell: Yeah. Okay. So look for that. Yeah. Or use spy based and just, you can just use one of the ports and like, yeah, there's, there's really a lot of options there.

Speaker ?: I,

Chris Gammell: I, I,

Chris Gammell: I,

Chris Gammell: I,

Chris Gammell: You can do a lot with a spy display. Oh yeah. yeah well i actually i love spy um and i've been playing with um a 2.8 inch tft at the moment on spy and i've just put it onto parallel actually and and on the you know the 16 megahertz um that i'm using it's it's just slow um but you know and i guess i'm not sure of the maximum spy speed that i can use on this particular one but i may be able to improve it but yeah just just the speed and the flash size and the peripherals and the usb and the lcd peripherals is the main reason for sort of getting away from you know the not the arm the um what are they called the mega type chips and the arduino style that's that's smaller 16 megahertz type processors the 8-bit processors up to the 32

Alicia White: you know we talk about the arduino as though it's a really functional thing and it is for a small number of of projects but when you look at a cortex m0 the very bottom of the cortex m0s the very top of the at mega chips they don't overlap i mean you get a lot more in the cortex m0

Chris Gammell: yep and if you go to an m4 you've got floating point and dsp and a bunch of other 60 megahertz

Alicia White: yeah and the chips by themselves the cortex m0s are cheaper than the at megas

Chris Gammell: yeah for the other i just can't get that was another reason i just couldn't get past you know why why i haven't done this and you know that's another reason that's sort of been pushing me to go that way of of having a very quick look and you know the i guess configuring the usb interface and um all of that stuff myself is a little bit i guess daunting to start with because it's just so easy you know you push a button and it's there but um a lot of the resetting the registers and a lot of that sort of information there's a lot of information out there to read on that i guess well for the

Alicia White: embed boards initially it should be just as simple as arduino i mean you shouldn't have to do anything with the usb it just pulls down and suddenly you have a new thumb drive yeah it's kind of magical

Chris Gammell: yeah cool cool well lee thanks for calling in man uh that's cool thank you very much for the opportunity and yeah it was really good to talk to you both and all three of you and you know shame dave's having a bit of a lazy attack and on vacation but you know we can't can't have it all can't stop aussies from taking vacations that's what i've learned it's uh it's a right right

Alicia White: i was wondering about his vacation and then i realized that it's the other season there it's

Chris Gammell: not dark and cold where he is yeah yeah actually we've probably just had about nearly 150 miler rain nearby so yeah a lot of people need it all right well thank you very much thanks again lee we'll talk to

Alicia White: you soon thank you thanks okay the next uh person on my list is stewart mcandrew he's another aussie what is happening wondering what is going on with this list and i have to warn you that stewart and i got in a battle on twitter who could come up with the worst ocean puns and i i i give in i mean i i i ceded the battle to him uh-huh but i i just want to warn you before you call him that whatever happens

Chris Gammell: here is probably my fault okay now i'm gonna try and think about a as one of my own i have to get on pun battles all right here here we go hello hey stewart how are you doing hi good good greetings hi stewart hello hello hello so where are you calling from i'm calling from perth western australia a long way from nowhere oh the left side yeah that's it yeah wow that's uh different than most aussies that we talk to and you're the third aussie in line did did anyone tell you you're gonna be the third aussie in line here i did and and i think there's more to come oh seriously oh geez it's all right it's a good it's right in the middle of their day dave's not here is he no no that's good so i heard uh that you and alicia were having uh ocean puns is that right yeah yeah uh i think chris started it me other chris that doesn't okay that's right yeah no it was it was you stony monster chris i did yeah yeah had to do with a dolphin picture oh well maybe i started the pun no i started it so what was the best one then uh i don't know if there's best when it comes to yeah you can't you can't grade them like that chris okay what was the worst i think it was when i

Alicia White: finally like admitted that i i was done and i couldn't go on and and and then he he okay maybe i did the prawn one yes but then stewart said he didn't harbor any ill will so

Chris Gammell: it had gone to that point okay yeah that's usually where the ship runs aground okay so stewart what what uh what is your question uh for us so i just wanted it was just a sort of a

Chris White: general sort of discussion point as uh talking about watchdogs um you know which uh you know which is better beagle or alsatian and sorry i had to slip in the puns um no external versus internal sort of the the pros and cons of of each type and um i guess you know which which one should be used

Alicia White: in in each situation well internal is cheaper because it comes with your processor external is a nice

Chris Gammell: fail safe the external there's a like a separate chip is that what we're talking about there yeah that's right a lot of them a lot of them are coupled with voltage supervisors so um ah okay i've used one of those and you those are nice because it doesn't depend on anything you're

Alicia White: doing in the processor your software can't disable it it happens whether you like it or not so if you have a safety critical system that you don't want any dependence on software for like faa or fda or space systems an external one has some good advantages you really it's gonna happen how do you how do you

Chris Gammell: keep that alive then you just like pull down on a pin every every couple exactly every 10 microseconds or something yes okay the downside of course is if you don't want it to work then you have to if you

Alicia White: want to update your firmware and you need a whole second to do it instead of the three quarters of a

Chris Gammell: second it gives you then you're they have disables on the chip or no yeah most of them don't have they're they're pretty the timers aren't very long there's you know you know some do have a you know multi-second sort of interval but um a lot of them seem to be fairly short yeah where uh i can't imagine so what are the reasons to not use internal again because that's usually i mean that's the ones i've seen and you know i see like you always disable them when you start debugging that kind of thing because if you hit a break point you're like no no no right well that's the very reason is they can be disabled so that if you're going through a safety critical product uh it could be argued that well what happens if your software screws up and disables the watchdog and then something goes wrong and you don't catch it or you write a bug that disables it and you forget to re-enable it with an external you don't have any choice right you know anything that happens with your processor is being managed from a different place entirely um so it feels like a more secure system have firmware people considered to not write bugs yeah that's been mathematically proven to be impossible

Alicia White: the external uh watchdog also gives you a power cycle where an internal watchdog may give you a soft

Chris Gammell: reset and so it's a different start state yeah okay yeah but don't don't uh watch the internal watchdog sometimes like pull an external pin and then you can do a reset yourself yes there are a huge variety

Alicia White: of internal ones yeah okay um most of the cortex though is a soft restart start uh which i have i have to admit i don't really like i wish it would do the external pin thing because a soft restart doesn't solve all the problems right yeah but an internal one is cheaper it's a lot easier for the software engineers to deal with most of the time that's fine and yeah it's mostly fine if you use an external one then you do all your development without it because if you want to j tag that puppy you can't have an external um you end up with the jumper to turn it on and off right and then someone forgets

Chris Gammell: to put it on right the jumper vibrates off you're like oh boy or you forget to disable it and then wonder why your processor keeps resetting because you haven't you haven't written the code to toggle

Alicia White: the pin but i mean this does also bring up how do you treat a watchdog and um i was talking to someone today about watchdogs and they were saying about timer interrupts and watchdogs and i was like that's a really really bad idea and i wasn't sure if the person i was talking to understood that that is a really really bad idea or whether he was trying to be clever because so the goal of a watchdog is to make sure your processor is doing what it should be if you have a timer interrupt that kick kicks or pets depending on which generation you're in uh your watchdog every 60 hertz or 60 nanoseconds or whatever you choose for your timer then it doesn't matter if your software is wedged as that only tells you that this high priority interrupt is happening it doesn't because an interrupt handler always is going to run unless things are totally hosed yeah unless you're other unless you're hosed in an interrupt and so you don't want to put your watchdog interactions in an interrupt what you really want to do is put your watchdog interactions in the lowest priority thing you're doing like your while loop so that you know it if that isn't happening because your interrupts are wedged or because you have some other thing you're waiting on a semaphore or whatever then you reset because the lowest priority thing has

Chris Gammell: to be the thing that's going if you expect a stable system yeah there's lots of ways to make watchdogs completely useless they really are i think um jack gansel wrote a good article on on watchdogs and and sort of you know where to put it um in the code so uh starting a sentence with jack gansel wrote a good blank is usually true yeah yeah yeah um but i think he i guess uh said what you said at least you just you know where to put it in the code you know don't don't put it inside a loop because it may just continue on that loop and never get out of it um what do you what do you steward what are you using this for what do you what do you does it involve space it does involve space yes yay um so uh as part of the little satellite that i'm building um i'm using a watchdog so i'm using an external one because i need the ability to i guess remove remove the power so um that's where i'm that's where i'm using in this instance but i just wanted to i guess discuss you know the benefits of of using either either way external and internal so just be sure that when you uh when you hook it up you don't have like when when the watchdog fires have the the piece of thing in space re-enter that'd be bad no there's no propulsion that's right i'm just just depending on on gravity and and inertia so hopefully once it's in orbit it stays there uh and only only slowly comes down good yeah that'd be

Alicia White: really good so there are some externals that don't start watching until the first time you interact with them and that lets you update the code and lets you have a little more freedom less safety but more freedom um are you using one of those are you using one of the old school you have to boot within a certain amount of time or you may be stuck in a watchdog reset loop i'll have to check the data

Chris Gammell: sheet um there's usually a an initial uh timeout period before the watchdog timer kicks in um and i don't know whether that's hard coded in into that watchdog or externally settable via resistor so i guess read the data sheet i'll have to figure it out be sure to enable it early yeah do you have like multiple versions of backups up there like multiple watchdog type things for for a piece of space gear i know that there's usually a lot of redundancies but i don't know how much there would be on this kind of thing which is pretty destructive we should ask him to tell us about what he's building yes that's right okay so um i'm building a pocket cube satellite so you probably haven't heard of a pocket cube it's it's pretty small it's it's smaller than the planet labs doves you've you've had the planet labs guys on the on the show alicia um talk about their their cube sats now they're 10 centimeters by 10 centimeters by 30 centimeters long so mine's a lot smaller it's only five centimeters by five centimeters by five centimeters so it's pretty small so there's not a huge amount of room for redundancy and sort of additional systems so it's it's a fairly simple beast and it's what's it up there for is it doing the same thing of like taking pictures or is it doing something else it's not up there yet it's sort of a i guess the the project in in progress but it's going to be a earth observation satellite so it's going to yeah take photos from space and it's like small enough so you can just throw it out there or or how does it get up there pretty pretty much um just throw it really hard for a big yeah exactly yeah or like a cube sat gun i don't know would you would you believe that the russians have actually deployed satellites from the space station by throwing them i do believe that i've yeah yeah yeah um so for these ones they're they're very small so they actually at the moment there's only one way to get into space and that's on board an italian satellite and so the italian satellite has the little you know cannons to to shoot out these pocket cubes and that italian satellite thing goes on board another rocket and at the moment uh the planned rocket is a russian rocket so yeah depending on funding is whether whether i actually get on board that one or not cool and this is all a personal project right yes yes it's just uh i guess a very strange hobby not there's not many people out there that are uh building satellites in their spare time so uh it's just i guess a project that i sort of decided to take upon myself to to do i wanted to uh i guess after listening to your shows and hearing about all this hardware prototyping i thought you know it's something that i can get involved you know doing so i wanted to make my own circuit boards and i figured a satellite's a good project as any to to start off with awesome i would disagree with did i mention i'm gonna build an electric vehicle i just uh why not just the hardest thing i could possibly think of yeah i imagine that at one point you're like i want to build electronics and then i want to get as far away from them as i possibly can yeah maybe it's a yeah i guess it's a bit ambitious but um i mean when you when you sort of engineer it and break it down it's it's uh the skills are out there and and you know if i just look at all the code that people have done with arduinos and and other microcontrollers and you combine with a you know the ham radio guys and the the uh amateur sort of balloon community they've already written the majority of of these bits of code so so you can sort of take bits and pieces from you know what these guys have developed and sort of you know just roll it up together just in a in a different platform and just so happens that this one is a little you know tiny five centimeter cube that's going to go into space and take photos sounds simple doesn't it yeah oh yeah it's something inspiring i mean i could see how

Alicia White: i mean all the personal projects there's always some bit of drudgery or you know just some point where you

Chris Gammell: stop and you think this isn't worth it but that's usually when i have to do anything mechanical yeah

Alicia White: yeah um but i can see how it's going to space would get you over some of those doldrums my my widget in space i don't know it just has a nice motivating factor to it yeah i mean

Chris Gammell: or maybe you just want to get far away from it no i think that even even sort of when they talk about space and governments you know no one sort of has a anything bad to say about it it's it's it's for the for the benefit of of mankind now fair enough there's you know there are commercial companies that that are in it to make money but uh generally when you're engineering something for for space it's sort of there's always a positive sort of spin on that unless you're the military right not you're not going to be doing a lot of mining with uh five five centimeter cubed uh or whatever the cubic version is but you can do you know remote sensing so you you can because they're you know smaller smaller and cheaper you can make more of them and uh you know spit a whole bunch of them around asteroids to to measure it um so that's that's sort of where where my goal is for for this is to eventually you know produce something that does more than just take you know low res pictures well stewart thank you for calling in we appreciate it and uh good luck with your project that's all right thank you very much and uh i guess i'll i'll keep listening to the show it's it's always entertaining i think you know even even though there's there's not a huge technical content every week uh there's always something that you can just learn from listening to people talking about you know their projects and and the things that they've they've learned and their experiences so it's uh always good to listen to other people we are the online water cooler excellent that's good that's good thank you take it easy all right thanks guys all right well that's cool i want to build something cool once again i am made to feel inadequate by our listeners or guests

Alicia White: so next on the docket is jonathan mcdonald his question's hard i was thinking no i know before

Chris Gammell: that too i was thinking um we need more like we've had like good these have been great questions so far but i feel like there's not enough like hardware versus software contention you know what i mean like the where should a debounce happen like that's what we should have started with right yes if we ever do

Alicia White: this call-in show again we we want you listeners to think about all of the times that as a hardware engineer your software engineer has been wrong about and as a software engineer your hardware engineer has just been a jerk about and call in and tell us about those so you can start the war

Chris Gammell: finally i don't want to start a war i just i just think that that's where the interesting stuff lies i mean mostly what i'm really saying here is i have no idea the answer to some of these questions guys here's here's a question i so i used to have i used to work with a really good firmware engineer who was just like gray hair you know old school you know wrote everything assembly awesome really fast code dude however when we'd be booting up a board that i designed he would turn to me and he'd say uh chris there's noise on this board do you guys ever say things like that not recently but yes and i'm like tom you mean it's not working right now he's like yeah there's noise on the board i'm like do you have any idea what you're talking about right now i mean like he was like 30 year like former ee kind of thing like kind of did but i'm just wondering are there uh are there things like that that you say to the hardware people that the hardware people make funny

Alicia White: faces oh there were so many things that i used to do and say so the before so i got out of school i was a software engineer i definitely cs and i had a lot of uh signal processing math because i was in this weird major that did engineering not specific just general and so i had a lot of systems and so i i went to hp and and got further further down into the hardware layer but i really wasn't the hardware engineer and i got transferred over to become a real firmware engineer really neat team but i was it was like my first day they sent me down to test boards in the heater or in the oven oh yeah right and i i didn't know i mean i didn't think about hardware enough and so i i set the board on the oven floor and turned it on

Chris Gammell: okay i mean i that's i think that's fair if you wouldn't have done it before

Alicia White: i mean why didn't he give me standoffs but yeah so i think it's on him for not like for just assuming

Chris Gammell: but maybe you should have given your background a little bit more but i don't think he listens to

Alicia White: your show so i still don't think he knows what happened that day and he asked me later was it something about the board or or did you make a mistake and i said oh no i made a mistake i don't think there's anything wrong with the board here but yeah um don't set boards on metal things if they have metal pokey bits out that was my first electrical lesson yeah i think i think the thing that uh i most

Chris Gammell: commonly irritate double e's with is i have no respect for static oh or rather i don't believe it exists i just you know nothing has ever gone gone wrong on a board because i've done an esd uh discharge and so i just don't believe it happens so i don't handle them with care he's sitting here

Alicia White: in socks on a wool rug surrounded by electronics on a winter day with i i've been fine i think it's a

Chris Gammell: conspiracy of electrical engineers to sell those little things on your wrist that's right they're just fashion accessories guys they are i don't know i mean honestly it's gotten a lot easier not it's not i'm gonna get myself in trouble here but it's gotten a lot better like you know it used to be really bad with like just when people were using jfets there's no no diodes anywhere you know

Alicia White: it's like it's gotten better but yeah you should care about the processors are a lot more tolerant

Chris Gammell: to handling than they used to be right yeah that's and again that's that's the diodes on every pin kind of thing they're super they can they can take a lot of crap so um yeah i bet i bet i was gonna i was thinking of what what i say to software firmware people that pushes them off um i just ask them to recompile i think that usually gets them going pretty good just do it again i think it's the insistence you know usually the thing that gets me is the insistence that it's a software problem and probably conversely the thing that gets selected yeah it's like my first instinct is well there's something wrong with the hardware that usually usually there isn't i think it's it's good to just keep a tally you know that that keeps people in line right you have a little tally on the wall once it gets past maybe 75 towards one person or another you know then you you have to buy lunch i

Alicia White: don't know so you keep a team together buying lunch really changed my career that when i started you know having lunch with the double these just being able to talk to them and say come by maybe i didn't hook up the o-scope in the right spot and you have a test point i didn't notice and they taught me so much and it stopped becoming a throw it over the fence sort of problem and started becoming a team sport and i feel like i sort of grew up where it really isn't it doesn't matter who's wrong it just

Chris Gammell: matters that we ship except yeah that's right except that they're wrong well but let's go let's be honest like there shouldn't be that much tension because it's really firmware and electrical people against the mechanical people let's be honest right mostly you guys we don't care so much that's true that's true you can fit software anything that's right yeah well that's good though i mean these are these are things that we need to work out as a community you know this is where the crossover happens this is when george jetzen and fred flintstone are figuring their stuff

Alicia White: out right yeah yeah who's who never mind i don't want to know yeah okay yeah next caller we still

Chris Gammell: have two more to get through hello hey welcome what's your name where are you from uh jonathan mcdonald from um newcastle australia seriously another one really that's four now what is happening man what is going on here i swear you'd think you'd think that silicon valley was in australia it's a conspiracy of australians it's all that it's dave's fault uh hey jonathan welcome i i hold nothing against australians it's just the four in a row come on that's that's a statistical anomaly right must be dave's influence i guess so that's that's great though i mean do you find that there's

Chris White: a good community where you are um oh that's the um local tafe um a couple of people i work with that's about it you said newcastle where's where is that uh yeah uh just above where um dave lives

Chris Gammell: in sydney oh okay okay cool huh i guess that's pretty spread out the city's pretty spread out i mean there's a lot of people in sydney but yeah that's uh that's crazy so what's going on what is

Chris White: your question um i was wondering how i'd get starting with um learning a proper programming language at the moment i've programmed pic micros using flow code and um it's just hoping to explain what that is i i don't know what flow code is yeah neither do i okay it's just a graphical i guess you'd drop boxes that generates c code so you drop like an if statement or a go-to statement or output to this pin um input from this pin you can input an analog voltage or to a 8-bit value in a register i guess and then it generates c code which then compiles to a hex file then and what is the what is the application that's used for um just writing programs for micros i think you

Chris Gammell: can use avrs as well as pics oh oh well i mean like are you using it for like industrial or like uh for scientific measurement stuff or what's what's the industrial it's um industrial yeah it virtually

Chris White: takes the place of an old plc we used to use at work um yeah right we use it to um automate ladder systems on um mining equipment so you've got large mining equipment and um there's an automated ladder that'll fold out in front of the machine so i'd write the software for for that huh why why does a

Alicia White: ladder fold out sorry no no actually you're right the mining equipment is far more interesting than

Chris White: the programming language go ahead no that's um i mean we've um i've used it to control a few lighting towers to um manage battery voltages and just cut lights out if the voltage is too low they're like an led lighting tower um solar trailers um used for radio repeaters so the i don't know i guess it just seems like a really basic programming language to be using when there's um i guess better stuff out there and i was wondering what books were available or online courses well that was sort of my first question for

Chris Gammell: you was what's the motivation to move beyond that yeah um it's it sounds effective to be honest it sounds like a lab view style kind of graphical stuff right yeah more like cubamx that it it starts out

Chris White: graphical and then it goes to c yeah true yeah i don't have any problem using it i was just wondering what other people are using and what communities out there and possibly learning c a bit better i mean we have to have to have to dive in to see the fix flow code sometimes there might be a bug in there where it doesn't set the oscillator correctly in the microcontroller so you have to insert a small bit

Alicia White: of c code in there but that was my question was how often do you read what it generates not a lot fair enough as little as possible probably yeah yeah i imagine doesn't generate pretty things

Chris Gammell: no uh are you looking to do this for for work or is this more for your interest as well uh both okay so i would highly suggest that you don't do it for work to start with yes definitely yeah i mean it seems like especially if you already have a flow going that if you uh yeah if you try and like i would just go i know myself and i would just go right back to the other thing and be like i need this to work right now you know click done whatever and that's why these kind of these kind of in between things exist anyways that's why flow code exists that's why labview exists because it's like you just want it right now um so yeah what's your background is it double e

Chris White: um no no i've um done a cert 3 in electronics um just in the tafe system here so that's i did a small microcontrollers course where we used assembly language to um program run a program on a computer to output an 8-bit value through the uh parallel port cool so not a lot of experience but plenty

Alicia White: to get started you don't have anything to be afraid of you've already done the worst possible thing with the microcontroller and the assembly even if it was small there other things will be easier and probably more fun once you get to a higher level language i think that you already understand too like

Chris Gammell: i think one big piece of jumping into micros as well is just understanding even what you can do and it seems like that piece you already have under your belt as well you know that you can take analog measurements digital measurements uh you know control things time things whatever you want to do there you know and so you already get the the broad picture stuff and now it's down to well how does this stuff differ how can i you know like what kind of what kind of code do i need to write how you know can is there is there a structure i can match and you know play around with kind of thing one thing i would

Chris White: like to implement in code is a an astronomical clock so um you input the gps coordinates into code i'm using a real-time clock to work out when the sun's going to come over the horizon and when it's

Chris Gammell: going to set have you heard of anything like that uh yes in fact i invented that for something a few months back yep for me actually yeah yeah so yeah it's pretty that's pretty straight pretty straightforward um okay just a formula yeah it is actually uh i think there were some corrections and things but it was it was not that you know you have to do a little bit of math and stuff like

Alicia White: that but since we moved to the beach i wanted to be able to know when sunset was every day so that we went to the down to the beach to see the sunset it was important enough to demand that he write code for me despite the fact that i could do this myself this is like the first worldiest problem of all

Chris Gammell: time if it makes you feel any better i i didn't finish so i got it to work i was too busy shining my first edition books and i got it to work once and then i stopped working on it so yeah well and i

Alicia White: needed not only sunset but like an hour before sunset so we had time to walk down there you know it was very yeah it was so first world zero with world i think yep well that's no i think that's uh that's

Chris Gammell: good to have jonathan that's good to have a project in mind as well right that's that's another huge piece is having something to work towards because honestly you know so like like i said you already have a lot of this understanding of what systems can do right so you already have some project ideas hopefully and then the rest is just finding community online and c obviously has a bunch arduino has a bunch you know you could you could go jump into a lot of other you could even go crazy if you wanted to and i'm sure chris and alicia will like this one you could go buy a tesle board and you could just program your micro and javascript if you really wanted to make some embedded people angry sorry you blanked out there uh chris i didn't yeah i think you blacked out actually you just couldn't handle that that statement uh that people are doing that which is crazy but it's it happens and you know so people do all kinds of things they do all kinds of things do you think that jonathan do you want to stay in the embedded realm as well is that something that you want to i mean like that you're going to keep these kind of things or do you want to learn languages that might go higher level um i'd like to

Chris White: no i'd like to stay low level if i can with um c maybe or um something like that a good stepping

Chris Gammell: stone would be arduino because that's basically c c plus plus maybe you're shaking your head at me

Alicia White: oh no i was i was i was agreeing shaking sorry an rtc real-time clock you can get a little module from from a number of places even in australia and um and a little arduino board and maybe a little led that lights up to tell you which astronomical phenomenon you're interested in whether it's sunset or sunrise you know blue or green or whatever yeah this is this is a weekend project because 75 of the code that you're going to look for you're going to type in you're going to search on google and it's going to be there for you and your goal is to be is to figure out how to put it together and

Chris Gammell: it's not going to be very hard yep excellent top down top down is very very important get that dopamine

Alicia White: hit man yeah well that's one of the things you've already got with the flow uh program you've been doing you already kind of know how to lay it out you know you you search that you you do this and then this it's people who try to get five things to go on at the same time and don't understand no you do have to do one thing at a time because the processor unless you jump to fgas right away come on

Chris Gammell: you can go right to fgas talk about a learning curve

Chris White: uh i don't i wouldn't suggest doing that by the way no i did have one other question about hardware um i want to intercept a 10 volt signal zero to 10 volt signal um reinterpret that and then output it to a zero to 10 volt signal through the micro controller i was originally thinking maybe just a voltage divider coming into the micro and then an r2r ladder leaving the micro but i don't know if is there off the shelf solutions for that or i can um yeah i use cry card at the moment to design boards so it wouldn't be a problem throwing half a dozen resistors down on the board yeah i mean uh

Chris Gammell: what kind of what kind of resolution i mean you say r2r ladder on the output and i usually just question

Chris White: the resolution you need so that would be the main thing there it's for a throttle pedal so i think um eight bits would be fine for that but i don't know what are your thoughts i don't know anything about

Chris Gammell: throttle pedals uh i mean it sounds like that's a lot of steps for a physical thing right so that's 255 steps right yes yeah so that would be that's that's that's quite a bit you know you can get a lot um especially if you're using your foot you're you might have more uh sensitivity with like tactile hand stuff but with a foot you're probably not going to feel that as much so yeah i think i think that would probably do it um but you could just as easily um you know if you had a uh a dac output on a so if you if you're using like an arduino you probably be um wouldn't be using you don't there's not a dac on a at mega 328 but you could upgrade to a mega uh arduino mega and they have dac outputs or you can just set up an rc filter and then do a pwm signal i don't know if you have ever done that kind of thing before never done that no okay i mean so basically the idea is you have a pwm signal that just kind of varies from zero to a hundred percent uh basically chopping up a standard clock signal and then you put it through an rc filter and doing that very simple thing just a resistor capacitor basically translates it back into a analog signal yep and you can get into the filtering if you want to but then you know basically uh if you want you put an op amp on board get a multiplier you know multiply by four type thing yep gain of four and uh you know so you output maybe zero to 2.5 and then scale it by four and you're ready to go uh might be harder than i'm saying there but shouldn't there be a mosfet in here um where some sort of transistor to scale it up

Alicia White: ap wm yeah just after the uh filter no yeah so like if you have just a well your processor may not have oh okay because your processor may not have it might not have enough current to drive things directly but with an op amp you should be good sorry i missed yeah there's yeah that's uh yeah

Chris Gammell: so if it's going through an rc to just an op amp input there'd be very very high input impedance so there'd be no real current flow going i mean there'd be a little bits but um yeah yeah it should be should be able to drive that no problem so excellent what about the stability of the

Chris White: voltage coming into the micro do you need to put a reference on there or can you use a zener on one of the other pins to to sort of do some maths against that um you know you can test that out i mean in

Chris Gammell: terms of how accurate you want it to actually be um yeah again i i think you'll probably be okay you know like with throttle type thing it probably as long as it's consistent over time then you can always scale it in in the firmware um yeah you know there's like yeah oh yeah but yeah the actual getting 10 volts down to a 5 volt input um i would probably take it down to 2.5 and then get you know lower scale or whatever the scale is on the the input jar do you know i don't even know

Chris White: what i was worried about was if the um if the micro is getting if one board's getting 4.5 volts into the micro and the next board's getting 5.5 that could affect the um input voltage both of the boards are both getting 5 volts on the analog pin but they're both reading different values in software

Chris Gammell: oh oh just because of accuracy of the internal reference yeah um yeah i mean i actually don't know what the spec on that thing is but i mean if again it depends on what steps you're looking for so if we take uh let me find my calculator app here um if we take 10 volts divided by 255 so that's a step of 40 millivolts which a lot of references should be fine with i mean that kind of that kind of uh it should so even even a a pretty bad reference should be okay there yeah it's almost 0.05 well it gets 5 reference tolerance yeah when i have done it in the past i've

Chris White: upped that to 0 to 50 volt range so that would become a 400 millivolt milliant yeah millivolt accuracy it's just something i was thinking about the other day thinking of pushing out of all this

Chris Gammell: hardware if you were i would be probably pretty cognizant of the the if you're doing a divider as well just make sure you have high quality resistors as well that's important um you know 0.1 are cheap 0.01 are not as cheap but you can do a lot with it so why am i using five percent

Chris White: because they're real cheap yes the application that i previously used that in was a um the lighting tower that i was talking about it's got some work lights on it and i didn't want to kill the battery bank in that so the work lights cut out if the battery bank drops below 11 volts i think

Alicia White: or 10.8 volts all right uh so any other thoughts or well i feel like chris gamel you should probably mention that you have some method to teach people software i i have this guest coming on and he says he's from constructual electronics which i hear you're associated with and so maybe you should like mention the whole i should mention that eric will be on embedded.fm next week

Chris Gammell: uh yeah no yeah we teach um we teach c yeah i mean using stm 32 parts and stuff like that using pretty standard tool chains but you know we also do arduino stuff um so yeah we we teach all those things thank you for making me be self-promotional here that's that's good i forget to do that sometimes it's not a plug if you if you demand somebody else do it right yeah yeah no that's uh but again like i i'm not sure so like someone like jonathan i would probably point them to uh like an arduino tutorial where it's just a bunch of projects to start with you know that's that's a good place to start because especially if you know especially people that don't have projects in mind where you're handed a project especially if you find a project you're interested in that's sometimes that's just just to get going you know it's all about momentum so and that's when when dan and um and michael were on the show we talked about a lot of that stuff just the momentum is really important i probably wouldn't promote myself on that for this in this case but but if you're

Alicia White: doing it for work then maybe you do want a formal thing so that boxes can be checked at work and on the side you can do more fun things true yeah yeah yeah well that's uh that's and what is

Chris White: this is for mining equipment you said right yeah yeah for for anything really um automotive i guess would be a the way to describe it okay oh cool okay yeah that's that's uh that's sounds like you

Alicia White: have a lot of really good challenges there so yeah australia is really big and they have really industrial things farming especially mining and fences fences oh the rabbit proof fence the rabbit

Chris Gammell: proof fence that steve dalton worked on we didn't really talk about it yeah right episode 63 of embedded fm well jonathan thank you thank you for calling in we that's uh that was a good set of questions thank you thanks jonathan thank you all right uh we are down to one left yeah yeah except he's not

Alicia White: australian so i don't think we should talk to him we should we should ask him to talk an australian accent what do you think i believe he's already been asked in the way in the green room that is skype he has to he has to have an australian accent okay well let's let's get it going then here we go so this is going to be christopher speck and he's actually been on our show too so this is cheating for us

Chris Gammell: how are you guys doing good good welcome back you're i mean past guest of embedded fm i mean another another one so that's right why don't you uh why didn't you tell us from memory which show you're on can he do it uh the number can he do it yeah no no looking up man no no i cannot remember the number um i couldn't even tell you the title number 78 happy cows oh i remember that one yeah that was good yeah so uh what part of australia are you from that's right yeah um the sydney that's the city you know trail right right good guess yes and it's it's currently summer here right yes that's right yes that's a good question excellent excellent did you guys did you did you excellent call call quality for getting all the way over there oh yeah so how bad was that pretty bad the delay with all them no no it's good old undersea cables man remember i talked on aussie every week so uh yeah it works you guys had some special setup as opposed to just regular old skype but uh so where are you actually from oh yeah that's a good question i'm actually from i grew up in anchorage alaska really yeah yeah i did i lived there till i went to college i ended up a few other places like reverse australia it is reverse australia some people think it's part of canada people think it's part of russia but yeah cool and your question where are you now oh i am i'm that's okay i'm in uh raleigh well near raleigh north carolina cool so what uh what kind of question you have for us um i've got i've got it up here i'm going to read it so i actually kind of get it right um so the the gist is what unmet needs exist in the embedded development world or put another way what embedded development product or service like hardware or software that you would pay money for what what product doesn't exist or product like that would you like a better or a different version of and the the thinking is that you know i i use a bunch of different products all day and some of them i wish were a lot better name names but i'm not sure if i'm not sure if i want to offend companies that i rely on but anyway some some of them it's time to renew your uh your subscription to our service and uh your price went up five thousand percent because we heard what you said exactly exactly my employer gets mad at me along the way or something but anyway i'm just wondering what um what kind of stuff would you all want that just isn't out there or that's out there but but just isn't uh isn't good enough or lacks a lot of features i was hoping for a micro that could run javascript is that is that coming and actually there are a couple i know i've made fun of him too he's just he's he's trying to make us mad yeah oh i see yeah that was just baby really i mean i just i know i've i've come on i listen to your show i get to hear it every time like i know i know what is coming too like i can hear it i'm like oh they didn't oh no they did it so i'll let you guys start i have no idea well i think you know the obvious answers chris so uh you know skipping past those is a little difficult i mean i don't think the development tools on the software side are good i mean like idees idees uh compilers debuggers i think that's all probably 10 or 15 years behind where desktop you know desktop targeting tools are and i think a lot of that is because i was gonna say sorry that's that's the that's the big tool that i use every day that i'm like why is this 15 years behind anything else partially because there's no

Alicia White: reason for it not to be yeah well the embedded community is so much smaller and so we don't have as much money to pay to get the good compilers and we have such diversity of needs with our compilers i mean sure cortex there are a lot of them now but a cortex compiler is going to be way different than on an stm versus a ti with bluetooth in it so we need all these new compilers all the time we're a small group and we really don't want to pay for anything well and all the resources

Chris Gammell: at those companies making those tools that we are have elected not to mention i think a lot of the resources go to targeting those new chips as they come out it's for time spent okay there's a new ti chip i've got to figure out uh how to optimize for that or there's a new uh st chip got optimized for that uh and so if you're a small company which most of those places are really are there's not a lot of time to go back and edit the user facing stuff sure no i totally i didn't mean to sidetrack my own question by saying why these tools are like that that they they are for very good reason but i'm just wondering what what else is out there that you all have run to that you know someone like me or someone like you know who wanted to actually like develop develop some products like this might might cook up and actually make a living at are you looking to to do this yourself is that the idea are you looking you want to build something it's always been sort of like you know hey someday or you know what kind of a side project might turn to like a little revenue on the side kind of a thing um note to anyone from my employer who's listening i'm not looking to leave the leave the job i'm good but uh but you know come on man be bold no i'm all right i've got i've got a good gig no just you know just scratch your own itch and you know make a make a little extra extra income on the side so

Alicia White: i'm just curious what uh what you all see as a need well if we look around and and not worry too much about like compilers or rtosses or even graphics libraries which i think could all be argued maybe what you're looking for as an answer instead look at what have some of the people around us done that have made a reasonable difference osh park comes to mind lane wanted boards and he wanted it not to be so hard and so he wrote some software and now suddenly we all get purple boards because it is the easiest way to get a small number of boards or you look at um bob cogshawe at at small batch assembly and he wanted well he wanted that really awesome assembly tool but he also wanted his boards assembled and so he figured out how to make a small business around the problem he had and you look at tindi and all of the things people are building there which are mostly solving their problems and then leaking that solution out to other people for a little bit of profit and i i like that method instead of like what problem can i solve everybody should have it what problem do i have and how much does it annoy me and can i solve it in a way that

Chris Gammell: like pays for itself yeah my problem is that i look at the problems that i have are things like my entire development environment think why why is this so bad to go start a huge company to fix that kind of a problem it's hard to even start a company to do that because it's not clear you'd make your money right yeah absolutely i i i do see some of that moving to open source i i think that you know compilers and things you know clang is going to be start targeting more chips and people are going to get interested in that and perhaps the problem on that side goes away on an electronics podcast can you explain what clang is uh it's a just it's a new compiler um open source compiler i think apple's had a lot to do with it um cool sort of a competitor gcc it's a more modern gcc yeah competitor oh cool uh but it it it does some interesting things um that allow it to be more flexible uh in the back end generating code but but people are getting excited about that and arm and you know i'd like to see some movement there i but i don't see i just can't see building a company around tools because there's there's so much stuff out there already and i don't know that people are really using a lot of it you know you look at the green hill stuff with um you know all the advanced trace tools for debugging or static analysis tools all that stuff is there um wow is it expensive and it's super super expensive and i'm not sure how much value that is really there i think it's a marketing i mean that's

Alicia White: a marketing problem more than a technical problem because how do you manage to sell the 15 licenses you need to employ an engineer for a year wait are you saying that at that price or a different any price well okay it's it's a it's a a market problem either you sell you have a thousand customers yeah and so maybe yeah if if they all buy at a thousand dollars well then that's actually

Chris Gammell: just fine for three engineers maybe four in shops yeah yeah yeah but more likely you're going to get

Alicia White: a hundred of them to buy for ten thousand dollars but now you have to go out and explain why ten thousand dollars is worth it and so in addition to your engineer you also have a marketing person to go out there and and do that and then because they cost ten thousand dollars as an engineer you're not just developing the thing you wanted you have to do all the drm crap which probably oh yeah for many engineers means hiring a different person because you just can't live with yourself afterwards

Chris Gammell: all right so so so those are things we can't do i'm gonna pie in the sky about things like i think i think i think you really kind of wanted more broad answers of the classes of things so things i would like to see i would like to see something i can stick on uh buses i squared c spy whatever and tell me something's wrong so something's pulled high when it shouldn't be or like an oscilloscope no i want it to tell me i don't want to have to look at it so you want like a magical debug yeah debug this bus all right i thought at first when you said bus i was like i was imagining an actual like bus i don't know why my brain went there yellow i was thinking like a big yellow cheeser yeah yeah right that too but i can tell you if sondra bulk is on this kind of a thing that that kind of a bus dude pull-ups

Alicia White: so you mean like those tools that go on your car and will and will sniff the can bus and tell you

Chris Gammell: that your engine is yeah i want something to say this has this has timing problems or you have a capacitance issue so you've got uh your rise time is screwed up or you've got excessive ringing um because are you getting lots of collisions because i want to put double e's out of business cool no i like this this is this is good this is good uh stuff for our podcast that's good okay i so in the i think i think i'm gonna go start working on this because it doesn't seem that hard it seems like a really good idea i like it i like it why do we keep these guys around yeah no i think i think but any tool where you're developing something like that it's always in you know like like i said at the beginning of the show you guys make the money in the corner cases right right you can have something that might and and there are things that are out there that do some of these type of things i mean even you know scopes with protocol analyzers that kind of thing uh however it's often knowing that you have the problem and then being able to apply it quickly and then hoping it's not a corner case yeah right those are all like and if but you're right i mean like

Alicia White: that is totally buildable for a lot of that stuff so so can you add a mode for me i wanted to connect to two things and it should um tell me if their impedances matches and if it they don't match it should give me very clear explanation on what i should do to make them match i hate it when the

Chris Gammell: impedances don't match you mean like for like a signal line or like what yeah for for signal lines okay

Alicia White: it would be it would be the same sort of yeah because i i'm thinking all of the times that i've had to go to a double e and say i give up right okay and there should be a power you should put it on power and ground and it should say how many capacitors you need and where they should be on the board where they should be on the board

Chris Gammell: just take a picture of the board and then it should generate a schematic reality near near the near the chips is the answer but but you know in all seriousness i do think but now you guys are asking you know yeah you i mean but this is embedded people wondering hoping for hardware tools and as a hardware person i'd hope for embedded tools but it's it's more of an accessibility thing right you're not yeah you're not gonna like so if i i mean something like embed would be something that probably someone like me would ask for right where it's like oh i just want the library so i can just get this thing running and then make it easy to customize if i really need to but just twiddling a couple things you know that's but it's still it's not chris's question was about embedded tools for embedded people right and it's like anytime you have a pro tool i don't think you can't i mean like without like there's a reason that these pro tools exist and that they're crappy interfaces because there's so few people working on them but yeah that's just

Alicia White: that's the market problem but there's also strangely a lot of people who are making a living on open source tools and so you don't have to invent the wheel you could just make the wheel better and if you can figure out how to get paid for that that's sort of exciting um whether it's doing support for open ocd and all of its lovely tool chain um or or working with a company who wants to use open ocd and so you you can make it better do people make money off open ocd not yet yeah

Chris White: yeah i do wonder i've used the ir tool chain and i've used all the like you know free open ocd gcc tool

Chris Gammell: chains and i look at the gcc open ocd you know you throw an st link that costs 10 bucks in the tool chain and you know you got your hardware get your software and i look at that and i think no one is making any money i mean st is making money because they're selling the chips that you used to program with the st link 10 piece of hardware but no i don't see anyone else making money in that chain so that that's part of what generates this question well it's kind of what happens on desktop right i mean nobody apple doesn't make money off their id and yet it's generally considered to be pretty good um so it's almost like you need one of the chip manufacturers to say you know what we're going to make the premier embedded id and it's only going to work with our chips and that happened with code composer for ti now look where we are oh yeah i've worked for a chip company several of them and i've i've been through a couple a couple of those a couple of those uh acquisition sort of thing

Alicia White: too those are those are tough yeah because it is a it is a never-ending problem i want to throw the

Chris Gammell: question back at you yeah i think you have stuff in your brain chris come on um it just the the protocol so here's the thing so like the the easiest thing i've thought of is you know i'm looking at urbs all day long i've got you know usually hardware urs usb to ur something it's a ur in most of the systems that i work on and so it's sort of annoying to go behind my computer plug in three or four usb cables to all these things and so i think oh i should i should make wi-fi or bly or whatever some sort of a wireless ur and i google for and there's actually i figured there should be like 10 of these out there because they shouldn't be that hard to create and shouldn't be that that expensive for the hardware but i don't see any of them and that makes me think well there's probably no money it's not worth my time

Alicia White: no wait a minute i mean i would the esp8266 sorry is your wi-fi and uh the there are a ton of ble to ur maybe i don't understand what you mean i mean

Chris Gammell: i think he means for debugging have having kind of a hub like streaming data back for debugging streaming back yeah exactly it you would use a wi-fi thing for streaming data he doesn't want to lean down behind his desk all the time right like a five dollar hub that is that is part of it i mean okay so i i work for a company that makes robots and uh my robots move a lot and so cabling them up to stuff is tough and so ah okay that's a better question yeah so yeah sorry i didn't i didn't explain that well at all chris well yeah chris uh white surprised me with the flipping the question up on me um yeah so you have all this is like and then all chris's i like it you guys really have a

Alicia White: kind of a it's true i can just say chris to anybody it could be all of you yeah yes uh so you okay i i actually have had this problem and solved it um by i had a serial port output logging system and then wanted to put that to a thumb drive so that i could have my not a robot but a toy that wandered around um and be able to log the responses to the thumb drive and then i could i could do my data analysis on the thumb drive and at the time that was a huge pain in the neck and now there are exist plenty of people already doing that as a tiny little addition to your system

Chris Gammell: but you just want to do that as a wi-fi thing this doesn't seem hard it doesn't seem hard i'm surprised maybe i just don't use google very well but uh and i don't just want to do logging i want to do by direction i mean i would worry about the actual connection for like the streaming because if you lose a packet or two you're going to be losing that but why not just do it through like a beagle bone and then have a bunch of serial outputs in the beagle bone um and then have a wi-fi chip to the beagle bone and then like elisa said like logging on that and then you could just go check it after the fact if you needed to yeah so the way that i actually have solved this is i use either raspberry pi or i've never used a beagle bone but i've used raspberry pi same kind of thing it has wi-fi you you know smash it on top of whatever device you're trying to test with you use the uarts under the hood there but those things that they're not exactly low power they change the um they change the system enough and they're not physically very small which if you're using the sorts of robots and systems that i am then they need to be like smaller is better you don't want to add to

Alicia White: the physical uh physical footprint very much you need an esp8266 it will do exactly what you want you program a little bit of code in there they're super small they're low power it will

Chris Gammell: do it over tcp and you're not going to worry about losing packets as long as you're not

Alicia White: doing something crazy and they've already got like a a ur output but he wants to yeah you know he

Chris Gammell: wants somebody else he's not going to build a company off of this though that's a service thing though right i'd be very happy to you know scratch my own itch and actually cook that up i guess when chris asked me oh what's the thing you know you've got to have something in mind that you uh you you want this is the prototypical thing that i think i want this and i'm also surprised why doesn't it exist like i realize that you know this is not technology that we don't have yet we just i haven't seen it packaged in a you know ada fruit style or or you know uh stuff that i can buy for 15

Alicia White: bucks or even even 50 bucks i just just plug it in and go and yeah exactly the bus pirate slash slash

Chris Gammell: wi-fi the wi-fi bus pirate yeah yes yes yeah that's yeah it's good point no i think there's a lot of opportunity for tools like that and and i think honestly people tend to make them when they need stuff desperately or they tend to you know do what you do and all right we'll put a raspberry pi on because make it work enough and then we don't think about it any further but i bet there's a lot of opportunity for hardware tools that's just going by the wayside because things have moved so quickly in the last couple of years that people haven't even caught up to what you can do for 10 bucks you know yeah well i think if you add a software layer on top of it too so it's even more accessible so like if you had a so like that's one thing with the the bus pirate is not just like i mean it has terminal commands but it is there's you know the special they have their own little terminal right but i mean it's accessible for people maybe outside of you know just running from a command line um if you if you can layer software on top that makes it look good on the the client side too where it's like oh okay this is the the tool right so now it's not an esp8266 it's it's built with an esp2 one of those things and it outputs serial but there's like a software interface then it becomes like a more product type thing right that's kind of the product design side of it yeah i mean when you say that it reminds me of the salia which is you know a logic analyzer right exactly i mean shipping bits across the um across usb is not exactly novel but their software looks good and it's actually

Alicia White: reasonably usable yeah yeah i love their stuff that's good so i i bet we get messages saying this exists and they tell us where to go to get the wi-fi bus pirate but i think the important thing from that is not to give up hope it's i had a good idea somebody else had one and to look at how they did it because the next time you get irritated and think if only i had this and you don't find it online that may be the time you say oh this is what i should be building and yeah and between now and then keep looking around to see what the path is i would say don't be afraid to tweet the question

Chris Gammell: too don't be like oh well if i if i tweet about it then someone else will take the idea most people no one's even people listening here like maybe one person would consider it and look it up and say oh i'd i'd like to do that idea and take that idea it's like who who cares you know like if you're gonna if you want a thing like that and you ask around on twitter or elsewhere i think that's good to put it out there yeah you're gonna find out for sure if the tool actually exists in the world and if not you might even have some collaborators that are like i'd be interested in working on that with you i'm assuming uh you all will take take some sort of a cut of the ip from this phone call here right i don't think you're gonna make much money on this cheap yeah yeah you can make a honorary donation to both embedded fm and the amp hour after the fact how about that nice yeah although we know that embedded fm does not take any uh does not take any sponsorship at this time that's true i did i did hear that if you heard any it was not put there by them and they received no revenue yeah i i heard that statement and i was like i bet their cpa wrote that or their lawyer i actually

Alicia White: wondered why they wrote why did you write yeah why what's up with that i was listening to a couple other podcasts in fact it was uh 99 invisible and he had a statement at the end like that and i liked it and then i was listening i don't know i i got onto somebody said something about stitcher and how it cut up one of our interviews to put ads in and i'm like that sucks i people should know that that is

Chris Gammell: not us i don't edit things that way yeah i would know not to use stitcher i wondered if you should all be using overkill that's what you should be using but yeah there's a lot of podcast aggregators that do weird things so it's probably a good thing to just remind people that what they're hearing is possibly not what we intended so so what you're saying is i should start my own podcast aggregator that adds ads ads and removes hosts saying there shouldn't be any ads thing is that what you're saying that's what i'm hearing i'm hearing that podcast tools are a big big market so yeah you possibly possibly bigger than the embedded tools market yeah it's true there's more people hopefully listening to podcasts than we're paying for them that's true is this your day job so you guys are going to edit my entire question this is gun south no no this is the show this is far north of where we've been yeah well that's good is that just because i'm in north america and not australia well

Alicia White: yeah yeah that was that was actually what he meant hoping you didn't notice

Chris Gammell: i'll take it let me know when you've got the uh the cool ide done yeah well it has been good to talk to you it's been great thank you very much i appreciate the call on show from uh from all three of you thank you all right we'll talk to you soon all right well

Alicia White: that seems like like it's a show i mean i don't know only an hour and 45 minutes in come on do you do you need me to find someone callers because i i'm i'm no i don't have any more sorry we're all out of australians yes there are only seven of them there and dave was on vacation

Chris Gammell: uh no i think i think we're good that uh that was a lot of great questions um really good stuff and hopefully we'll be able to do this kind of thing in the future as well dave does go on a lot of vacation so that that happens that was fun um yeah i was thinking actually maybe next time too we could pull in a mechanical person and really have a uh knock down drag out you know i don't know what

Alicia White: i'd ask a mechanical person i mean all of my mechanical questions involve can you show me how to do this

Chris Gammell: when i have a pile of parts in my hand i'm just thinking of one mechanical person and never mind that it would be not that person it would be a fantastic show but but but not that kind of show

Alicia White: well maybe somebody can suggest a mechanical person who would be a wonderful personality and can answer questions without seeing any of the parts involved so tolerances why are you so upset about that

Chris Gammell: what's with you and your micrometer what is the deal with vernier calipers are for wimps

Alicia White: why can't i put the screw hole underneath the chip it's a ball grid array it'll be fine yeah i have an answer for this

Chris Gammell: uh yeah i guess i didn't even think about that you guys don't really ever had to you never really had to interface with the mechanical people huh i have in the past a lot i have but it's usually outside my job job um i was working on i guess i can say i was working on micro micro telescopes to adjust uh beam focus of lasers and so the entire multidisciplinary thing was there was a lot of back and forth between software electronics and hardware and that kind of thing yeah i guess if you had a piece of software that was controlling yeah the physical world like through you know a motor or whatever then yeah for sure that that actually yeah that makes a lot of sense and we had you know we had markers and things and and feedback so we had to say well you know you need to put this here

Alicia White: so we can see where home is and that kind of stuff well and i worked with mechanical engineers mostly at leapfrog making toys where they would want to put together a switch that cost only two pennies and i would say that switch really sucks can we have something better they'd say well maybe if we add something it will cost 2.1 pennies but that's a lot and so yeah wow right

Chris Gammell: that could be another interesting crossover too with like uh code versus manufacturing people or i mean obviously electric electrical plus manufacturing type thing like you know large-scale manufacturing we kind of did that with jeff a couple weeks ago talking about that stuff that was the mighty ohm show right jeff geezer yeah cool what did you call keezer is that right rhymes with geezer it's kaiser

Alicia White: guy i have no i only know him as mighty ohm i'm sorry that works too sadly i'm far more into twitter

Chris Gammell: than i ever ever thought i would be yeah so which is where people can find us uh which is how we should wrap up the show what are your uh respective handles i'm at stony monster don't ask why do you want to spell that s-t-o-n-e-y m-o-n-s-t-e-r and i'm at logical elegance

Alicia White: which is also the name of you all y'all's company right that is our consulting company we do embedded

Chris Gammell: software consulting we're busy that's right so people are listening and they get if they have a challenging problem and can pay well uh call up logical elegance and we'll tell you we're busy the phones are open and probably busy yes uh i am chris underscore gamble on twitter if people don't know and the amp hours on twitter as is dave as ev blog guys thanks for being on the show it's a lot

Alicia White: of fun yeah this is great thanks for having us um and can i also say uh thank you to andre from the great white north for keeping our skype green room warm yes he's quite the character yeah it was fun

Chris Gammell: all right guys we'll talk soon thanks chris bye

Alicia White: x

Dave Jones: x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x To work out the heart. To work out the heart. To work out the heart.