#631 – A Noisy Rude Bus

Chris Gammell: This is The Amp Hour Podcast. Release May 7th, 2023. Episode 631. A noisy, rude bus.

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

Chris Gammell: And I'm Chris Gammell of Contextual Electronics.

Dave Jones: As always.

Chris Gammell: Yeah. I was watching your 10 years of... Dumpster diving. Digging through the dumpster diving. And you know what the number one thought I had was? What? Where did all that crap go? Do you actually put it back in the dumpster? Because I really hope you do most of the time.

Dave Jones: A very, very little of it goes back into the dumpster. Wow. Like the computers and the monitors and stuff, I keep, give away, use. Like I'm using dumpster computers at home, here, everywhere. So, yep.

Chris Gammell: Bitcoin mining. He's still doing it. One person. He's the last person. He's the last holdout. Just because he finds so much crap at the dumpster room.

Dave Jones: And no, I did well. A lot of them are stockpiled. I've got a lot of... Even though I keep... I gave away 15 monitors once. And then I gave more away when I moved the labs. I gave 15 computers, like small form factor computers away when I moved lab last time. You know? Psst. Hey, buddy.

Chris Gammell: You want to buy a CRT?

Dave Jones: I've still got the two CRT TVs. They're kind of handy for old school retro stuff. You know? Things like that. Sure. I'm using a network switch I found in the dumpster. I'm using a couple of network switches I found in the dumpster. You know?

Chris Gammell: Yeah. All sorts of stuff. This is what I'm going to recall next time you complain that the lab's really cluttered. Right. Yeah, yeah. Yeah. Yeah.

Dave Jones: Well, that's why I've got two storage units, you know? Yeah, exactly. Yeah. Well, the chair I'm sitting on now found in the dumpster.

Chris Gammell: You should just move the sign for the dumpster room to your storage unit. Right, yeah. Exactly. Dump it directly in there. Yeah.

Dave Jones: They have actually installed a camera in there now. After threatening, after the sign's threatening to do it for like a year and a half or something, there's finally a camera in there now.

Chris Gammell: Are they real? I always love those kits you can buy where it's just a blinking LED.

Dave Jones: I did actually trace the camera and it goes, they've got one inside, they've got one outside and it's tied into a, it turns out that it was, the reason I didn't know about it is because it was installed by the building next door, which also has access to the same garbage room and they have a different strata committee because I'm on the strata committee for this building. So there's no way they would have expensed it and installed it without me knowing. So I wondered how or why they did this. And yeah, it was, it was the building next door that paid for it. So, yep. Yeah. So they must've been having issues. They got their strata committee got fed up or something with people dumping so much crap there and, but they upgraded the whole security system for the building. So they've got cameras all over the place now for the building next door. Whereas.

Chris Gammell: It is really easy to upgrade security systems now with like, you know, wifi sensors and all like the little thread and all the different sensors that are out there. Yeah.

Dave Jones: But these ones are all wired. So yeah, I, I actually. I followed the conduit. I followed the conduit a lot, you know, they screwed it into the concrete roof, right? That is the right way to do it. Car park.

Chris Gammell: And you know, if you really want to do it right, that is definitely the right way to do it.

Dave Jones: But yeah.

Chris Gammell: But yeah, like, you know, like a lot of, there's a lot of.

Dave Jones: Well, you need power. Right. So it's like, you know, you're going to have to install it where there's power. If you're going to get the wifi thing, you can't just magically install the wifi thing.

Chris Gammell: Right. Well, no, a lot of the, a lot of the, like, so like simply safe and other ones that are like kind of pop it up, at least in the U S a lot of them are battery based sensors. And then there'll be like Zigbee or like.

Dave Jones: Yeah, but you can't have a camera. You can't have a camera with a battery.

Chris Gammell: Although, no, actually they are doing, that is actually another thing too. I've been surprised about that because I looked at this for my house actually. And there was like, there's a nest camera that's like wireless. And I was like, how the hell, you know, like there's no installation. Okay, fine. Whatever. What you do is you basically, it's like, it's like this magnetic mount. Yeah. And then like once every, I guess during daylight or, you know, whatever, you pull these things off their mounts and you go plug them in and then you charge them up and they come back out. And it's like, oh, that's.

Dave Jones: But how long is a camera going to last on a charge? I don't know. I mean, that's just BS, right?

Chris Gammell: Yeah. I didn't buy it for a reason.

Dave Jones: No, I just can't picture that working. That's just dumb. And if you want to install something like that, you don't want to have to maintain it. You just want it to.

Chris Gammell: Yeah, I think that's right. You want it to be like an appliance, right? Set and forget. Yeah, totally. Totally. Yeah. Yeah. Agree. Agree.

Dave Jones: Yep. Yeah. You can install, you know, per motion sensors, you know, you install like a CR123 battery in there and it lasts five years or something, you know, it's fine.

Chris Gammell: Yeah, exactly.

Dave Jones: But no, a bloody camera. No. Yeah.

Chris Gammell: Like read relays and motion sensors and stuff like that, where it actually activates and like cuts off the switch otherwise. Yeah. Yep. Yep.

Dave Jones: Yeah, nah. All right. What have you been up to?

Chris Gammell: I have been, oh, I've been, wait, oh, let's see if I can, I can turn it on. I've been writing music, Dave, for a demo I made. Here we go. Ready?

Dave Jones: Yep.

Speaker ?: What?

Dave Jones: What? Is that it?

Chris Gammell: That's it. Yeah. So I've been working on Thingy 91, which is like a little platform for Nordic for their cell thing.

Dave Jones: Right.

Chris Gammell: And they have a buzzer on there. You know, it's just a little piece of buzzer. Yeah. But like there are no examples on the entire internet. And I was like, I'm going to go make some examples for this. And like, I was, I was making demos for work.

Dave Jones: Yeah.

Chris Gammell: And so, oh, I guess I could probably turn, I should probably turn it on and get the other ones playing.

Dave Jones: So it's an actual buzzer at a fixed frequency. It's not a piezo transducer, which you can operate at any frequency. Which one is it?

Chris Gammell: It's the latter. Yeah. So it's like a NPN driving a piezo that you can drive. Oh, right. Okay.

Dave Jones: So in theory, you can drive it at any frequency. Okay. Right. So it's not a buzzer. A buzzer.

Chris Gammell: Oh, interesting.

Dave Jones: Oh, I didn't know this. A buzzer is defined. Yeah. A buzzer is defined as something that has its own internal oscillator and it oscillates at one frequency. It just buzzes at one frequency. A piezo? I would call it a piezo transducer.

Chris Gammell: Transducer.

Dave Jones: Okay. Yeah.

Chris Gammell: Yeah. That's great. Okay. So this is a piezo transducer. Right. And yeah, I've been using this thing to just make some silliness and- Sweet. It's been pretty good fun. Triggering stuff from the internet, of course. It's basically the most trivial internet machine I could make.

Dave Jones: Right. Yeah.

Chris Gammell: Among many other things that it does, but that's the most trivial part of it for sure.

Dave Jones: Right. Right. So is that the extent of your musical skills at the moment? I thought you were an actual muso, dude.

Chris Gammell: Well, I had to actually, because it was just like, I had to set up kind of like a framework for it because it's literally, when you do a piezo, it's like, or sorry, when you do a PWM, it's like, well, okay. You give it a period and you give it a, how much duty cycle it is. And that doesn't really translate to like notes. Yeah.

Dave Jones: Right.

Chris Gammell: So then I had to go and like, right, this note is this. And so I put in all the notes. Oh, okay.

Dave Jones: So you've got a note library, a note table or something. Okay. So you can choose your notes. Okay. Right. Yeah, exactly. That makes sense. Rather than, you know, write a whole lot, you know, you could have a library that takes in a waveform or something and then, you know.

Chris Gammell: Yeah.

Dave Jones: So in theory, you could actually make that talk.

Chris Gammell: So yeah, exactly. If I like learn phonemes and like did all that sort of stuff. Yeah. Like I could, I could go pretty insane with this sort of thing, but I, I probably will start stop at a certain point. All right. Let's see if, let's see if the internet part works. Yeah.

Dave Jones: All right. It's very old school Merlin. I've, I've done a video on that old school Merlin, you know, game from the 1970s, which used the world's first microcontroller, you know, and. Oh yeah. Okay. It's yeah. No, I'll try and get that video.

Chris Gammell: Okay. I mean, you know, that was the Mario theme, right? That it wasn't very well done. Oh, I, I, I, that was the, sorry. That was the Mario brothers theme. Okay. Oh, okay. And then I also did this one. Got it. Unfortunately, like part of my lab does not. Oh, well, that didn't quite catch it, but that was funky town.

Dave Jones: All right. As you do, as you do.

Chris Gammell: Got it.

Dave Jones: Anyway, I sent you the link to the Merlin. Yeah.

Chris Gammell: And then I pretty much got bored with like putting in new songs. So, but I have the framework now, so I could put in other stuff, but yeah, monophonic and, you know, I wrote a little song and, you know, I put in two covers and, and then that's about it, you know, and then I moved on to the other stuff I had to do with it, but they're triggerable over cellular from the internet. Right. So that part, that part's actually pretty cool.

Speaker ?: Right.

Dave Jones: Is there like an online tool that you can put, like, you can just give it a song and then it gives you the notes out? Is that like a thing?

Chris Gammell: Hmm.

Dave Jones: Surely there is.

Chris Gammell: Well, I tried that with ChachiBT and it didn't quite get it right.

Dave Jones: Okay.

Chris Gammell: You know, the easy way to do that, I think is like, just go and look up sheet music, but of course you need to like read sheet music, which I do, but other people might, if you don't have that experience, you might struggle with that.

Dave Jones: That's why I'm thinking an online tool that can just spit out notes. Oh, it's a B, G, A or whatever. Yeah.

Chris Gammell: I bet, I bet you could find something that would do that. I was actually looking for something. I was looking to see if there was other AI services. Of course, we already got into AI. I was looking for other AI services to see if it would like write me a song, you know, like I tried, I tried like, Oh, like make it sound like a John William-esque theme, you know, like basically like a mid journey for music. And apparently that is coming. Google has some demo. There's like a bunch of demo stuff out there, but none that are like services where I can just type stuff in and have it spit something out, which is fine.

Dave Jones: Interesting.

Chris Gammell: But it is on the way. I forget what it's called for Google. Right. You know, there's so many repeating elements in music and there's actually like a court case that just concluded where Ed Sheeran was being, Ed Sheeran's like a pop singer. Yeah.

Dave Jones: He was being sued by the estate of Marvin Gaye. Copying something or what?

Chris Gammell: Yeah. They said that one of Ed Sheeran's songs sounded like, let's get it on. So.

Dave Jones: I don't know it, but okay. Right. Right. So they're claiming it was, yeah.

Chris Gammell: Yeah. They're claiming that. And like, and it's, it's interesting. There's a, there's a couple of Adam Neely videos about it. Who's like a music YouTuber. Pretty good.

Dave Jones: Yep.

Chris Gammell: But it's basically like, he makes a really good point that like, even though this is about that specific song, it's actually probably much more pertinent in the, the oncoming age of AI, because like, can you copyright kind of like a feeling of a song? Right. Right. Because it's not like, not the same words. It's not the same, but it's like definitely in the same style. But like that song, like the Marvin Gaye song is like basically defining the style. And so like, that does have some implications, I think going forwards.

Dave Jones: Yeah.

Chris Gammell: It seems for now, AI is going to win the day.

Dave Jones: Yeah. Well, that, I, I coined a new term the other day on Twitter. I just sent you a link. Okay. Anxiety or anxiety, AI, anxiety, you know, it's, it's, it is the feeling that you have when you think you may not have chosen the best, most optimum AI tool for the job. Because there's so many AI tools, like there's a dozen new AI tools coming out every day. Now it is insane. And it's like, which one do you pick? Oh, I've got this task, but, or have I picked the right AI tool? Is this one going to, you know?

Chris Gammell: The only redeeming factor is that there's not really a whole lot of like skill building that goes into each new one. So that part's good.

Dave Jones: It's just, oh, but there's just, there's thousands of them, literally thousands of AI tools. It's just, and it's only going to get worse. It's just nuts. Yeah. So how, which, which one do you choose? Which one's the best for your tasks that you need doing? You know, there's a hundred choices now.

Chris Gammell: I mean, it's got, it's coming for us, Dave, over time. I think it's all about like existing in the, in the, in the corners for now. Right. I mean, so that's, that is honestly part of the reason I, you know, it's not like a true concern, but I, I am interested more in like getting, continuing to get better at firmware and stuff like that too, because it does feel like the board layout piece, it's much more like ubiquitous, which is actually great. I think from an access perspective, but also like if I was going to hire, you'd be hired for my services in the future. I'm not sure it's as hireable. I guess you always have like people that, you know, you always have your reputation and like being able to solve problems that, that part's not going away. I don't think, but I do wonder about the hardware versus the firmware piece. It does seem like the, the firmware piece is harder to solve in a lot of projects or there's more, there's more stuff there. That's my, that's my gut feel.

Dave Jones: Yeah, but you still need a program. You still need a good programmer to make sure the out, to combine it all together and make sure the output from the AI is good, right? So you still need those. Yeah. Yeah. Yeah.

Chris Gammell: Some kind of like judgment engine, right? Whatever it is, it's like, yeah, I think AIs will have fever dreams and I think they've achieved the task, but they're, you know, being, being the third party in that case is still valuable.

Dave Jones: You need to, yes. There needs to be an adult in the room. So there you go. You need to be the AI adult. And here is my tip for every, every youngster out there, right? There is a new technology that comes along once in a lifetime, once or maybe twice in a lifetime. AI is it, right? So if you are not getting on this now, you should become the expert at whatever company you're working at. You should become the expert. Spend every waking hour. I'm, I'm, I'm not kidding. Spend every waking hour learning all the different things that AI can do, become the expert. So when your company goes, oh shit, we have to do something with this AI or, or, you know, how can we optimize things? You can go, Hey, I've, I've been doing all this work on the side. I'm an AI expert. I can, I can show you how it does this, this, and this, and this, and boom, you'll be the most valuable employee in the company.

Chris Gammell: So that's interesting. It's not a bad, I don't disagree with it, but like when you say an expert.

Dave Jones: You can't lose, you can't lose by spending your time learning all the different stuff that AI can do. Yeah. But like. You don't necessarily have to be an expert. You, you just need to be smarter than, you need to know more about AI than everyone else in the company. That's what I'm saying.

Chris Gammell: Yeah. You just got to be faster than, faster than your slowest friend, right?

Dave Jones: Yes, exactly. So no, no, no, you don't have to be some absolute guru, but you will appear as a guru, right?

Chris Gammell: No, I know. I know. But what I'm getting at here is, do you think it's learning how to program and build your own, or is it just as a user?

Dave Jones: Doesn't necessarily matter. You just have to, even if you just know which tools are applicable to which application and, and how to utilize them, that is more than enough skill. That's more than 99.9% of people who work at your company will know, right? Is, is how to use a particular AI skill for a particular AI tool for a particular purpose and then how to drive it. And if you know that.

Chris Gammell: Okay.

Dave Jones: Yeah. Yeah. You are invaluable.

Chris Gammell: I guarantee it. I don't know. I guess I'm, I'm discounting the, like the general ubiquity of all these things too, that like to truly stand out in that case, it doesn't really just knowing the tools doesn't feel like you would stand out in like a large organization that has potentially other young people there. That's, that's the main thing that I, I'm resisting it.

Dave Jones: But what you're not counting on is that most people are lazy.

Chris Gammell: Most people are lazy. Yeah. Most people are lazy.

Dave Jones: You can be the, you be the person who's put in the hard yards at night on weekends, learning.

Chris Gammell: Yeah. Doing projects for yourself. Doing projects for yourself. Showing.

Dave Jones: Showing stuff. Have stuff that you can immediately show them. That's a good point. That you know what the hell you're talking about and you can actually apply this school immediately. Yeah. That's not bad.

Chris Gammell: Okay.

Dave Jones: Right. And trust me, you'll be the most valuable person in the company because every company is going to have to get on the bandwagon. They're going to, oh, how do we, but, but they have no idea how to use it.

Chris Gammell: Right. You'll be the Ryan Howard of your company. You'll go from being the temp. Who's that?

Dave Jones: What's Ryan Howard?

Chris Gammell: That's a U S office, the office from the U S. Oh, right. Okay. It's a reference to him. Right. Sorry. Yep. The people who knew it though, they were like, yeah, I know that reference.

Dave Jones: But the thing I'm saying is that this is like a once in a lifetime thing. Very rare. Like the internet was one of them, right? If you got on the early internet, stuff like that. Right. Right. Name another thing. Right. Well, it, it'd be a personal computers in the late seventies, early eighties. The printing press. Yeah. Personal computers, desktop publishing, the internet. Like I'm not thinking of much else. Right.

Chris Gammell: Right. So I guess this is another piece of my resistance here is that like, okay, so if you, so like, let's talk about the internet, right? Yep. Some people said, Hey, I'm on Usenet forums. I know how to use this stuff. And they stayed in their company and they said, yeah, I'm the, I'm the guy. And then they became the IT guy. Right. Other people did that. And then they went and started a business and sold to those IT guys and they're billionaires. Oh, sure.

Dave Jones: Yeah.

Chris Gammell: So those kind of like, that's, I guess that's kind of just the stuff that's sitting through my head. I'm not saying that people should like put their jobs and go and become AI.

Dave Jones: But, but you need to get on this now so that either you can branch out on your own, right? And form your own company or form your own products or whatever it is, or you can be the most valuable person in your company. Right. Either way is fine. Right. It's yeah. I'm not talking about how to become super rich. I'm just, you know, how to secure your future. Yeah. Right. And this is a once in a lifetime opportunity.

Chris Gammell: Do not pass it up. If we are molding young minds, Dave, maybe we should, we should tell them to shoot high and then, you know, worst case scenario, you become the IT guy.

Dave Jones: Oh, maybe. Okay. Yeah. Well, that's why work on stuff on your own time. And then that either, this is maybe that lands midnight engineering stuff, right? This is classic midnight engineering we've talked about forever. Right. But do it for AI because AI is the new game changer. Right.

Chris Gammell: Yeah.

Dave Jones: So yeah. Be the most valuable person in the room when it comes to AI.

Chris Gammell: Yeah.

Speaker ?: Yeah.

Dave Jones: You, it will not be a waste of time spending your time on that. Right. Don't go out and party. Kitties. I'm telling you. Right. Spend your waking hours learning AI.

Chris Gammell: I think what we might be deluding ourselves is that anyone under 18 or even under 25 is listening to the show right now. But if they are, first off, you're awesome. And also you older folks, you're awesome too.

Dave Jones: No. Yeah. But how many college students would we have? Right. Tons. I'm sure. I don't know. Right. Heaps.

Chris Gammell: I don't know, Dave.

Dave Jones: Right. Stop partying. Learn bloody AI. Right. Get it. Don't be a lazy bastard. Right. Yeah.

Chris Gammell: Yeah. Yeah. That's true. Go hard. Yeah. I agree with that base. Yeah. I don't know if I would have listened to it in my younger days.

Dave Jones: Yeah. Like I totally wish I had that advice back in the day. Yeah. You know? Yeah. And just learn it. Go hard. Yeah. I wasted so much time just bumming around, being lazy. You know? Yep. Yep. Totally regret that. Waste of time.

Chris Gammell: Yep. Dave, you can't regret it. It brought you here to this very moment.

Dave Jones: I know, but yeah, every step I did.

Chris Gammell: Talking to the awesomest co-host you've ever had.

Speaker ?: Right.

Chris Gammell: I am not worthy. I am not worthy. So, yeah. Yeah. I was going to say. I was going to say. Love it. Anyway, that's my tip for this week. Well, another thing I'm interested in, in terms of like future trends, I was just watching Wendover today, had a really interesting video about manufacturing in Mexico.

Dave Jones: Oh, I have not seen that one yet.

Chris Gammell: It peaked. Yeah. It just came out today. Yeah. And it peaked my interest because it was basically, I think the title was like, Mexico is the new China kind of thing. Oh, right. That's like a buzzword-y kind of thing that's been written over and over again about Vietnam and India and Brazil and all of, you know, like lots and lots of those. But I think, you know, he lays out a case for the logistical aspects of it in relation to the proximity to a very, very large market.

Dave Jones: Oh, just simply being close is. Yep.

Chris Gammell: Totally.

Dave Jones: That is. Yeah. Yeah. Like the border, right? It's right there.

Chris Gammell: Yeah.

Dave Jones: Sometimes it's miles away, right? If you're talking like San Diego, it's like, you can just walk over.

Chris Gammell: You literally walk over the bridge. Yep. Yeah. Yeah.

Chris Gammell: It's like, you know. Yeah. Yeah.

Chris Gammell: And what I was wondering though, is like, I would love to hear from listeners is if you are doing manufacturing in Mexico and you're allowed to talk about it, of course, just like, what's it like? I mean, like, first off, I felt like a little validation. I'm like, oh yeah, I'm learning Spanish and not Mandarin right now. Yeah. Yeah. Right. But that's actually because I don't feel like I could learn Mandarin and I don't really, you know, like, it's that easier.

Dave Jones: I would have that advice. I wouldn't be learning Mandarin. I wouldn't bother. Yeah.

Chris Gammell: You know, I, did you, did I ever mention on the show? I think I probably mentioned on the show that I took Japanese in college.

Dave Jones: Oh, maybe. Yeah.

Chris Gammell: Yeah. It was like this, this trailing edge advice of like, you know, growing up in the 90s, people were like, oh yeah, you got to learn Japanese.

Dave Jones: When you're in college, the Japan thing was. Yeah. Yeah. Like, like, like Japan was huge in the eighties, right? Japan. Well, like. Basically.

Chris Gammell: And through the nineties, I think too, right? I mean like.

Dave Jones: Yeah, maybe. Yeah. Early, early nineties. Yeah. Yeah. Yeah. I mean like until. Yeah, for sure.

Chris Gammell: Until a lot of the, the stuff shook out with like the memory wars and stuff like that. I feel like. Yeah. Yeah. Yeah. Yeah. That was. I also like Japanese culture, of course, but like, but it definitely was on the back of like, oh, well I'm doing engineering. This would be a good idea for like my career too. Sort of thing. Yeah.

Dave Jones: Well, maybe that's something I can ask Steve Senge next week.

Chris Gammell: Oh, a little teaser there. Yeah. A little teaser.

Dave Jones: We, we have him booked. We have the CEO of iCrochip. Steve Senge will be on next week. So. Yep. I'm going to record that on Tuesday. Yeah. Assuming nothing goes wrong.

Speaker ?: Yeah.

Dave Jones: Yeah. So. Yes. Send questions. I do have a Twitter thing down below.

Chris Gammell: Oh, great. Great.

Dave Jones: You know. And I, of course I asked that question. Everyone bitches me. You know, everyone bitches. When can I get my parts? When can I get my parts? Why does your software suck? You know, I can't ask him, why does your software suck? You know, I mean.

Chris Gammell: No, it'd be kind of rude.

Dave Jones: Yeah.

Chris Gammell: You could ask that in a nicer way of like, what are some resources for engineers looking to, to learn your unique style of software?

Dave Jones: I can ask chat GPT for a nicer way to say it.

Chris Gammell: Nice way to do it. Yeah.

Dave Jones: Yeah. Right.

Chris Gammell: Yeah. I think even if you ask really rough questions, like, you know, he's the CEO, he like, he knows how to like, you know, they've all had media training too. They know how to like get around questions.

Dave Jones: They know how to get around questions. Yeah. They know how to work.

Chris Gammell: No hard hitting answers for, for hard hitting questions on the Amp Hour. We're all about softballs around here, folks. But yeah, what would be, what's the thing in relation to like Japan and China? Yeah.

Dave Jones: I don't even know where they're making. I assume they're still making some chips in Arizona, right?

Chris Gammell: Yes. Yeah. Are they?

Dave Jones: Yeah.

Chris Gammell: They are purported to be growing their fabs, I believe. Oh, okay. I think they were a part of the Chips Act. So that I think would be a good question. Yep. Chips Act and just like what are their plans there? Yes, yes, yes.

Dave Jones: The Chips Act and everything. Yeah. Yeah, definitely. Yep.

Chris Gammell: I'd also be interested about like, you know, just like, how's it going with eating Atmel? Yeah, no, totally. Are they digested yet?

Dave Jones: Are they fully digested? Yeah.

Chris Gammell: Yeah. I mean, you know, and like, I think that's got to be tough from a, you know, like you and I still talk about Burr Brown parts, right?

Dave Jones: Oh, yeah. No, that's hard.

Chris Gammell: I ate them, what, 35, 40 years ago. And like, there's just, there's different operating groups. There's geographical differences. It's basically like, like there's the European Union, right? Which is this collection of many, many countries in the continental Europe. But like, we wouldn't refer to like an Italian person the same way as you would a French person unless you want to get punched twice. Yeah. Right? I mean, like, you know, like there's just like these cultural differences that like persist and people take pride in. And it's like, you just, it wouldn't be smart to do that culturally, I think, within a company either. No, totally not. I'd be interested in that dynamic.

Dave Jones: There's so much legacy in these, you know, because they all get gobbled up. We've covered it here on The Amp Hour. Like, it's just insane the amount of companies, semiconductor companies that have been gobbled up. But yeah.

Chris Gammell: Yeah. So that would definitely be interesting. What else would I ask Steve?

Dave Jones: Well, I can ask him if, if anyone's attempting to buy microchip.

Chris Gammell: Yeah. I'm sure he'll, he'll be right out with that. Yeah. Right.

Dave Jones: He'll be right on that, you know? Yeah.

Chris Gammell: Yeah. Yeah. Oh boy. Yeah. Maybe the, I would be interested actually the, what it's like with a public company, you know, like for, you know, they make very complex things and like, and then you have people that are on wall street, you know, it's a public company. And just like, what is it like interacting with investors and do they have any semblance of the actual difficulty? You know, is it just like, Hey, money machine crank out more money? Or is it like, actually there's actual interest there and like, does that drive business and growth and that sort of thing? You know?

Dave Jones: One, one, one of my big questions is going to be supply chain, right? Cause people think, Oh, I, you know, us, uh, engineers, we need the chips, right? We think the chip is the bottom. Like if we can't get our chip, right? That to us, we don't care about the rest of it. Can we get the chips or not? Right. But then the companies that manufacture the chips, they're, they're going, Oh, well, we need, we, we need our silicon. We need our, you know, our wafers. We need all the chemicals that go into making them, you know, like that's a supply chain in itself. So everything ripples, everything ripples right from the raw material upwards. So they've, if they've got supply chain problems, like, like we think they're the bottom of the supply chain, it's turtles all the way down. Right.

Chris Gammell: That's right. Exactly. Exactly. Yeah. Right.

Dave Jones: Oh boy.

Chris Gammell: We can't process the wafers until we have more machines and the machines need wafers that are inside of chips. And Oh my God.

Dave Jones: I think I don't actually know that. Like, why is there a chip shortage? What is stopping you making the chips? There must be something further down the supply chain that's stopping them making it. Yeah. I know it was the whole COVID thing and they did predictions wrong and shut down. Then there was sudden demand and they couldn't get them and the big companies were buying them up. But is there something beyond that? I'm just curious.

Chris Gammell: I bet that's pretty big. Yeah. Yeah. That's, it's interesting. I mean, like, and also like, you know, Steve's insight into the whole supply chain too. Right. I mean, like the CEO needs to know a lot, but I'm guessing that it would be a big waste of his very expensive time that, you know, if like an operations manager in the silicon sourcing was like, you know, like he's giving him like one sentence for the investor meetings versus like, you know, like Steve, can you help me solve this problem, please? The fact that he's talking to you is actually like, that's, that's, we should ask how much it would cost. Like, what is he really giving up? What is the opportunity cost? Yes. What is it?

Dave Jones: Well, I was struggling to get the full hour. It was like, oh, we might be able to get you half an hour. You know, it was like, but apparently. We'll take whatever we can. Yeah. Yeah. Yeah. We'll take whatever.

Chris Gammell: Whatever he doesn't fill into the full hour, then it's going to be me and you filling in with like, uh, like an after, like an after game show, like a sporting event, you know, and we're just going to fill in the details. So the more time he gives, the better it's going to be for the, you know, how they look.

Dave Jones: Got it. Ah, boy. Anyway.

Chris Gammell: We'll take whatever we get though. Yeah. Yeah.

Dave Jones: We totally will. Yep.

Chris Gammell: Yeah. That's cool. No, I'm excited. That's really great. Yeah. It is. I mean, I don't know. It's, I don't know if you've been seeing stuff. We talked about, oh, we've been talking about a little bit, I guess when other day was here in Durham and we recorded, he was talking about it. He's a consultant, you know, just kind of like general loosening up of supply chain. And I forget who was on before Dave. We've been talking about supply chain a little bit though. And it seems like it is loosening up. Right. That's, that's my general feel. And also like my digi keying.

Dave Jones: Depends on your luck. Like, you know, I hear people still know I, my business is totally shut down because I still can't get parts. You know, it, it depends on what, what part you've actually got.

Chris Gammell: And yeah. Yeah. I dropped in on the, uh, the Adafruit live show yesterday too, is Phil and Lamor were talking about like, like, oh, we got a reel of, we got, we managed to get this one reel and we're able to make this product now. And it felt like, you know, it's kind of like that stop, start nature of it too. That felt very realistic too. Of just like, yes, you just do what you can as you can. And then just hope you have the other parts that are necessary in the moment. Exactly. But I think that's how like the train kind of starts back up, you know, like it has to happen like that. And I think the thing that's really helping a lot of the, the things shake loose is people not scrambling and they can go back to building like they had before. And like, then they can start to build, you know, so now I'm not building with the parts that you needed and you're not then scrambling and building with the parts that, you know, someone else needed and then on and on and on down the chain, like how it ripples like that too.

Dave Jones: What's all this IQbed C rubbish?

Chris Gammell: IQbed C. Oh, that is a new standard that is on our list.

Dave Jones: Because there aren't, because there aren't enough standards. So we'll make a new standard and that'll be the standard, the XKCD thing.

Chris Gammell: Well, this is just the, it's the next one, Dave. It's the IIC.

Dave Jones: So this is brand new. Is this analog device, what's this? It's on the plan analog device. It's the I3C, which I'll call the IQbed C. Right. Because this is an update to I squared C, right? This is update to the Philips I squared C standard.

Chris Gammell: That's right. Yeah. So Philips was the original and it was licensed at some point, but you know, it's obviously become genericized.

Dave Jones: It's a trademark thing. If you want to use the I squared C name, then you have to pay, well, I don't know if it's still the case, but it used to be the case. You had to pay them a trademark, you know, royalty fee and you had to license. And you know, back, yeah, as you said, like they would give you a license to dress that you could use and all sorts of things, you know?

Chris Gammell: Mm-hmm.

Dave Jones: But yeah. So a lot of companies will just put it, it'll just be like a generic, they'll just call it a two wire. Two wire interface. TWI. Yeah. Yeah. Yeah, exactly. But it's actually I squared C, right? That's right. They just don't want to use the trademark. Yeah. Yep. So this is I squared C. So who's, who's come up with this?

Chris Gammell: This is I3C. This paper is from TI, but I'm guessing, I think the MIPI Alliance is actually, actually doing it. So MIPI.

Dave Jones: Oh, it's a MIPI, right. MIPI is the thing that a lot of cameras use and stuff, like a serial interface for cameras.

Chris Gammell: Yeah. And so I think they're probably, I think that's the other thing that's kind of interesting too, is that like back when Philips did the I squared C, it's like Philips was just so big and monopolistic and they, you know, companies didn't really do this, this whole standard thing anyways. And now it's like, this is a little bit better of a art, you know, of working in alliances and that sort of thing. It's not easy, but.

Dave Jones: No. For those who want a history lesson of how I squared C started, it was like a Philips, of course, this is back in the days of television, none of this digital rubbish, right? Analog television. Philips made a chip for every aspect of a TV. So you'd have the tuner chip, you'd have the filter chip, you know, you'd have like, you'd have the CRT driver chip, you'd have the onscreen fancy, fancy ones in the later in the day, you'd have like an onscreen display chip and you'd have all these chips and, you know, a, and a TV would have dozens and dozens of these chips. And Philips needed a way. Philips came up with the I squared C bus so that they could join all these chips, standard protocol to join all their own chips together. That's how it came about. It was the standard for TVs.

Chris Gammell: Right.

Dave Jones: Every TV used the I squared C bus.

Speaker ?: Yeah.

Chris Gammell: Whatever you do is a standard then because you just keep using it internally. And then other people are like, Ooh, I want to play.

Dave Jones: Yeah. You just keep using it internally. And then all the other competitors, all the other manufacturers went, Oh, well, I would like to talk to the I squared. I would like to make a chip that talks to the I squared C bus, please. I want part of that action. And they went, okay, we'll license it to you. And, you know, and then just, yeah.

Chris Gammell: And I think, you know, luckily we are getting back to the age of chip companies buying each other and being massive enough. So they just get to make their own standards again. So why not? Yeah.

Dave Jones: We're back there. So I don't see any technical detail in here. Is this still like a pull-up base? Cause I can't, it's, it's 33 in like in ternary mode. It's 33 meg bits per second. I don't, I have a hard time believing you can do, you can't, you could do that without like an active pull-up or something. Does it use an active pull-up? Because basically the speed limit of an I squared C bus, you should know this, right? This is bread and butter. Is that it uses a, usually a 2.2 K pull-up resistor, right? But then the capacitance of your line limits the, you know, you've got, then you've got an RC charge constant, time constant, and that limits the slew rate and that limits the speed of your bus. So is this like an active pull-up? I assume it's like a, still a two wire thing. I, I don't see any details.

Chris Gammell: Yeah. I just, I just linked you a wiki page about it as well. I think it is still pretty, let's see, 2016. There was some stuff, some early stuff. Oh wow. It's been around for a while actually.

Dave Jones: Oh, there's a Venn diagram. It's, it's half Iceberg C and half SPI.

Chris Gammell: It seems like this is still, yeah. So it seems like this has been a long time coming, but it definitely is like a superset type of thing. Yeah. Like Dave mentioned the superset or the Venn diagram shows it like crossing over with SPI as well, which is kind of nuts, but.

Dave Jones: Oh, it's part of the new DDR5 standard. That's interesting.

Chris Gammell: Yeah.

Dave Jones: That'd be for, that'd be for memory chip configuration or something. Maybe.

Chris Gammell: Yeah. I, the thing that I don't understand about all this stuff though, is it does seem like they're just trying to force more and more stuff on, they're trying to make like a mid range bus, which is still like. Yeah. It's like a jack of all trades. Yeah. It's trying to be. Yeah, exactly. And what, and so like, I don't know. I just think I squared C is like, it's going to be slow and it's going to be, you know, it's not super noise resistant.

Dave Jones: But that's all you need. But sometimes a good lot of time that that's all you need. You just need to. Exactly. I just need to update a few bytes in a real time clock chip or something. I don't know. Exactly.

Chris Gammell: I need to take a temperature on some other part of my board. I need to get that across the board and like, I can just, you know, put pull ups at either end and I'm good to go. I can slow down the clock if it doesn't look good. Right. So like that, but like, but it does get, I mean, by my, by yours truly included, like it does get bastardized quite often and like loaded down with lots and lots of sensors. And, you know, there are limitations on addressing as well. You know, so if you have the same sensor, you don't have address switching, you need to insert like a I squared C controller or switching chip that handles that sort of stuff. So it does get overused a little bit sometimes, but I'm not sure how then this plays into it as well.

Dave Jones: Right. I still can't find any info on the hardware layer. I'm not. Yeah. Yeah. It's there somewhere. Yeah.

Chris Gammell: Yeah.

Dave Jones: Whatever. Yeah. It's, it's, it's trying to bridge the gap between spy bus and I squared C and it's like, do we need to?

Chris Gammell: I don't. Well, and I think this kind of comes back to like the, uh, the, the Daveism that's like, uh, it doesn't exist until it's on digi key. Right. And it's like, right. And I think that that's a pretty good rule, especially in this case, like, well, we'll have to worry about it when we see it on microcontrollers and parts that we actually care about. Yes. I'm actually much more concerned about, what was it? Oh, on the, the new Nordic chip we talked about a couple of weeks ago, the 54 H20.

Dave Jones: Right.

Chris Gammell: So they don't have 3.3 volts. They only have one bank out of all of their IO that has 3.3 volts.

Dave Jones: Oh, one bank. Is that it?

Chris Gammell: One bank. Yeah.

Dave Jones: Usually it's chip wide.

Chris Gammell: I think it's because they're moving down on the, the size of the, the, like they're on like 22 nanometer now.

Dave Jones: Right. Yeah. Yeah.

Chris Gammell: Like just starts to not be realistic. So you, then you have to go external with like a power controller. And that worries me a little bit more than this sort of stuff, you know, like, right. Just because it's like sensors are also going that way. It's not like there's not sensors available, but just like the multiple power domains.

Dave Jones: Like I'm just finally getting my brain. It really got annoying when, you know, everything moved from five volt to 3.3. It's like, Oh, you don't have five volt compatible IO anymore. That's annoying.

Chris Gammell: Right. And if we have anyone out there listening, who's like, Oh, I designed cell phones or like I'm on like leading edge stuff. They're like shouting right now. Tell somebody cares, you know?

Dave Jones: Yeah, exactly. It's like, you know, like we are just working on our industrial widgets.

Chris Gammell: Yeah, exactly. Like slow and slow, baby. Slow and slow.

Dave Jones: We, we've just got, it feels like we've just gotten used to the fact that, Oh, okay. It does it. You just assume everything doesn't have five volt IO anymore. And that's fine. Right. I seem to have like just accepted that now, you know?

Chris Gammell: Yeah. And there are still some specialized ones like the new Arduino thing that's using a Renaissance chip. That's five volt tolerant. But I think it's also like they chose it because they know people are going to be trying to shove 12 volts. There's all backward compatible. Yeah.

Dave Jones: And there's all backward compatible boards and whatnot, hats or whatever they are. Yeah. Yeah. Shields. Yeah. Shields. Things. Yeah. I don't know. What's the world coming to?

Chris Gammell: Well, you know, just more and more standards. I actually, I've been playing around a little bit with CAN, which is an old standard as well. Oh, okay.

Dave Jones: You're in the automotive. That's, to me, CAN is like, it's just in cars and that's it. It's like.

Chris Gammell: That's a lot of it. Yeah. But it's also like, it bleeds over into other.

Dave Jones: Yeah, it can do.

Chris Gammell: Mobile industries, basically. Like things that have basically anything that has like a motor controller and a battery system as well. It's probably going to have a CAN bus. Okay.

Dave Jones: Right. So like a humanoid robot or something might have a CAN bus.

Chris Gammell: Robot, scooters, that sort of thing. Right. It's all the same kind of stuff there.

Dave Jones: Yep.

Chris Gammell: I'm not doing like a lot of active work on it. I'm mostly like reviewing someone else's work, but I played with a very, very little bit. And what a hot mess. Like there's no standardization at all. Like, so like basically, I think this is maybe, maybe the same kind of effect that we're talking about here where the companies are just so big and I think secretive as well, you know, think about like automotive companies that they just, you know, so CAN is basically like a, it's, it is the physical layer. And I think one layer up from that in like the, the OSI diagrams.

Dave Jones: Basically. Well, in the automotive market, they're playing for all the marbles, right? You're not going to share. You're not going to be friendly with your competitors, right? You're going to lock in that standard for yourself so that, so that the car designer has to buy your chips, right? They don't want a second source, right?

Chris Gammell: They write all of the rules for their subcontractors and all like the OEMs and the OEMs might make the same thing for Toyota and Mitsubishi, but they might be completely different communication standards. It might be the same hardware with completely different firmware. Right.

Dave Jones: Yeah.

Chris Gammell: And like, yeah. So then, you know, and everything just is like shouting on the bus, you know, it's just like such a noisy, rude bus, you know.

Dave Jones: Noisy, rude bus. It's the Ethernet of serial buses, isn't it? Yeah. Right, right. For those who don't know, the Ethernet standard is incredibly rude. It's just like, you can jump on there anytime you like, and then they'll tell you to bugger off. Stop. And then you come back at another random time and try again and try to annoy the bus again. And, oh, if it happens to be available, yay, winner. Otherwise you keep trying at another random time. It's bloody, is there a ruder bus? Is there a ruder bus than the Ethernet? Rude, rude buses can be the name of this episode.

Chris Gammell: Yeah. I like that.

Dave Jones: What do you think? Rude buses.

Speaker ?: I like it.

Dave Jones: Yep. Yep. And, and, and you can have like a bar, you can do the, uh, mid journey thing and get a, well, after this we'll, we'll try the mid, we'll, we'll, we'll, we'll do a mid journey battle and see who can generate the best.

Chris Gammell: Okay.

Dave Jones: All right. Rude bus. Oh boy.

Chris Gammell: Yeah. I think, uh, with can, I've used a similar kind of thing on, uh, like a bass bastardized RS 45 in the past. And it was like the same kind of thing where it was like basically lowest address wins. And so at least can is a little bit kinder in that like you basically like everybody shouts their address first, I think. Yeah. And then whatever's listening for it, then you respond. It's, it's kind of like somewhere between an Ethernet and a I squared C and that there is, there is specific addressing though as well.

Dave Jones: Right.

Chris Gammell: Yeah.

Dave Jones: Hmm.

Chris Gammell: Yeah. So it's, um, it, I, I, I hopefully have more, more to report later, but, uh, it is, it is an interesting, interesting space to be in for sure.

Dave Jones: Oh, well.

Chris Gammell: Yeah.

Dave Jones: What else we got this week?

Chris Gammell: What else do we have?

Dave Jones: Oh, I, I do like practical engineering's video on how sensors keep bridges from, from, from collapsing. Have, so, so have you worked with a strain gauges at all?

Chris Gammell: No, just like not, not in any practical context, just, you know, like. They are very cool.

Dave Jones: Strain gauges are very cool. Like you can super glue them to any, you know, I've super glued them to all sorts of weird and wonderful things and, uh, or, you know, used epoxy or some other thing to attach them. And, you know, you can measure the stress and strain on your product. If you've got a physical product, then learning how to use a strain gauges is, uh, is a really useful thing, you know? So.

Chris Gammell: Yeah.

Dave Jones: Yeah.

Chris Gammell: Yeah. And he's showing it like with a kind of an industrial, so he is actually epoxying it. He shows a, an eye beam.

Dave Jones: Yep.

Chris Gammell: And then he stands on the eye beam and it shows the deflection and it's like much more than I would have guessed. You know, like.

Dave Jones: You can get really minute deflection. You can measure really, really minute stuff. Yeah. With a strain gauge. They're really remarkable. For those who don't know what a strain gauge is in this instance, it's basically a printed resister in like a zigzaggy pattern, you know, a big, you know, back and forth zigzaggy pattern thing that you, that's like on like a Mylar sheet. There's many different ways to do it, but like, you know, often the ones you'll get are on like a Mylar sheet. So they're printed this resistive thing. And then you glue down that Mylar sheet to your steel beam or whatever. And then when that Mylar sheet stretches, then the trace on there stretches and it changes its resistance. So you put it in a Wheatstone bridge thing, which is very good. A Wheatstone bridge of configuration, of course, is very good at picking up minute differences in a resistance and then it will give you a differential output and then you can amplify that and you can get. Yeah. It just changes the resistance ever so slightly.

Chris Gammell: Right. And it's already pretty high resistance to start with too, isn't it?

Dave Jones: Yeah. Yeah. They're yes. Yes. They're quite high. Some of them could be hundreds of ohms. You know, it depends. Oh, okay. All right. Right.

Chris Gammell: Yeah.

Dave Jones: Yep. And yeah, very cool. Very cool.

Chris Gammell: What did you use? So when you said you did it for products, like what were you measuring the strain on a product?

Dave Jones: Oh, we were, well, because we were manufacturing towed array streamers for not only, well, I worked on submarine ones. I've worked on ones for the survey in oil and stuff like that. So they drag behind the boat. So everything is torsional and, and, you know, everything's flexing. Everything's, you know.

Chris Gammell: But is it the case that you're measuring usually, or is it like the actual circuit?

Dave Jones: Oh, it can be case. It can be a strain on wiring harnesses. It can be strain on shells that hold hydrophones and transducers and other things. Yeah. I can imagine pressure vessels for sure. Yeah. All sorts of stuff. Target thing. It's, yeah. Yeah. Yeah. Stress and strain measurement. And yeah, I, I actually built my own, uh, amps for that, you know, so you build up your own amplifiers, you know, really precision. So you use a nice analog devices, you know, differential amp, you know, with bugger all offset and, you know, really nice, really nice stuff. You know, the chip is like 20 bucks. Yeah. Yeah.

Chris Gammell: That's great.

Dave Jones: That's awesome. Really fun stuff.

Chris Gammell: Yeah. That's cool.

Dave Jones: Anyway, I will have to watch that. That'll be my lunchtime video today. Yeah.

Chris Gammell: Yeah. That's a good one. That's a good one.

Dave Jones: Yep. Very, very, once again, that's another tool to having your toolkit. You know, if somebody goes, oh, you know, like if you're working at a company and you happen to be manufacturing some sort of physical product, you can go, oh, I can measure to see at what point that fails or at what, you know, see if it has any impact or, you know, things like that. So, yeah.

Chris Gammell: This is going to sound stupid, but I remember the first time that someone like just had a thermocouple and they were like, yeah, I'll just measure the, I just like glue it to this thing and just measure the temperature. Yeah. Like a chip or something like that. I was like.

Chris Gammell: Magic.

Chris Gammell: Oh, oh yeah. You could just do that. You can do that. You can learn so much. I know that it's like a, like a thing that now I'm like, yeah, of course you would. But like, then I was like, wait, you can just do that. You can just measure the temperature like that. And yeah, yeah, you can. It's, it's actually a really good way to do it.

Dave Jones: Can we come up with a list of valuable skills that you wouldn't, yeah. Like, you know, I have a particular set of skills. If you take my product.

Chris Gammell: Yeah. If you take my product, I will find you.

Dave Jones: Liam Neeson. Channel our inner Liam Neeson. And for those who don't get the reference, taken. Taken. Of course. Yeah. Yeah. And yeah, like, you know, thermal, thermal imaging, thermal measurement.

Chris Gammell: Yeah.

Dave Jones: Stress measurement with these strain gauges that we're talking about. I've done videos on vibration measurement using accelerometers and stuff. Right. Very useful skill for, you know, is my PCB going to survive transport? Is my PCB going to survive in an operational mode? Right. Huge, valuable skill to have.

Chris Gammell: How would you do that now? Like, just like.

Dave Jones: Exactly the same way. You would, you would glue an accelerometer. Super glue. You get an accelerometer, a little miniature one. And you would super glue it to your PCB, often in multiple locations, often in multiple orientations, because you want to shake it in different orientations, but you can use just one if you want, and then mount your board in different configurations when you shake it, you know, vertical, horizontal.

Chris Gammell: And you like, wire it off board and just measure it like out of a DAC or something like that?

Dave Jones: Yeah. Well, you send it into a charge amplifier, and then you send that into a dynamic signal analyzer, which then you can get the frequency sweep of. So, you know, the resonant point of your PCB, like how it's mounted. So, your piece of, oh, I've got this giant PCB, and it's only got four mounting points in the corners, and I've got this real heavy transformer in the middle. Well, dude, you're going to come and gut some. Right? Resonant frequencies, yeah, for sure. Just transporting your product on the truck to go to your customer can damage it. Right? So, there are standards for all these things. There's standards for transport of products, and you can test that your product meets those transport standards by rail, by road, by air, by ship, and all sorts of things. Right? Yeah, that's great. It's a huge rabbit hole there. And if you've got the skills to do that, once again, you're a valuable employee. Right?

Chris Gammell: Yeah, that's great. I think, I forget where I was talking about. I think this might have been at work, but I was talking to someone about the importance of getting off your bench, because it's so easy to lull yourself into a false sense of security of like, of course, I've done all this testing on my bench, and it works great. Right, right. But then, like you said, it's in your pocket, and oh, the corner exposed, or something on the board shorts to the case, or the chassis, and then it just shorts out or something.

Dave Jones: Or simply getting in the outside. Oh, it's in the sun. Oh, it's in the sun. Oh, it's 40 degrees now. Oh, it's not 20 degrees in the lab. Oh, it's snowing outside. It's zero degrees. Yeah. Does my battery work? I don't know.

Chris Gammell: Right? And just like, I think that would have been great advice to my younger self of just like, get it in situ as fast as possible, even though it's not perfect yet. But whatever amount of jewelry rig I need to do of super glue and tape and whatever to get it into a real enough situation, that's another skill I feel like that is important. Because I especially feel like young Chris, and sometimes current Chris, there's that safety of being on the bench. Yes. Yeah, yeah. No, totally.

Dave Jones: Totally.

Chris Gammell: It's a learned behavior to move it off the bench quickly. Yeah. And if you can do that, you're going to find so many more problems that you're going to save yourself many late night engineering adventures, I feel like. Not the good kind.

Dave Jones: That's one aspect of my rather boring career, my boring professional career that I'm actually glad of is that I worked on stuff that's used in the field, that's used in extreme environments, whether it's the bottom of the ocean, whether it's towed behind a ship, whether it's used on the back deck of the ship in 60 degree hot sun, you know, and also, you know, we like whether it's dropped from a helicopter, I've designed products that have to survive being dropped from a helicopter into the ocean, right? Huge impact forces. So shock, temperature, vibration, transport, all these sorts of, you know, temperature, extremes, all these sorts of things that, you know, I've had experience in is taught me quite a lot of valuable skill. It's taught me the importance that those things actually happen in the real world. So, yeah.

Chris Gammell: Yeah. And I think just kind of like, it's almost like being a prepper for your own board. A prepper. Electronics prepper. Yeah. Right. All the things that could go wrong, you kind of just, you trigger them. You do drills. Yeah, yeah. You know, like.

Dave Jones: So, you know, or whether or not you're designing a product for, you know, for that kids use, you know, you'd be surprised at the abuse that kids will, you know, put on a product or something like that. It's easy to design on the bench and everyone comes up and, oh, it works. They press the buttons nice and gently. And, you know, like. Yeah.

Chris Gammell: I feel like, like having like in-situ battery testing too, like with like real use case, you know. So like if you could get it again, like you could do, you could do a battery test on your bench and measure the, you know, the discharge curve, whatever. But can you also potentially measure it when it's in someone's pocket and they've depressed one of the buttons by accident. And like that actually is draining the battery as well because the light's on or something like that. You know, like all of these other failure modes that are just real world type of stuff. I don't know. How do you, how do you get to those points faster and still, and still hopefully measure them.

Dave Jones: Right. That, that gives me an idea for a video.

Chris Gammell: Yeah.

Dave Jones: Actually.

Chris Gammell: Yeah. I mean, this is a good, this is a definitely good fodder for video of like the, it's like a bootcamp. It's like a bootcamp. Yeah.

Dave Jones: It could be like top five skills you need to know.

Chris Gammell: Dave Jones bootcamp of measurements. And then you could actually like put on a hat and whistle. Yeah. Right.

Chris Gammell: I think I might. Listen up maggots. I'm going to teach you how to shake a box so hard or your circuit board falls apart. That makes a good whiteboard. You know.

Chris Gammell: Not whiteboard. Come on, man. What?

Dave Jones: No.

Chris Gammell: Put some theatrics into it. Come on. Yeah.

Dave Jones: Come on.

Chris Gammell: Come on. It's really easy for me to say as someone not doing all the recording.

Dave Jones: Yeah, exactly.

Chris Gammell: You should spend a little more time on that, Dave. Right. Okay.

Dave Jones: I know. I just want to churn out cheap and easy content. Oh boy. Anyway. Yep. Yeah. Valuable skills. Those, those skills that you don't, you don't learn these, right? You don't learn these in your typical engineering course, right?

Chris Gammell: No, you learn them on site. You learn them on site later. From someone who's already been through it.

Dave Jones: Yeah. Yeah. There is no course on vibration testing. You know, hands up. Oh, there is, but it might be. Maybe, but it's some niche thing that nobody, you know.

Chris Gammell: This is why co-ops are important. I think that's, that's the, because there's one job they're also going to make the co-op do. It's monitoring the vibration test. I mean, that was many of my friends had that as well.

Dave Jones: Yeah. But the thing is like, if you get, like, if you work at a company, you get work experience at a company and all they do is design, you know, some little PCB that doesn't get used in, you know, any real, like, you know, any like harsh environment or something like that. Then you're not going to consider any of these things.

Chris Gammell: See, that's another attack you could take on the video of like things to look for in jobs, like internships. Right. That would be definitely one of them. Okay.

Dave Jones: That's interesting. That's interesting. Take. Yeah.

Chris Gammell: Yeah. Because yeah, I, I definitely remember thinking, I remember thinking that like, I want to go to an internship at Intel. Right. That was like, that was like the top spot. And then the one guy that did it, he came back and he's like, yeah, I was reading Verilog in a cube for eight hours a day for all summer. Yeah. You know, it's like, oh yeah, of course.

Dave Jones: You know, that's right. Duh. Of course. You know, so all we got out of that is well, valuable Verilog skills, but the. I mean, probably a lot of money too. Prestep money and the, and, and the prestige of having that name on you.

Chris Gammell: Yeah. Yeah. It can, I mean, it totally can, it could springboard you into another thing, you know, some Intel on a resume might help the next internship or your first job, whatever, but, but it also might mean you're just put into another cube. Right. Yeah.

Dave Jones: Anyway, I'm, I'm going to, I'm going to think about a video. All right.

Chris Gammell: We've started the Dave engine here. We started the Dave engine. Well, your imagination is spinning. Mine is spinning from yet another antenna simulation from, oh, Sam, that his name?

Dave Jones: Sam.

Chris Gammell: Sam A. I think. Sam A. Yes. So if people saw, I think we mentioned this on the show a couple of weeks ago, it was basically, it was, he'd tagged us on Twitter with like a simulation of a transmission line. And it was like an overlay of a blender thing. And then he overlaid like the simulation using open EMS. And it was like, you could see the reflections. Super cool. Well, this new one, I really liked this, but we talked about a little bit as well about whether it's actually useful is a wifi.

Dave Jones: We talked about this just briefly before the show. It's like, yeah, it's very cool to see what he's done.

Chris Gammell: Oh, it is so cool to see. Like you have to go watch this right now. We will link in it, of course. But is it useful in a troubleshooting and or like design context? That's actually an interesting question. Yeah.

Dave Jones: That's what I was thinking about before we were discussing this before the show. I'm going, this looks really cool. And then it dawned on me. What am I actually looking at? Is this useful information to me?

Chris Gammell: Right. Because I had said to Dave, I said, I would love to be able to see this kind of like visualization of like, you know, so just as like an audio picture here, it's basically it's a, it's an F antenna and off of like, I think it's the TI recommend F antenna for like a 2.4 gigahertz. And then it's like the waves propagating off of this thing. And again, it looks so cool. But I had said to Dave, like I w I've always wanted this in like a, like a smart glasses kind of thing. You could like see this in the world. So you could see, you know, waves coming off your wifi router or your.

Dave Jones: And then how does it couple into the receiving antenna? There is another thing. Like, how does it like, yeah.

Chris Gammell: But does it actually change like your design stuff? So like, I think the, one of the key things that neither Dave or I have is like, okay, now you see this, what are you going to do to make it better or worse? Like that's really a missing piece there that neither of us have. Like, we're not gonna be like, oh, well, at the end that 90 degrees should be 80 degrees, you know?

Dave Jones: Yeah, we don't.

Chris Gammell: Yeah, exactly.

Dave Jones: Yeah. No, apart from the fact that, okay, I'm not going to put that next to a metal object, or I'm not going to have, have the user's hand covering it or something like common sense stuff like that. Right. But, but apart from that, it's like, well, I don't know, keep the antenna as far away from metal and humans and other objects as possible. And Bob's your uncle, you know, kind of like rule of thumb. I mean, I don't know.

Chris Gammell: Well, I think there are some RF experts that would, would have more knowledge and, and use case for this sort of thing, but that actionable intelligence, not, not for us so much.

Dave Jones: It's, it's really more of a skill for if you are designing the shoe phone, you know, you're designing the latest Apple shoe phone and you have to integrate the antenna on the PCB, you're going to go to town, right? You're going to ride that donkey all the way into town, right? Trying to figure out how, you know, like every minute detail of how that antenna radiates, right? Yeah. That's, that's super important, but you know, I, otherwise, I don't know. Yeah. It's cool. Um, but I can't see how I can practically, how that gives me a practical insight.

Chris Gammell: Right. What I think this is kind of like a, if you, if we zoom out a little bit from this, these absolutely gorgeous renderings and we should just restate them, Sam, very nice job. This is, please keep making these. I, I, I will look at these all day long. Uh, but like generally I think as well, like, like simulation has tons of value, but again, like then what is, what is your action as a result? I think, I think at this point, this is like, this is one of the things where I think I've talked to Mr. Jeffrey Kaiser about this before too. And I say like, well, what do you do then? What do you, what is it like? Yeah. And Jeff's like, look, you got to go back to fundamentals at that point then. Right. You need to make sure you understand your R's and your C's and your L's and that you understand how signals normally propagate in a best case scenario, worst case scenario, whatever. And I'm like, oh, right. I don't have, I don't have that. Exactly. Or I've forgotten it, you know? I've forgotten it. Yeah. Yeah.

Dave Jones: I've totally forgotten it. There's no way I'm getting it back.

Speaker ?: Yeah. Right.

Chris Gammell: And I think the other thing too, is that like, you know, so Sam's showing a inverted F antenna here with like the recommended pattern. If it doesn't work, I wouldn't say you go and if it doesn't work how you expect it to, I wouldn't go and start tweaking an antenna at that point. You know what I mean? I'd be like really be digging in to make sure like, is it not working because I did something wrong? Because like, this is a proven like reference design over and over again.

Dave Jones: Because I use the wrong, wrong dielectric PCB. Is it not working because other components are too close to it? Is it? Exactly. Have I got the wrong, wrong impedance trace going to it? You know? Right.

Chris Gammell: Yeah. And if you start from something like that, well-known and like that, like, and something that is in a reference design, not like you can't find it, but like the default assumption should always be like, what did I do versus what did they do? You know, it's not like I need to go and change, change the geometry here. It's like, no, I need to understand the actual situation first. And that's actually, I think where simulation could help you kind of dig in a little bit more and like understand more about it because it's putting visuals to something that does not, is not in the visual, visual spectrum. So that, that part's always really tough.

Dave Jones: You do know that everyone's just screaming at it. Everyone who knows stuff is just screaming at it. It's going, oh, you guys have no clue. Just shut up. Well, they ain't wrong.

Chris Gammell: They ain't wrong.

Dave Jones: Exactly. But anyway, yeah, good stuff.

Chris Gammell: Well, we're not making RF stuff that off anyways. I mostly buy prefab RF stuff. Thank you very much, which is a pretty good solution. The correct answer is buy stuff with SMA cables on it or SMA. Right. Yes.

Dave Jones: Oh boy. Yeah. Okay. What else we got?

Chris Gammell: I don't know if we have much time for much else. Yeah.

Dave Jones: So I am powers up last, last thing. I did like, uh, uh, uh, let your boom former guest. Yep. Made a very cool looking lamp, a fire, like a fire. He calls it a firefight lamp. I guess it is, but it's got little in inductive leads inside that, you know, light up and dance around and, you know, it's fan at the bottom and flies them around and, you know, it's cool.

Chris Gammell: In the, uh, in certain parts of the U S, including where I grew up, they were called lightning bugs, not fireflies. Right.

Dave Jones: There you go.

Chris Gammell: That's actually a way you can tell, uh, where people grew up in the U S is what they called. What they call the thing is that light up their bums.

Dave Jones: Got it. And, um, somebody did a smart, uh, uh, tear down of, I didn't know Ray-Ban are doing smart glasses. Yeah.

Chris Gammell: Becky did this. Yeah. Yep.

Dave Jones: And what the hell, uh, but why are we doing stuff? I thought they went the day way of the dodo smart glasses.

Chris Gammell: I am surprised actually that they're, that they're still doing this. Yeah. I mean, cause like, what's their name? Uh, snap snap chat. Yeah. Yeah. Another Snapchat. Snapchat was another, uh, social media thing.

Dave Jones: Oh yes. Right. Yeah.

Dave Jones: Snapchat prior to Tik TOK. Okay. Yeah. Right.

Chris Gammell: Yeah. Yeah. They had glasses as well that were a little bit chunkier than this. The Ray-Ban ones, they, they look kind of normal-ish.

Dave Jones: Yeah. But what do they do? What do you use them for? What's the killer app? What's the, you know?

Chris Gammell: Oh, the, you know, actually Dave, I happen to be holding my Google glass.

Dave Jones: Your ancient Google glass.

Chris Gammell: In my hand. It's actually, it's actually the, the headphone thingy is actually on my desk for some reason. Uh, and I, I, I play with it as we speak.

Dave Jones: Yeah. Yeah.

Chris Gammell: I think the only thing that was actually good about it was like, like having something in situ and like, that is, you don't have to pull your, there's always the creepy factor. There's no, no denying that there is the creepy factor, but being able to like capture quasi decent photos at the drop of a hat without pulling out a phone is so different. Right. And like, and now I'm not going to say liberating cause it's not, it feels like that's like a religious experience or something like that, but it's just.

Dave Jones: It's, it's just a convenience thing. It's so much more convenient. It's right on your head. And what do you, uh, to take a photo, you just tap the double tap the side or something. Yeah.

Chris Gammell: I don't know. I don't know. I don't know on the Ray-Ban ones, but it is having ubiquitous. The camera is like, you get to like, just have this.

Dave Jones: See for me, that is probably the killer app. Don't make it do anything else. Don't make it do any of the bullshit projection stuff. Just double, double click the side. Like it's always on sleeping. And then when you double tap, tap the side, it takes a photo of whatever you're looking at. And then it beeps to confirm that it's taken it. So you don't have to check.

Chris Gammell: Right. Yeah. Exactly. And then. Right. And if you want to check you later, you go look at your phone and you can download the photo.

Dave Jones: You build up confidence in it that. Yep. If it's beeped. Yep. It's taken the photo and it's got a wide angle, fixed focus, you know, like.

Chris Gammell: I think that's pretty much what this is doing. Right. I mean, cause it's, it's trying to kind of replicate just like the eyeballs. It's like, you're looking at someone else's eyeballs. That is an interesting perspective from like a film filmmaker kind of thing too. Right. It's not a common perspective, even if you put a phone in front of your face and it's not like it gets stereotypical.

Dave Jones: Yeah. What's this 3d heat map modeling thing software they got? They've got like, like, like an x-ray kind of.

Chris Gammell: I think sponsored the video. That's like a scanner thing.

Dave Jones: Right.

Chris Gammell: Yeah. That is, what is it called? Oh. How's it?

Dave Jones: I just want to know how that's working. How they've got a 3d.

Chris Gammell: That is a, a cat scan. You remember how I showed you a cat scan at one point? Yeah. Yeah.

Dave Jones: Because they're spinning the thing around. They've got the model. And so they must've mapped it. So they must've put it in this big 3d mapping x-ray box scanner box thing. And then got a 3d x-ray of it or something. And they're able to.

Chris Gammell: Yeah. That is. So I had access to one of those at mhub when I was there, and I think they still have it there, but it was based, or it wasn't a cat scan. Sorry. It was a CT scan.

Dave Jones: Yeah.

Chris Gammell: And then there was like a software they layered on top of it. This looks pretty similar. This actually looks, is it LumaField? Is that who it is? I think it's LumaField.

Dave Jones: Hmm. Don't know. Anyway.

Chris Gammell: And so I think that's who is doing this stuff. Okay. But I think they were one of the quasi sponsors of this video as well.

Dave Jones: Okay.

Chris Gammell: I don't know. I know one of the people that's there. I met one person who's there. So yeah, it's definitely cool software though. Yeah.

Dave Jones: But anyway, it's very, it's very impressive. Anyway, I'm kind of, kind of jealous of, no, no, I'm sure it's not cheap.

Chris Gammell: Not cheap in the slightest. It would be very, very expensive. But I think like the sort of thing where like, you know, when you need it, you really need it sort of thing. Yeah. So like same, same with that CT scan that I did. It was, I actually got a friend's manufacturing problem board. And like, he had like, this actually looks like this is not CT scan. Cause it's not, it doesn't have high enough resolution. Right. The one that I was doing, like, it actually had like, remember I showed you like the, the BGA balls, like you could actually see voiding underneath. Yeah. Like you can get 3d viewing and stuff. Yeah. Oh, so.

Dave Jones: Yeah.

Chris Gammell: It's so cool. And like, when you need it, you really, really need it, you know? Right. So, cause you're captive underneath parts. Becky does this teardown and she actually shows there's a Qualcomm chip with like encapsulate, like top by memory, you know? So like the, how they do memory stacked on top of processing chips as well.

Dave Jones: Right.

Chris Gammell: Yeah. So very, very complex assemblies like that. You kind of just, you can't really take it apart at that point. You just kind of need something to look inside of it, you know?

Dave Jones: Very cool. Great video.

Chris Gammell: Great video. I like, I like these kinds of things.

Dave Jones: I have to watch the whole thing. Yeah. All right. Well, that's it. Our MPL is up.

Chris Gammell: Yeah. All right. Well, good luck with Steve next week. Tell him I said hello, even though he doesn't know me.

Dave Jones: He doesn't know you from Bar of Soap, dude.

Chris Gammell: Please send my regards. Please send my regards. And I like their parts. Blah, blah, blah. Thanks for being on the amp hour.

Speaker ?: Right.

Chris Gammell: And please stick around for a whole hour, Steve. That's the other thing.

Dave Jones: All right.

Chris Gammell: And if you listen, Steve, as you're listening to this, or if one of your staff members is, more likely, ignore what we said about your parts prior, early in the show.

Dave Jones: That cancellation email is coming through right now. There it is. Okay. Cool. Catch you next time. See ya. Bye.

Speaker ?: Bye. ! Thank you.