#455 – Bill and Dave's Excellent Equipment

Bill: This is The Amp Hour Podcast. Released August 19th, 2019. Episode 455. Bill and Dave's Excellent Equipment.

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: Back with the same intro we've been doing for a decade.

Chris Gammell: Not a decade. I was actually thinking, I think August 10th was our anniversary. Oh, was it? Happy nine years. We're entering our 10th year. Yes, yes. Started in 2010. Yep. The 10th year of the Amp Hour commences, and we're sorry. Sorry. Yeah.

Dave Jones: That's about all we can say.

Chris Gammell: We'll try harder next decade. Right. Well, let's ask what's new then. What's new with you, Dave?

Dave Jones: What's new with me? Yeah, with you. There's not much new with me.

Chris Gammell: I don't know.

Dave Jones: We can talk about my latest video, if you like. Sure. The debacle that was.

Chris Gammell: Why don't we start with that? I've got some things to talk about. But yeah, I think let's start with the debacle. So this is the multimeter.

Dave Jones: This is the Vion multimeter. For those who aren't aware, it was a Kickstarter and Indiegogo, because you've got to have two sucks of the sav these days. Of course. Nobody just goes on one platform anymore. It's always dual platform, usually at the same time. You know? Yeah. Anyway. It's a thing. So they raise over half a million Yankee bucks, I think. Oh, wow. For this little dinky multimeter. It looks like two probes, basically, with a curly cord between them. And it's a Bluetooth multimeter that hooks up to your phone. You know, it doesn't do much unless you hook it up to your phone. It's got some leads in the tip of it to light up red, green, and blue and stuff to show you various statuses of things. But apart from that, yeah, you've got to hook it up to your phone app. And it was, they promised everything. Oh, my God. Yeah. And the worst part is they promised, like, a shopping cart experience, so that if you test a light bulb or a fuse and it's blown, it would automatically order you a new part.

Chris Gammell: Okay.

Dave Jones: Yeah. Instantly, you can kind of start seeing the problem with that, can't you? It's a marketplace. It's a marketplace. Oh, it's just unbelievable. For those whose brains aren't engaged at the moment because you're listening to the amp hour, what that means is that, well, they're probably on their exercise bike, you know, or walking the dog or whatever, you know, not just, you know, spacing out, listening to the amp hour. Sure. Who the hell wants to engage their brain? Takes energy.

Chris Gammell: Right.

Dave Jones: Anyway, let us analyze it for you. We, obviously, you have to know exactly the type, rating, and all the specs of the product that you need that you're actually testing first. So what's the point? Like, having a shopping cart thing that can just, anyway. Like, so dumb. And they promised it was, you know, like the world's easiest multimeter so that anyone can use it, of course. And it would, you know, just like nobody, people who know nothing about multimeters can use this thing. Right? Right? And you'd never have to call a repairman again because you'd be able to fix stuff yourself because you'd be able to test that something's broken and then automatically order it on your shopping cart. It'd come and you'd just replace it and you don't, oh God. It's, yeah. It's that bad. Anyway.

Chris Gammell: So that's a bad business model, but that's not a bad multimeter necessarily. So keep going. No.

Dave Jones: Okay. Yeah. Anyway, it actually took them a year. So they're, no, more than a year, year and a, you know, year and a quarter later. Which is not a problem, right? These things happen. And they were really good in their updates, actually. They were, you know, posting photos and they got bad parts from AliExpress and, you know.

Chris Gammell: Well, of course.

Dave Jones: It's like shock horror, right? So they, anyway, so they were really good at all the update and everything. And they kept having excuses, excuses. People said scam, scam, scam. But, you know, they were posting, you know, it was obvious that they weren't scamming. They were doing, you know, real hard work to get this thing to market. So hats off to them, right? And anyway, they finally shipped. But before they shipped, they did this update that said, oh, look. Well, you know, great news. We're about to ship. And sort of like in the small print at the bottom said, oh, by the way, the capacitance mode doesn't work. The inductor mode doesn't work. The voltage output mode doesn't work. We couldn't get that work in the app as buggy. It doesn't really work properly. We'll, like, need another month. It's not really a multimeter. It doesn't really do anything. It doesn't really do anything. But we're going to ship it anyway. Yay. And everyone started shouting, you know, refund. And they actually offered refunds. But I don't believe, I haven't heard that anyone's actually gotten a refund yet. So I'm not sure what the deal is there. But they did actually offer it. But anyway, they started shipping all these things out. And somebody sent me one. So it turned up in the mailbag. So I decided to do this video on it. And as always, there's a big thread over on the EEV blog forum called the world's silliest multimeter. Because the concept and the app just look so silly. It was just all the stuff they were trying to do with it. It was just dumb. But like you said, right? It's like, yeah, they'd eventually find out that all the stuff they'll try to do is dumb. But it could have been a potentially useful USB multimeter for want of, you know. I mean, you know, I'm not going to write home to my mom about it, right? But it could have worked. Anyway, I got one. And like...

Chris Gammell: So the idea is like that it's Bluetooth. Okay. So that's not bad.

Dave Jones: It's Bluetooth only.

Chris Gammell: Yeah. So what's the name? Mushimeter did that, right? We had Eric on the show.

Dave Jones: Yes, that was one of the first. There's the PoKit one, which was another Kickstarter successful one done by a company here in Sydney, which used to be David Tu's former employer, actually.

Chris Gammell: Oh, okay. Yeah. I mean, and like the concept isn't bad. I mean, like the idea of a wireless multimeter...

Dave Jones: It's not necessarily bad. There's a niche market for it. Like it's not going to set the world on fire, right? So...

Chris Gammell: Right. But I like the idea of like the voltage isolation and, you know, basically having streaming capabilities. You could be isolated. You could actually be like physically isolated from, you know, like I remember they showed it like within a test chamber on the other side of like... Right. Yep. Yeah, of course. When they were measuring, what's it called? Tesla coils, stuff like that. You know, you're far away.

Chris Gammell: Yeah, of course. You're safe. Yeah. So I like that idea.

Dave Jones: Inside. Yeah. Yep. Absolutely. But this wasn't really designed for... No, this was designed for Joe Average. ...to, you know, test things. Yeah, I guess... Test things around their home, you know.

Chris Gammell: At the end of the day, I'm wondering like who's going... Are people like browsing Kickstarter and Indiegogo still or is it like they're getting sucked up into like... They do. ...into like ad campaigns?

Dave Jones: No, they do. People browse the tech categories. They just sit there all day browsing. There's these super backers who back hundreds and hundreds of projects. And obviously, they're just... Yeah. And I've talked to a few of them. I know a couple of super backers and they say, yeah, I just browse the thing and anything that sounds interesting, I just want to... Sometimes they'll like just buy in for a dollar, right? I know a few people who bought into this Vion, even though they knew like it wasn't going to do what they claim. They bought in for a dollar just so that they could follow the saga, right?

Chris Gammell: Yeah, they're like looky-loos, you know, but you have to pay the ticket in order to get through the gate.

Dave Jones: Yeah, you have to pay the dollar or five dollars or whatever the minimum buy-in is to follow the hilarity just in case they do the private updates and stuff like that. Anyway...

Chris Gammell: I feel like there's like... That's like the potential for like a stronger community, you know, like... And obviously, a lot of people are great with like backing Patreon and of like different channels and stuff like that, but like it just feels like wasted energy, you know?

Dave Jones: Yeah, I know. It's... Yeah, but people just love their tech gadgets and they just, you know, and granted, Kickstarter is a place to find new... I'm not going to... Yeah, I'm going to put innovative in quote marks, right? Because we don't need another 3D printer, you know? It's like... Right. Or another cooler or another, you know?

Chris Gammell: Yeah, the coolest cooler.

Dave Jones: The coolest cooler. And... But yeah, like people just love their tech stuff, so they just browse. And maybe you can get... I don't know. I've never tried. Can you get like email alerts from Kickstarter and Indiegogo about new programs? Maybe. Anyway, that's where a ton of people come from.

Chris Gammell: Yeah, I was wondering... I had this discussion with someone recently. I forget who. But I was like... I was trying to estimate like... You know, so... Oh, that's... It was someone who actually makes dev boards. That's what it was. And I was kind of asking like, well, what percentage of the boards that are kind of like... That are ordered, right? At the beginning of a project, you know, you see a flashy video and you see like, oh, this is all the things you could do with this thing. What percentage are actually using them? Because I know I have a box. I literally have a box of dev boards, right? Yeah, same thing. And like... And like some are back to the campaign. Some of them are just, oh, that looks interesting. It's like the story that I told myself was like, oh, I'll definitely use this for something. But like, of course not, you know? No, that's right. Just from a practical perspective, like I can't use every platform. I can't... It's stupid to do that and keep trying these things. Anyways, like... So what is the percentage of like utilization of dev boards? And then I said it, you know, the extreme example would be like, I think Raspberry Pi is at like 25, 30 million boards sold at this point. How many are actually in like active use on a daily basis?

Dave Jones: I would be stunned if it's 10%.

Chris Gammell: Oh, yeah. I would be stunned if it's reaching double digits.

Dave Jones: Really? Yeah. Yeah. Yeah. I wouldn't argue with that. Yeah.

Chris Gammell: Just because I think about like, okay, so like 10%. So like, let's just use... Let's use 10 million. That's an easy number, even though I know it's more, right? So that would then mean like, so 10% of that would be 1 million. 1% would be 100,000. Like 100,000 people using a Raspberry Pi every day? Yeah, maybe. I don't know.

Dave Jones: Yeah, maybe. Yeah.

Chris Gammell: I guess you'd have to multiply that by four or whatever they've sold all told. Yeah. But yeah. Yeah. It's just an interesting thing. I just feel like a lot of dev boards and projects and stuff like that are supported and like, you know, God bless them because that props up to people who are using it. Yeah, yeah. That's right.

Dave Jones: And it makes these things actually exist for when you do need them. Yeah. I mean, just imagine if we didn't have the ecosystem of dev boards that we have now. Sure. They actually take for granted. Now, like if you want to get a job done, right, there's a dev board or, you know, some sort of module out there to do it.

Chris Gammell: Yeah. It's great. Whereas when I was a boy... As a consultant now, it definitely has saved my bacon a couple of times.

Dave Jones: Exactly.

Chris Gammell: Yeah. Yeah.

Dave Jones: Yeah. But it's obviously dependent on the product. Dev boards are like that, but something like a multimeter is like, it's a tool that you can use daily. Sure. Sure. So, you know, it's much greater usage.

Chris Gammell: Yeah. Yeah. Well, and I think that what that points to is that like in any of these projects where you're doing crowdfunding or anything like that, you know, it can't just be like, hey, here are the raw specs. It has to also be like, you know, here's the project you could do with it. Even, you know, even if it's kind of contrived, it's like, yeah, you know, so it's like you have another microcontroller platform. Yeah. You could do a lot of things with it, but you still need to show a couple of things that you did do with it and actual thing, actual like end products or projects rather so that people are like, I want to do that thing, you know, like. Exactly.

Dave Jones: Or I might need to do that thing or I better have it in my toolkit.

Chris Gammell: Yeah. Right.

Dave Jones: Just in case, you know.

Chris Gammell: I don't know when I need to set up a thousand tap for a filter, you know, I might need to do that. That's right. Never know. I should have that, you know, Xilinx Zinc dev board, you know, I need to pay $200 for this right now. I may have had this conversation in my head before.

Dave Jones: Well, you shouldn't laugh, but it is kind of funny how, you know, most of them, like only single digit percentages end up being, you know, used for angle. Like people usually get it and then power on and have a fiddle with the app that comes out of the box. I'd say, you know, probably more than half would be doing that. Right. But after that, it's like, oh yeah, yeah, it works. Let's light up. Yep. Yep. Got my warm, fuzzy feeling knowing that I backed this project and turned up. Yep.

Chris Gammell: Yeah.

Dave Jones: But in the case of the Vion, it's like there, nobody, this is going straight. It's going straight in the bin as soon as people got it. There are people who have paid for it and it's come into their country and they've got, you know, it's stuck in customs or whatever. So they've got to pay, you know, an extra $20 to get it through customs in these pain in the ass countries. And yeah, all you see in the comments is, nope, I didn't even bother. I didn't even bother paying it. It's just, no, I'm not going to spend any more money on this. It's like, just based on, just based on their own updates of the thing, let alone when, when my review hits the airwaves, you know.

Chris Gammell: Yeah. Right.

Dave Jones: Oh boy.

Chris Gammell: Yep. That's funny.

Dave Jones: Anyway. Yeah. Like there was like nothing on it that was commended, like nothing on it. That was good. I think the only thing I actually commended was that it was a reasonably fast update rate. That was it. Like everything else either didn't work or was flawed or was, you know, oh my, like, wow. Wow.

Chris Gammell: Yeah.

Dave Jones: It's, yeah. So unfortunately, like the hardware, it might've been useful, you know, if, if the app was perfect, but the app was just absolute garbage. So, yeah. That's too bad. Unfortunately, the app makes the product.

Dave Jones: Right. Right. Right. And anyway.

Chris Gammell: Well, that's something that could be updated, but it's not necessarily going to be. It could be.

Dave Jones: It's not going to be. No. Unless somebody reverse engineers the whole thing and then just does their own from scratch. Nobody's going to bother because the hardware is not that great anyway. And the, one of the funniest things was is, is like, it's the world's simplest multimeter. Right. But the app has this sensitivity button that you can sensitivity plus minus. It doesn't explain what it does. And if you hit it, it can give you completely wrong values. I'm, I'm, I'm, I'm talking like five volts from a nine volt battery or 10 and a half volts from a nine volt battery. Like, wow. You know, it's, yeah. Why is that bad?

Chris Gammell: The precision, the digits of precision, you know, go the wrong direction there. Yeah. Just rounding up or rounding down.

Dave Jones: Oh, from, from a technical point of view though, right. This is actually a real interesting example of where they're completely kamigata because they didn't use a multimeter chipset in the front end. Right. They've just got a microcontroller on there. It's one of those Bluetoothy, you know, um, micro, you know, integrated thing. So presumably. Yeah. Yes. I think, yeah, I think it's one of those and use, yeah, it's got to be a building analog to digital converters or whatnot. And, you know, so they're using all that. So they're doing all their own switching and everything. You know, so there's a bunch of transistors on there that just switch in the various, you know, uh, loads and configurations for the different functionality. Right. Whether or not it's measuring resistance, it's measuring voltage, AC or DC. And, uh, and so they rolled all that themselves. And I've, I'm not sure if you've seen inside the data sheet of a multimeter chipset, but they're even like a $2 multimeter, right? They are incredibly complex. There's so much going on to make and to give a multimeter its functionality, especially it's auto arranging functionality and every, and all the different modes that a multimeter in quote marks can actually do. And it's that, and, but they, they chose, oh no, we'll just use a microcontroller with an ADC, which is okay, fine. But you've got to know what the hell you're doing. And, uh, yeah, they've completely come and guts up.

Chris Gammell: So I haven't seen the, uh, the chipset, but I've seen the, uh, uh, what it takes to actually have a device do that when I used to work at a test measurement company, you know, like basically it was, uh, it was the more complicated version of what they were trying to do. And it's like, no, no, no, that, that was like a huge software stack to get all those things doing. And then, you know, you have to have all the switching, right. You have to have all the timing, right. And the capacitance on your gate set up so that you're, you know, switching at the right time too. So it's, uh, it's pretty, it's really important.

Dave Jones: I just sent you a link to the one used inside the 121 GW multimeter. Oh yeah.

Chris Gammell: Okay.

Dave Jones: And, and, and if you scroll down to like page 21, right, you'll, you'll, you'll start seeing the function network of what's, you know, happening inside a, like an example app. It's just, yeah. Yeah.

Chris Gammell: This is like a hard logic version of, of what was happening inside of, you know, inside

Dave Jones: a Keith Lee electrometer or something. Yeah.

Chris Gammell: Yeah.

Dave Jones: Yep. And yeah, it's like, okay, yes, you can roll your own, but no.

Chris Gammell: So they, but not for the low costs that you really, yeah, not in a short amount of time.

Dave Jones: Exactly. And no, you just, it's hard enough to drive these things, right? Let alone if you might use a multimeter chipset, but there's hundreds of registers in there that if you don't set them up, right, you're going to completely come a gutter, but at least you can fix that in software.

Chris Gammell: Yeah. Right.

Dave Jones: You fix that in firmware. But if you screw it up in hardware, you are gonski.

Chris Gammell: Yep.

Dave Jones: So yeah. Anyway, so they had to add this sensitivity. I thought it was hilarious that they had to add this sensitivity feature. And it's like, yeah, Joe Average is going to know what sensitivity is. And, and they're going to think that their nine volt battery is given out five volts. Yep. Like, wow. Anyway, complete and epic file. We were talking about your former employer, Keith Lee.

Chris Gammell: We were. And you were making fun of me.

Dave Jones: Just before the show, I was making fun of you because you just casually dropped.

Chris Gammell: GPib.

Dave Jones: We called it GPib. GPib. GPib. And, and I instantly like, and my mind, my mind just rattled for a second. And I kind of got like a, like a twitch in my neck. And it's like, what? Nobody calls it GPib.

Chris Gammell: On the other side of the pond they do. GPib. If you don't know what.

Dave Jones: But apparently you say they do it Keith Lee. That's the.

Chris Gammell: That's the ones that I knew there. Yeah. Yeah.

Dave Jones: Wow. Really? I've never heard anyone call it GPib. It's always GPib.

Chris Gammell: I'd be okay with it going away. Uh, it's, it is a, uh, old and tired standard, but, uh, it's out there still. And you, you had made a video about it. That's why I brought it up.

Dave Jones: I did a recent video about an ethernet interface. Yeah. Which is. Yeah.

Chris Gammell: It's a good idea. I mean, like it really is, um, to have that kind of thing out there.

Dave Jones: How you can network. Like it's, it's not good if you just want to control one instrument. What it's good is if you want to network like 20 different instruments in a rack. Right. Like if, if you don't know, GPib is like a parallel, uh, multi-drop bus that you, so you stack these cables all in parallel, all these instruments. And it can have up to 30 addresses if memory serves me correctly. So you can have up to like 30 instruments. Um, and it's kind of like that. I don't know. Do you remember that?

Chris Gammell: Did you see that tweet that the, I forget who it was. I'm looking up on Twitter. Someone had set up like that crazy, crazy setup. Here it is. Robbie Newell. Newell. Sorry. Robbie Newell.

Dave Jones: Oh, is that that one in like, in like, there was just like three or four days ago. Battle station is, uh, armed and operational. That was great. Yeah. I retweeted. Oh, that's just brilliant. That's like a research.

Chris Gammell: Yeah.

Dave Jones: It was, uh, set up. It was for.

Chris Gammell: Yeah. So it was lock-in amplifiers. What was it doing? And there's super, there's two superconducting magnets and a bunch of measurements around it. Yeah. We'll share the Twitter thread. It really was.

Dave Jones: Do you see the, uh, like a pegboard they had with the like splitters and everything? So they had these. Oh no. Uh, splitters attached to these pegboards and then the coaxes running out of them. And, and yeah, rather than just have it like flapping around in the breeze, I actually went to the effort to stick it onto these panels. It was, it's really great. Yeah. It was really great. Have you ever used one of those lock-in, uh, amplifiers?

Chris Gammell: No, like a Stanford research system kind of thing? Yeah. Yeah. Stanford research lock-in amp. I was always doing like, like that's like higher frequency stuff, right?

Dave Jones: It's, well, it's no, no, it doesn't have to be. It can be extracting, uh, basically if you don't know what a lock-in amplifier is, it's the, it gives you the ability to extract your expected signal out of the noise. Like, so if you've got, you know, like they're often using physics experiments. I won't pretend to understand why or, you know, how, but, uh, yeah. So if, so if you're expecting a, like a, a, a one, a one megahertz carrier to be coming back from some test thing, but it's buried in like volts of noise, you're expecting a, you know, a five microvolt signal to come back, but it's buried in volts of noise. A lock-in amplifier can extract that signal out of ridiculously high noise. It can go down to nanovolts even, right? It can extract a signal down in the nanovolt range, like crazy stuff. So apparently it's, it's big in the physics community. So as you can see in that photo, uh, this is a visual. They were doing too.

Chris Gammell: So further down, it says they were testing 2D crystals like graphene, multiple taps bonded sample sandwiched between two plates, inducing a variable electric field running in the three helium, four helium dilution fridge. I assume that's the, uh, isotope. And then ramping up to plus minus 18 Tesla. Oh my God.

Dave Jones: For those who don't know, a Tesla is a pretty big unit.

Chris Gammell: That's right.

Dave Jones: Like a, uh, Farad. A Farad is a pretty big unit. Right, right.

Chris Gammell: When you get to Farad's, you're, yeah, you're really, you're really chooching there. Uh, what, uh, I feel like, you know, like I look at this kind of stuff too. And I'm like, man, I just don't do anything important at all. You know? No, yeah, yeah, I know. These kind of like science experiment things like there too, like obviously this is well funded. It's got a lot of equipment in here too. Like, but man, this like raw science stuff. Holy shit. Like they really, well, hats off, Robbie.

Dave Jones: I've done some raw research stuff, but it's not, you know, physics research. But I've done, you know, research on, uh, crystals and things like that. Um, and yeah, yeah, it's fun. Like it really feels like you're really doing, you know, real meaningful experiments. It's like, oh, nobody's ever, I don't think anyone's ever done this before. It's like, wow, what are we going to discover? Right. Kind of thing. Or maybe someone has, but it's in some obscure research paper that you can't find on the interwebs, you know? So you've got to just run the experience yourself and find out what happens.

Chris Gammell: Yeah.

Dave Jones: So.

Chris Gammell: Yeah. It's the same kind of feeling when I look at like space, like older space, not even older space stuff, like space stuff too. Where like people are like doing this kind of stuff, like launching stuff into space. I'm just like, uh, I don't even, where do you even start? You know, and obviously there's tons of institutional knowledge. You would never do it as an, as a solo person. Uh, you know, so you're part of a bigger team.

Dave Jones: And then like even, even SpaceX, for example, they, you know, it's not like they just hired all these green engineers. No, they, they hired like, you know, people who had done self, yeah. People who had done self landing rockets before. Right. Because they didn't invent it. It was like, yeah, it was done in the eighties. Right. So they hired a lot of these old people and they knew how to do it. Oh yeah.

Chris Gammell: Actually. I don't think I mentioned that on here before. One of my friends from co-op, uh, he, he had contacted me to be on the show at one point. So I'm not going to say his name here, but, uh, and hopefully he still will be. But like one day I found out, I was like, you know, it's kind of lost track of him. And then someone's like, Hey, yeah, have you talked to him in a while? I'm like, no. He's like, yeah, did you see, let me send you a link. They sent me a link of that grasshopper video. Remember the early Tesla, like the rocket that went up and like then hovered and landed again. And he's like, someone, they're like, yeah, uh, he, uh, he, he designed the control system for that. And it's like, oh my God. Jeez. I mean, like, yeah, people doing that kind of thing. Unsubstantiated too, but like, you know, like that's the kind of thing where it's just like, you know, uh, that's, that's some legit things. Whoever designed that.

Dave Jones: What the hell were we talking about?

Chris Gammell: We were talking about research and before that, uh, GPIB. Yeah. Oh, you just said it. GPIB. Yeah.

Dave Jones: Does anyone else call it GPIB outside of Keithley? Please let us know.

Chris Gammell: Yeah. That'd be good to know. Wow. Yeah. Yeah. I mean, the, the real fun thing with GPIB is when you're, when you've got like, when you do have the stacks and stacks of machines kind of like networked together and you start like piling up the cables, you know, and they're like, you know, cause like the, the cable itself is probably what? Like, I don't know.

Dave Jones: Oh, they're big and thick and they're heavy and they're like 10 centimeters each one.

Chris Gammell: I mean, they're so, they really start to stack up and they're heavy. Yeah. The cables themselves are super heavy and they're like, they're just industrial old school and they screw together because they'll fall off if you don't cause you got so much weight to them. Uh, they're not fun. I'm not a big fan, but, uh, but like, but you, like you said in that video too, um, it's a good way to get access to old, uh, to old gear, which is really nice. And if you can get it on the internet, then, you know, you can start to do more scripting. The scripting with it, with GPIB itself is pretty rough. I mean, like if you look at the language that most of them have accepted, it's, it's pretty annoying, but like, you know, then you put it in a Python and you start to, or something similar, you could, you could really start to make it a little bit easier and, uh, yeah. Get some old gear off eBay too. You know, all that stuff can really start to help.

Dave Jones: Cause all, because all that old gear, they haven't improved on the actual performance that much, right? They've just like, you know, changed the UI and they've, you know, done a few other. Oh, well, some of them. Nicies. It's like, you know, some of them are like, like I've got a Keithley Electrometer here, right? It's only beat by like one other, like it's like 40 years old or something. It's only beat by what, like one other on the market, I think, and not by much. It's like, yeah, it's, it's performances pretty kick ass. Right.

Chris Gammell: But it's. Yeah. I mean, yeah, I guess the, the limits of, uh, you know, component sourcing and stuff like that. If you're gonna have like a really good resistor, it's always gonna be really good. Well, not always, but that, you know, whatever the defining spec is.

Dave Jones: Right.

Chris Gammell: But some of the, you know, some of the silicon in there has gotten better. I think it depends on what you're doing.

Dave Jones: Of course. And they've gotten smaller as well. Like I just did a teardown of a gorgeous old HP, uh, jitter analyzer. Did you see that video?

Chris Gammell: I did not see that one. No.

Dave Jones: Oh, you have to. Oh, it's gorgeous. It's gorgeous. It's got these, you know, it's all plugin boards and it's huge, right? This thing, this thing today would fit on a Raspberry Pi size board. Right. Yeah. But it's just like you open up and the engineering that goes into that and the service ability, it comes with a built-in, uh, service extender card in, in the slot. So you can actually lift the boards out, put it on the extender card so that you can work on it. Yeah. Like, oh. That's nice. That's nice. Oh, it's beautiful.

Chris Gammell: Resistor color. Someone pointed out it was resistor color coded. Yes. Okay. I saw that tweet about it. I did an update on my second video. Yeah.

Dave Jones: Yeah. Yep. Yep. They, if, if you don't know, they lay like the little handles that allow you to pull up the cards, they actually color coded those. It turns out they were in the resistor color code order. Yeah. Black, brown, red, orange, yellow, green, blue, purple, gray, white. That is nice. Somebody at HP was like, yep, I'm doing these as color code, as the color code. I know how to get this done.

Chris Gammell: Yeah.

Dave Jones: Yeah.

Chris Gammell: That's great.

Dave Jones: Oh, fantastic. And that old gear is just, yeah, it's, but, but it's big and it's, you know, power hungry and, you know, big fans were and, yep.

Chris Gammell: Yeah.

Dave Jones: Yeah. So, but anyway, they don't make them like that anymore. For repairability, serviceability things. Yeah. Well, they still do in some like big in, in industrial and other markets. They do make, and military as we'll get into. Sure. They, you know, make things repairable still and they make them, you know, because like space really isn't like a real huge issue for those, you know, for things that go inside racks and, you know, stuff like that, like size isn't really a big deal. Right. So you can often make things serviceable.

Chris Gammell: Always like the perceived, uh, uh, sturdiness too. There is that.

Dave Jones: Oh yes, of course. Yeah. Yeah. Of course. Yeah. It might be able to fit inside a little raspberry pie size board, but you need to put it in a big 19 inch rack to make it look. Right. You know, big and grunty and yeah.

Chris Gammell: Yeah.

Dave Jones: Yeah.

Chris Gammell: I was really surprised. I was talking to someone today about the, uh, the, that old, the raggle scope that you've done a bunch of stuff on, like the upgradable one. The, I think the 1054, not the 1052, the 1054 Zed, which is hackable.

Dave Jones: Well, yeah, that one. Yeah. Yeah.

Chris Gammell: Yep. Um, that's 200 bucks now. That's crazy.

Dave Jones: Like it's 200 bucks from where?

Chris Gammell: That's what, that's what my friend said. Uh, I don't know, but like, because there's some new version that's out there, maybe it was like a, like that was like remaining stock or something.

Dave Jones: It's gotta be a remaining stock. Wow. Cause that's, that's crazy. Like 400 bucks used to be the going price. It actually used to be 800, right? Right, right. When I first got the 1052 E, which was in my first, very first video, I actually, actually reviewed that. My first ever video, that was $800. Then, then the price went to 400, like a year or two later. And then it sat at 400 forever. Yeah. And, you know.

Chris Gammell: I kind of put that at like the lower limit of like, you know, an actual scope of the screen and probes and stuff like that. But maybe not. Maybe it's, and that's the thing. Like, you know, things keep getting cheaper, but there are some limits on like, you know, you got enclosures, you got some decent chips in there. And I guess maybe this is kind of, if, if the scope started to go towards like how the multimeters were, if there was more custom silicon that could not custom silicon, but I guess like cheaper commoditized silicon because everything has, you know, silicon in it. Yeah. We're like, like the multimeter chip you brought up. Like if there was that for the scope side of things, that would be interesting to keep bringing down the cost of scopes. But I'll have to double check on that.

Dave Jones: It's, I've, I pulled it up here. One of the cheapest prices on sale is $349. So I don't know where 200 bucks came from. Wow. That's like, geez, that'd be, geez, you'd just buy a spare one for that. Yeah. Right. You know, if it was like, you just buy a second one because then you can have an eight channel scope. Yeah. Right. You know, it's like, yeah, you would just, that'd be a no brainer. Geez. You know, that's, that's just nuts. The speaking about custom silicon and Rygold, that's interesting because you know how Rygold rolled their own, they're now one of the big players. They rolled their own ASIC. It took them years to do. Right.

Chris Gammell: Oh, I thought they were still doing FPGA stuff. No, no, no.

Dave Jones: They rolled their own ASICs. They're the new 5,000, 7,000 series scopes. They use the new Phoenix chipset, they call it. And yeah. And they're one of the big players because they rolled their own ASIC. But if you use any of their new scopes using this new ASIC, it lacks the high res box car averaging feature, which like every scope on the market has. And for some reason, it's mysteriously missing in all of the new Rygold models. And they don't, and they don't want to say why. I suspect, my theory is that it's not they forgot to put it in. My theory is that they put it into the chipset and it didn't work. And because it was the ASIC and they don't want to spend the millions of dollars to re-mask that thing, they just went, well, we'll just switch it off and hope nobody notices. So if you do know, if you do know the exact reason why Rygold's new, brand new scopes don't have high res feature, then please let us know. Can you explain that?

Chris Gammell: Sorry, I'm not enough of a scope person to know what you're talking about.

Dave Jones: Oh, right. Most scopes have an 8-bit analog to digital converter. Sure. But you can enable a high res mode where it does a box car averaging. A box car averaging, that's just the technical name for the averaging filter, used. But it basically uses a rolling average filter after the ADC stage, like, you know, at the sampling stage before it gets processed and everything else, so that by averaging your samples, you can get a higher effective number of bits. Ah, I see. So you get a cleaner waveform, right? You get a higher resolution. And some scopes, you can actually, the way they do that is in the user interface, it'll go, how many bits do you want? It's like, do you want 8 bits, which is your normal mode? Do you want 9 bits, 10 bits, 11 bits, 12 bits, right? It doesn't have a 12-bit analog to digital converter in. It's only got an 8-bit one. But by doing the box car averaging on the fly, you can get an effective greater number of bits.

Chris Gammell: Do you lower your sample rate because, like, your effective sample rate drops then too?

Dave Jones: You do effectively lower your, yeah. Yes, yes, you do. So, yep. You see, yeah, but it does, like, so it's not done at the post-processing stage.

Chris Gammell: Yeah, it's not in the GUI where you're just, like, doing an overlay.

Dave Jones: It's not in the GUI where it's just doing, no. It's actually just at the output of the analog. It's inside the sampling chip, right? Sure, yeah, yeah. The ASIC or the FPGA. It's implementing this rolling box car average.

Chris Gammell: Yeah, okay, okay.

Dave Jones: And, yeah, and, you know, and it's effective technique. And even the cheapest scopes, even the $200 scopes have this.

Chris Gammell: Right, you get free cleaner, quote-unquote, signals, right?

Dave Jones: Yes, yeah, exactly. So if your waveform looks noisy, in quote marks, I've done two videos on that of why digital scopes look noisy. If you don't know, there are technical reasons behind that. And compared to an old-school analog scope, the answer is they aren't. They aren't inherently noisier. It's just something else is going on.

Chris Gammell: Right, it's just you're seeing more of it, yeah, or the sampling errors or something. Yep.

Dave Jones: But anyway, you have to watch my videos to understand that. And, like, they have this brilliant feature in there, and it's just – Riegel don't have it.

Chris Gammell: It's just gone.

Dave Jones: The lineeoscopes is gone. It's gone. It's unbelievable.

Chris Gammell: These things happen. Oh, man. Can you imagine?

Dave Jones: So if you know the story behind that, please let us know.

Chris Gammell: Yeah, it's like the – man, can you imagine, like, just getting that wrong? It's like, you know, I get boards wrong sometimes, right? You get boards wrong sometimes. There ain't no bodge in a chip.

Dave Jones: Has anyone out there gotten a mask wrong? Have they goofed it?

Chris Gammell: Oh, yeah.

Dave Jones: Jerry, Jerry Ellsworth, I think. She told the story about how her – when she was working at that toy company, she told the story about how she wanted to skip the country because, like, she was in China or Singapore or somewhere. She was working on this toy chip and it was, like, her first ever ASIC. Like, she had no idea what she was doing, but she thought she'd have a go anyway, right? And the chip – she designed the ASIC for the Commodore 64, you know, the all-in-one chip. And the chip came back and it didn't work. And it was like – and she thought about – seriously thought about skipping the country. It's a hilarious story. We might have to link it in. Yeah. Was that on the first episode?

Chris Gammell: I mean, it was the first one? Okay. I was wondering because we had Jerry on, like, four times, five times.

Dave Jones: Right. And she wanted to skip the country. But then she thought, nah, look, I'll have a last crack at this. And she figured out that there was some unused pin on the chip and she could put it into some weird-ass boot mode that would overcome this problem. And she was able to hack it with just, like, a pin lift or something. So she lifted a pin and tied in a thing and, boom, fixed it. And, whew, it's like, yeah, she figured out this way to overcome a silicon – actually goofing up the silicon. Yep. So – but that was really lucky. Yeah. That was like, yeah. Oh, boy. Yeah. So if anyone out there has screwed up a huge ASIC like that, let us know. I don't know. Because there's no way they could have forgotten. Yeah. Would you admit it? Yeah, exactly. Yeah. So, anyway. Hmm. Yep. Speaking of –

Chris Gammell: Episode 35. Episode 35. 35? There you go.

Dave Jones: Yeah. Okay. Speaking of goofs, the U.S. Navy goofed big time.

Chris Gammell: Yeah.

Dave Jones: And I was talking to Chris about this before the show and I'm surprised it got this far. I am not surprised it got this far. What?

Chris Gammell: I'm not surprised it got this far.

Dave Jones: You're not this – you're not surprised? Well, I am, having worked in the Navy – having worked in the military and the Navy industry. I didn't – what didn't work in the Navy, but I worked, you know, as a contractor for the Navy. And the headline is, the U.S. Navy will replace its touchscreen controls with mechanical ones on its destroyers.

Chris Gammell: Yeah. Yeah. Oops. UI problems.

Dave Jones: UI problems. Apparently, there was a collision in 2017 and they've done the study, you know, to find out what the heck happened. Was it a Liberian oil tanker or something? The USS John McCain actually collided with a Liberian oil tanker. And they haven't read the whole thing, but it basically – yeah, it killed 10 sailors, right? This was a serious, you know, serious damn collision. So, they investigated this to the hilt. And it looks like they found that the touchscreens that they were using to actually control the ship, right? Actually steer the ship. They just – yeah, it somehow goofed it.

Chris Gammell: I'm going to step outside of reality for a while here. Yeah. Since this is, you know, this is a serious thing. And I'm going to talk about the Avengers because on the Avengers, they also have a ship that has a bunch of touchscreens too. And then, like –

Dave Jones: Oh, I've seen that. Yes.

Chris Gammell: Yeah, and then, like – Yeah, yeah, yeah. And then everything goes dark. And I just always wonder, like, okay, say you were going to have a – your redundant power system needs are just, like, amplified, aren't they? Because now all of your touchscreen controllers have to be bulletproof. All of your power systems have to be bulletproof. Everything else. And, you know, in the movie, at least, they're like, oh, no, the power is gone. And then they're still sitting there hitting the – tapping the screen. Tap, tap, tap. Tap, tap. Tap, tap, tap. Tap, tap, yeah, yeah, yeah. And then, like, I don't know. Blue screen of death is a thing, you know? Like, what operating system are these things running? You know? Like, I don't know.

Dave Jones: I guarantee these are running a version of Windows. Guarantee.

Chris Gammell: Windows CE.

Dave Jones: I don't know about that, but they'd be running a variation of that, I'm sure. Yeah. Yeah. Oh, boy. That's just – yeah. No, it's not good. It's like the planes that have fly-by-wired out instead of the cables. Yeah. There is a reason. There is a very good reason why Air Force One, spared no expense, is still a 747.

Chris Gammell: Mm.

Dave Jones: Because – Why? Because it uses full manual cable controls. They can be a – you know, they can fly through some nuclear blast EMP thing or whatever, takes out every computer on that plane, every piece of electronics on that plane. They can still land that thing, right? Because they've got the manual controls, which go through cables right through to the back of the plane. They go to the ailerons. They go to the flaps. Like, you know.

Chris Gammell: Yeah.

Dave Jones: These manual controls, they can still land it. Yeah. Yeah. I don't know.

Chris Gammell: Like, I feel like some of it is – you know, so we're moving, obviously, into an age – not moving into – we're in an age where, like, a lot of things are drive-by-wire, right? Even cars these days are – and, you know, all the car people are probably screaming at their radios right now, too.

Dave Jones: The throttle does not – the actual throttle does not have a cable which goes to a little thing which opens the valve. Right. Right. Exactly. Right. It's – but they still make it feel like that's the case, right? Sure, sure. So they actually put these resistance, these artificial resistance things in the pedal to make it feel like the old-school pedal. Yeah.

Chris Gammell: I mean, and, like, there are benefits to doing some of these things, right? The benefits are, you know, you can have greater control, greater monitoring, stuff like that, you know, like, just having –

Dave Jones: 737 max.

Chris Gammell: Well, yeah. I mean, sure, sure. But –

Dave Jones: Right. Yeah.

Chris Gammell: But I just mean that, like – so thinking about – maybe not braking, but, like, you know, more instrumentation around a braking type thing in a car, right? You can have ABS type stuff and – Oh, yeah, of course. You can have other things in there instead of just having a pedal that ties to the, you know, pneumatics that clamp down on a disc brake or whatever. Yep.

Dave Jones: Yeah, but you can still have that just those automated ones. You can still have those. If you engineer it right, you can still have the automated ones do their business with the manual cable controls.

Chris Gammell: Yeah. Yeah. Yeah. I mean, yeah. We're standing way outside our usual area here. But it is interesting as more things – you know, like, everything has a touchscreen these days. Everything is controlled by wire, that kind of thing. And it's just – if there's not money to continually test reliability, and in this case, it sounds like it's more even of, like, a human factors UI type thing as well. If you don't have, like, each link just completely bulletproof, like, I don't know how it works as a system level, right? It's like you can't trust every link in the chain then, and then something's going to break.

Dave Jones: Well, I don't know what happened with this touchscreen system, but I'm surprised that they got this through. I'm surprised somebody approved this. Oh, yes. You mentioned this is your Navy experience. Because they're really old school. So, yeah. This is my Navy experience talking. And it's like, yeah, they love their manual controls. I worked – I kid you not – I worked for, like, a month on a Switch.

Chris Gammell: Wow.

Dave Jones: One Switch. Like, I'm not kidding, right? It had to be the right feel Switch. You know, sourcing this Switch, actually building it into a test platform so that then it could go to a panel of people so that they can play with the Switch and see that it has the right tactile feel, that the actual lead lit up bright enough under various conditions, like the backlit thing. Wow. It wasn't even a lead back then. It may not have been.

Dave Jones: It may not have been old still. Yeah. Anyway. Yeah. They, you know, they were very particular about, you know, the look and feel of that Switch. Right? And it was like, you know, it had to be the right way. They just didn't accept it otherwise, you know? It just had to be, you know, perfect. Otherwise, you know, they wouldn't have accepted some touch switch. It just, you know, it just wasn't going to happen. Interesting. But, like, is it – Yeah.

Chris Gammell: This is at a much bigger level, though, too. So, if, like, everything's changing. You know how that goes? Like, if, like, a committee of people are like, well, we're just changing this one thing. We need approval on this one thing. But if you're like, we're going to rip out the entire bridge and do it completely different. Yeah. That could be – that could be very, you know – that could be more drastic changes, I think.

Dave Jones: Hmm. Anyway, I am looking at a photo of this. Presumably, it's the console under question. And they do have manual controls below the screen. There's, like, these big lever things. I actually presume they're for, like, the throttle or something like that to, you know, change the blade speed or whatever, you know, blade angle, blade speed, and that sort of stuff.

Chris Gammell: Well, this is – it says it's a different – it's a different ship, though. USS Truxton. Yeah.

Dave Jones: Oh, okay. Right. Yeah. Okay.

Chris Gammell: I'm guessing they're not going to show the bridge of the John McCain. Of the one under question. Yeah, that's a pretty big – yeah.

Dave Jones: Anyway, they're changing back. They went, bugger this. Like, it's not the sole reason for the accident, but it contributed to it. They're, you know, so they're going well. Yep. Let's go back.

Chris Gammell: Yeah. Yep.

Dave Jones: I don't blame them. That's just – yeah. It's just silly. That's tough. I mean, yeah.

Chris Gammell: Working on, like – working on interface projects. I've got one coming up that I'm working on. It's, like – it's – Yeah. There's a lot of – I think it just enters another realm, too, right? It's, like, you know, humans react very specifically to their graphical interface. Obviously, everything has an – everybody has an iPhone or a smartphone these days, too. So there's, like, very – there's, like, new conventions that people are used to. And, you know, you have to kind of play along with those human factors and make sure you're doing the right thing.

Dave Jones: So can we go back to Rigol for a second? Sure. It's about user interface. Have you looked at the modern Rigols?

Chris Gammell: No, I haven't seen any of them.

Dave Jones: Oh, like, just call up any – just Google, you know, Rigol 5000 series or – I don't know. Sure. And, like, just look at the front panels of these things or – what's another – is it the – no, it's not the Siglet one. But anyway, some brands just look – like O1 or something like that. You know, if you just look at them, they just look so – like, I think it's the Rigol that famously had five different fonts on the front panel. I think we've talked about this before.

Chris Gammell: No, this is – the same friend who told me about this Rigol today. He said that – he watched a video where you complained about reverse italics for 10 minutes. Yes, right.

Dave Jones: Yeah, they had these reverse italics and the – yep.

Chris Gammell: And it's just a dog's breakfast. Just let it go, man. Just let it go.

Dave Jones: No, I can't let it go because we're talking about user interfaces, right, and how people expect certain things. And if you told me, like, explain what's wrong with the scope and why a Keysight one just looks much better and much more professional, I'd have a hard time explaining it. It's just lots of subtle cues. Yeah, but some of it is just your – is it the different fonts, the color, the –

Chris Gammell: It's your expectations, though, too. It's what you've known for a long time.

Speaker ?: Sure.

Dave Jones: And, well, there's nothing wrong. What's wrong with expectations? They've been built up over decades and decades of – you know.

Chris Gammell: Well, it doesn't mean it's wrong. It just means that you're not used to it. You know.

Dave Jones: True, but then when you go to analyze it, you go, aha, this is why my brain thinks this looks really amateurish because it does use reverse italics and five different fonts and all these weird colors that don't match and all these, you know. Like, it's just – yeah.

Chris Gammell: Dave enters the design world.

Dave Jones: No, we've actually – we've talked about this John Kenny from Keysight, right? He came on – he's one of the senior designers at Keysight and a great five-part series for anyone who – I think it's five parts on my channel. I think it's at least three parts or something. And I just sit down with him in the lab and just talk about design in Keysight products, right? Agilent back in the HP. He used to work HP back in the day. You know, he actually knew Bill and Ted, right? It's – Bill and Ted. Bill and Dave. Bill and Ted's excellent adventure. Oh, goodness. Bill and Dave. Apparently, they're making a new one. Yeah, I'm not a huge fan. Maybe they're making a new one. John Wick's going to be in it. Yep. Yep. Yep. Yep. I'm excited. I'm excited. Anyway, yeah, and he's talking about – you know, I'm sure he talked about, like, user interface design and how important it is and how they have teams that, you know, select the right colors and the right style and everything else. And you can tell, like, companies that, like, Rygol, they just give it to whoever and it's, like – and they just don't care. And it shows, right? It really does show. You put them side by side and you end up with the reverse italics and the five different fonts and all the colors and weird grouping and convoluted layout. Okay.

Chris Gammell: We got to get into something else here. We got to get – No, why?

Dave Jones: It's important.

Chris Gammell: If I hear you talking about reverse italic for 10 minutes, I don't know, man.

Dave Jones: You're the one who bought it up, you know, user interface. Yeah, yeah, yeah. Come on.

Chris Gammell: It's important. Reverse italics is –

Speaker ?: Why?

Chris Gammell: Why is it not important? I draw the line at reverse italics, Dave.

Dave Jones: Why?

Chris Gammell: So, I don't know. Like, I don't care about it. I mean, like, I don't care about, like, the look and feel, really. I mean, like, I don't look at test equipment like you do, I guess.

Dave Jones: Well, there's a lot of people who do. Yeah, right, right, right. There's a lot of people who do, right? Right. And it can matter. If you're designing a product, seriously, if you're designing a product and it doesn't look professional, people won't treat it as a professional product. And that reason might be because you've got some dummy like me to do the user interface. Right? It's like, yeah.

Chris Gammell: Yeah.

Dave Jones: It's going to look amateur.

Chris Gammell: Sure, sure, sure.

Dave Jones: So, yeah. Anyway, I know you don't care about the small details like this, but they're important. That's why I like discussing them. It matters.

Chris Gammell: Yeah.

Dave Jones: All right. Go on to your next topic. Fine. Fine. Go ahead.

Chris Gammell: I don't...

Dave Jones: Come on. Make a good... Better be good and interesting. I know. As interesting as reverse italics.

Chris Gammell: Let's see. How can I segue this? You should send the scope back if you don't like it.

Dave Jones: Well, you shouldn't buy it in the first place. You can obviously see it from the screenshots.

Chris Gammell: You should send it back if you don't like it, unless you can't, because it's non-returnable. That was my segue. Well done.

Dave Jones: Well done. Yes. Yes, golf.

Chris Gammell: The parts that I've talked about last week, or two weeks ago, rather, I was like, you know, I think I'm actually going to use different parts. And I went to go return them. And Dave and I were talking about it before the show, too. It's like, yes, parts are actually... You're cheaper.

Dave Jones: Cheapest ones you bought because they were, like, rock bottom price. You want to return them.

Chris Gammell: Right. Because I just... I figured out another solution, right? And I could talk about that later, too. Right. But, like... And then I, like, went to go hit the return button. And, yes, there is a return button on, like, a digi-key, right? And then it's like, no, no, no. Those are NCNR. It's like, oh... What's NCNR? Non-cancellable, non-returnable. As soon as I hit the buy button, oh, those are coming to me, no matter what. So...

Dave Jones: Isn't that default if you get them on tape?

Chris Gammell: No.

Dave Jones: Sorry. If you get them on cut tape? I don't think so, actually. Will they accept cut tape as a return?

Chris Gammell: Really?

Dave Jones: I thought you could only return, like, full reels and stuff. Or full tubes.

Chris Gammell: I can check this in real time, actually, because I have a digi-key order out right now. But I think you can return tape. Maybe not. Maybe that's not right. But definitely reels you can.

Dave Jones: And... Yeah, right.

Chris Gammell: Yeah. So, like, you think about if you're a CM and you, like, over-order the number of parts, I think there's probably a restocking fee. But... Oh, yeah, of course. You got to pay, like, handling and stuff like that. But you can't, like... You know, if you buy, like... You know, like, a reel of components very easily can hit in the thousands of dollars. And then you sit in the stock...

Speaker ?: Oh, easily.

Dave Jones: Tens of thousands, even. Yeah. Yeah.

Chris Gammell: And you don't want to sell it on the gray market. So, yeah, it's... It's... You can basically just return it. And that's kind of nice. Not everywhere. And obviously, if it's a marked NCNR, you cannot. But that's nice for some things. So... Yep.

Dave Jones: It's just never occurred to me to actually return anything.

Chris Gammell: Right. Well, you and I aren't ordering a lot of things that are like that. So... Okay. So I...

Dave Jones: Well, sometimes I... You know, sometimes I do pay thousands of dollars for one reel of components.

Chris Gammell: Yeah. You know? Sure, sure.

Dave Jones: And it's like... But I've actually used them. Like, if I did pay thousands of dollars, yes, I would be looking to get my money back.

Chris Gammell: Right, right.

Dave Jones: Either sell them on the gray market, as you said, or actually return them. Yeah.

Chris Gammell: Right, right. Yep. So I'm on the site right now. And I can return... Let's see. The 10 1.8 picofarad capacitors that I bought on my last order. Wow. I can return those for a credit of... Do it. Do it. ...of 30 cents. Do it just for kicks. I will not do it because it's a $7.50 return shipping. So I will not be doing that. But I could.

Dave Jones: So it costs more to ship them back than it does to keep them.

Speaker ?: Yeah, it would.

Dave Jones: Yeah, right. Yeah. And I think that's... You know, logistically, that makes a lot of sense.

Speaker ?: Of course.

Dave Jones: But... Yeah, it does.

Chris Gammell: Yeah. So it turns out you can. That's great. All right. I didn't know that. But we were also talking about... Because I've got a bigger build coming up with a... You know, like in a couple... Well, as soon as I get the parts. But we were talking about like, you know, when you put in orders at lead time and, you know, if I get a 16-week lead time on something, I need to get in line for those parts. And I'm not 100% on this. But I think you don't necessarily have to take delivery on all those parts. You at least need to get in line, right? So if I want parts... Right. I need to forecast for 16 weeks from now that like, I'm going to be doing a build. But in the event, I don't do the build. It's possible I could just cancel them. You're not forced to pay. Yeah, right, right, right. Right. And so I don't think that, you know, it would be a smart idea to like go and order all the parts because at some point they're going to be like, okay, well, give us some money before we're going to, you know, put you in line. But like from a risk mitigation kind of standpoint, like getting an order in, you know, I've pushed for that before where it's like, no, well, you know, especially if there's something that's super long lead time. It's like, I'm definitely going to need that. I'll at least put an order in and then see, you know, see what happens. It's always worked out that I've used the parts, but I think you don't have to take delivery for all of them. So TBD on that one. So.

Dave Jones: Okay. I don't think I've ever done that. Maybe our, maybe the company purchasing people have or something like that, but I've, I don't recall ever having personally done that.

Chris Gammell: Yeah.

Dave Jones: I think I paid up front, you know, like I've, I've ordered stuff that's like 30 weeks lead time. Yeah.

Chris Gammell: Right. You know, it hurts. Back in the day. Yeah. It still happens. It still happens. Yeah.

Dave Jones: Yeah.

Chris Gammell: Yeah. It's, it's tough.

Dave Jones: But we're so, so used to multiple source parts these days. It's like where we're really spoiled for choice.

Chris Gammell: Uh, like. It's nuts. What do you mean?

Dave Jones: I mean, you, when, when you actually design something, there's, there's usually some sort of alternative. You know, you, if you're smart, you'll try and design your, uh, product around, uh, something that has multiple sources. Or if you can't, at least, you know, it's in stock from multiple suppliers. So, um, unless, you know, Murphy steps in and, and all of a sudden, when, when you go to buy it, oops, all the stock in, you know, five different locations is all sucked up at once. Right.

Chris Gammell: Right.

Dave Jones: Right. Then generally you're, you're able to get it. You know, there's not too many people these days that are designing a product and then waiting, you know, 20 weeks for the parts. Uh. Like not in the small side of things. In the larger end of town. Yeah. Yeah.

Chris Gammell: I guess, yeah, I guess, I guess that's right. Yeah. Yeah. Usually people aren't waiting 30 weeks for a sample or, uh, you know, the five, that's right. A five off cut from the reel. Yeah. I think you're, you're right about that. But from then a reel.

Dave Jones: With the, with the lean manufacturing, these, you know, all the, all the, you know, the, everyone wants some, everyone's used to getting something now. They, they, they get their boards turned in less than a week. They get, you know, everything. They order their parts from DigiCam, Mouser, they do it. You know, you can design and build something in a week. Yeah. Right. Yep. Like waiting, you know, 20, 30 weeks. It's just, you know, nobody does that these days.

Chris Gammell: Right. Right. Yeah. Well, the thing that I always wonder about is people that like, you know, there's like some really funky variants of like certain MCUs where it's like, oh, well. Oh yeah. Yeah. Yeah. We want, uh, you know, we have to have the floating point unit, but we need the 64, you know. Yep. And that's not in stock anywhere. It's like, but I need, I know I need that one. Well, they don't even do it unless it's bespoke. You know, like, it's just like, they have to do a, a fab spin for a, you have to do a wafer start for it. But it's like, those are the ones where, you know, like people then are, you know, you add an external memory on a dev board or whatever it is, but, um, it's, that's, that's disruptive to a development process.

Dave Jones: As I said, it's, yeah. But as I said, that's only going to be in the reasonably big end of town. Yeah. I think you're right. You know, your average midnight engineer or small company is just not going to do that. Right. We're going to find a different way to do it. Yeah. I think you're right. Yeah.

Chris Gammell: Yeah. Yeah. If you're, if you're digging into the spec sheet or the, you know, the ordering spec part of the data sheet and you're like, no, no, I have to have this part. It's like, yeah, you're going to, you're going to hurt for that one if it's not actually a stock. Yeah.

Dave Jones: No, you're all, there's always, almost always alternatives to this. Yeah. If I was going for a micro and I, oh, look, they, they do make this part, but nobody in the world has stock. And it's like, yeah, they, they're going to have to do a fab run for it. There's no way I'm going to design that into my product. It's like, I'm going to, I would rather take the pain of switching to a different brand micro than actually waiting for that part.

Chris Gammell: Yeah. Yep. Yep.

Dave Jones: So yeah.

Chris Gammell: Speaking of micros, there was a article about the, it was inspired by your video on the three cent microcontroller. Yes, it was.

Dave Jones: It's actually Tim who's on the forum.

Chris Gammell: Oh, great. Okay. And so this was talking about other, other, other low cost micros and stuff like that.

Dave Jones: But he basically did a shootout of sub 10 cent micros and came up with this nice little chart. I hadn't heard of most of these, most of these manufacturers. Yeah. Right. It's, it's, it's, it's a really great article. And there's a, there's a PDF version. Let me download the, the terrible three cent microcontroller comparison chart. Let's have a look at some of the company names. See if you recognize them. Hands up. Let's play bingo. Pull up.

Chris Gammell: Pull up. Pull up. Pull up. Pull up. It's like pull up.

Dave Jones: Duke. We've heard of Paduk because it's a three cent micro I've done a video on. Bojang Jing Industrial.

Chris Gammell: This is going to be, this is going to be a terrible mangling from Dave. Yeah. And me too. To be honest. It's a terrible mangling. Yeah.

Dave Jones: Oh, they've got, they are Y, Y Spring Tech.

Chris Gammell: Uh huh.

Dave Jones: Yeah. Okay. East Soft Micro. These are all China or Taiwan, by the way. And Holtec. You might've heard of Holtec. So. Yeah. Holtec's very common in, they make chips, chips sets for multimeters and they make microcontrollers. We're using a Holtec part in the micro supply, actually. Oh yeah. Not a micro, but they make an LCD controller.

Chris Gammell: Oh, okay.

Dave Jones: So we're actually using that. It was one of the few LCD controllers. That's one of the things where like we, you know, finding an alternative would have been hard. This Holtec one actually did the business and did it for a decent price. And there was nothing else. Yeah.

Chris Gammell: Right.

Dave Jones: There was nothing else out there that could match it for the price. Not even close. I think. And it was like, yeah, we had to use that Holtec part. Yeah. And anyway. Yeah. There's all these different manufacturers. Great table. And, but the Paduk one, they've now reverse engineered. Yeah. We mentioned this the other week. They've reverse engineered it. They've got a, now an open source compiler for it. There's an open source programmer. Yep.

Chris Gammell: It was the SDCC was the, was the, it was the, the compiler. So yeah. SDCC. For those of you playing at home.

Dave Jones: Which now supports. Yep. And yeah. Anyway, it's a great article comparing these and yeah. And all the different pros and cons of them.

Chris Gammell: So it's great. Yeah. These 10-10 micros. Were you the one telling me about the company that makes STMicro compatible parts? Was that you or someone else?

Dave Jones: Don't think that was me.

Chris Gammell: I think that is, no, which one is it? They make ST compatible. It's like, it's, it's, it's register compatible.

Dave Jones: Yeah. But ST, it's registered like as in an ST arm. Are we talking about like an ST32 or something?

Chris Gammell: Yeah.

Dave Jones: Really? They make a full register compatible version?

Chris Gammell: Yep. Yep. Actually, man, I think.

Dave Jones: Because if, if, if people don't know, right, you can't just buy an ARM chip from one manufacturer and then move to the ARM chip of, like you can't just buy an ST ARM micro and then move to a, an Atmel ARM micro. Right. Right. You just can't do it. You've got to remap all the peripherals. They're physically different.

Dave Jones: Implement. Yeah. Yeah. All, all the peripherals, all the registers, they're all different. Right. So, yep. Yes. Your, your code, your base code will still work. Right. You can, you can just import that, but all your register settings and all your peripheral setups, your clocks, your timers, your UARTs, your ADCs and all that, you know, all these external peripherals, they're, they're all custom for each individual manufacturer, which is a shame, but that's the way it is. So you're saying there's someone fully compatible.

Chris Gammell: Yeah. I was in the wrong, let me see if I can find, I was in the supervisors thing.

Dave Jones: You're in the, uh, the deepest, darkest alley of, uh, Shenzhen, were you? No, no, no. This is just. I came out and go, psst. L-C-S-C. Yeah.

Chris Gammell: Well, I was, I think the reason this came up is so after looking at the parts that I ended up buying and not buying, well, not wanting them and then realizing I'm keeping them. Um, I ended up going on, I was on L-C-S-C and I was just looking at parts there and there was a 35 cent STM32F0, 3-0 part on there. And then like 15 cents for like a 433 megahertz receiver and a 15 cents for a trans, for a transmitter. And so it's like. Wow. Yeah. Yeah. That's real cheap. Wow. So that's what, that's my current, uh, playing. Um. All right. And we don't.

Dave Jones: And, and it's a compatible part. It's not a genuine.

Chris Gammell: These are.

Dave Jones: Because they do sell genuine ones as well.

Chris Gammell: That's right. No, this is genuine. This is a genuine.

Dave Jones: Oh, okay. Right. Oh, wow.

Chris Gammell: Yeah. Um.

Dave Jones: And this is coming from a catalog supplier because they are essentially a catalog supplier. Right. Yes. Which, which is generally not the cheapest way to get a part yet. They're still so dirt cheap. Yeah.

Chris Gammell: So it's one fifth the price of the equivalent part at DigiKey is, is a buck 50 for, for the same quantity. So like.

Dave Jones: You did hear about L-C-S-C, didn't you? I don't know if we mentioned it on the show.

Chris Gammell: I don't know.

Dave Jones: Um. Somebody, uh, busted them. Somebody, they, they did a video. A video tour of, or a photo tour of their production facility.

Chris Gammell: Oh, okay. Um.

Dave Jones: As in, like, as, as in their pick and place, you know, where they, uh, pick, pick and place. As in, they're picking, you know, the stuff and the shelves and all that sort of stuff. Okay. And they'll use in one of those wireless anti-static wrist straps. Yeah. So, so I actually contacted them and I said, Hey, you might want to watch my video about that. And they came back and said, Oh, um, yeah. Oops. Sorry. We'll fix that. Yeah. So, yeah. ESD was precautions and LCSC were kind of out the window. So, but they've fixed that now, I think. That's good. Yeah. They, they said they would. Yeah. So.

Chris Gammell: So I figured out the, uh, the name of the, it's giga device. That's who does the, uh, that's the. Gigadevice. Yeah. So if you look at the part numbers, that's where it really becomes obvious. Right. So if you look at, so SDM 32 normally would be, so here's an example part. SDM 32 F1 0 3 C8 T6. Right. So that's the C8 package. That's the name of the company. No, no. That's the manufacturer part number. GD. Yeah.

Dave Jones: But no, no, no. Sorry.

Chris Gammell: Yeah. What? What? Our wires are crossed here.

Dave Jones: Gigadevice Semicon is the name of the company in Beijing.

Chris Gammell: That's right. Yeah.

Dave Jones: Gigadevice. Wow. And they make.

Chris Gammell: So then, so SDM 32 F0, F1 0 3 C8 2 6. The gigadevice part is GD 3 2 F1 0 3. GD in front instead of SDM. And so like they're, after SDM, they basically completely replicated it. I don't understand in the modern world. Like obviously this happens with 74 series parts, right? It's just like, that's all out there. Yeah, of course. But this is. Yeah. Something different. And, uh, and these are dirt cheap.

Dave Jones: And there's a lot of effort that goes into that too.

Chris Gammell: Yeah. Yeah.

Dave Jones: Like to, to actually copy a chip at the register level. Yeah. You know?

Chris Gammell: Yeah. So. Wow. Interesting. I don't know. Like, uh, like the cost, like the SDM 32 parts are pretty cheap on there too. So.

Dave Jones: Are companies like this actually paying the, um, uh, fee?

Chris Gammell: Yeah. I don't know.

Dave Jones: Or are they just stealing that as well? Yeah. Like, are they just like.

Chris Gammell: Here's the real question. Wow. Would you put this in a product? Cause I don't think I would, I'd rather just pay right now.

Dave Jones: I, I, I wouldn't, I would definitely pay the genuine. Yeah.

Chris Gammell: Yep.

Dave Jones: If I was, if I was making a million greeting cards or something and it didn't matter, of course.

Chris Gammell: You're using a SDM 32 for that, of course, because why wouldn't you? Yeah.

Dave Jones: Well, yeah. Yeah. Right. I should be using the three cent micro, but you know, yeah, no, I wouldn't put it into any serious product if it was, if it was non-serious and I just needed, you know, if I was just doing something for, you know, uh, something for fun, like I was, I was designing the world's largest lead array or something like that. And I just needed, you know, a thousand chips to actually do this. Then yeah, I would, you know, and I had to do something simple and easy. Yeah. I would use something like this, but otherwise no. Right.

Chris Gammell: Um, so this is actually not new apparently. So I just found a article from, uh, dangerous prototypes from back in January, 2018. So this is obviously not new. And then it's linked over to a comparison. So we'll have that link in our show notes too, just for the, um, yeah.

Dave Jones: Okay. So they've got to do, how deep do they go into the comparison?

Chris Gammell: Uh, solderability, blinking LED, UART and Spymaster. So comparing all that stuff. Oh, okay. Right.

Dave Jones: So they're just some basic. It's still something though. Like actually putting it on a board and doing it. Yeah. No, you've, you've, you've, you've got to get the oscillator configurations working. You've got to get the IO configurations working.

Chris Gammell: You know, we may have even brought this up before too. Who God knows. Oh really? Yeah. Well, cause Zepto bars did a shoot of it too. So, um, ah, right.

Dave Jones: They actually, and usually when Zepto bars do something, we mentioned it. So, um, and is it a physical, is it a physical copy or is it just a register actual, actual reverse engineering copy? I cannot tell. Cause surely they don't get access to the masks. Yeah. To actually copy the masks. Right. So I'm sure they're different.

Chris Gammell: Yeah. Well.

Dave Jones: Which is, it's just remarkable. But the Chinese market is so big. Right.

Chris Gammell: Especially in like, in like, in the SM32 is like in everything, you know, like they really.

Dave Jones: It's in everything. It's in everything. It's the de facto standard. Yeah, exactly. Yeah. Yeah. It's, it's sort of, yeah, it's like, like that arm market, like five, five years ago, like seven, like the arm market was so flooded. It was like, I, like you didn't know who to choose these days. It's like, eh, you're probably going to use the STM32. Right. You know, that, that's kind of like, if you're coming in and going, I need to use an arm chip, that's probably the one you're going to choose just because of like everyone else is using it kind of thing.

Chris Gammell: Right. Yeah. Yeah.

Dave Jones: It's like, yeah.

Chris Gammell: I mean, there's a lot of, yeah, there's a lot of stuff out there. Supply is obviously pretty high. So that helps. Obviously you have two vendors now. So yeah.

Dave Jones: Wow. Well, that could work at a pinch, right? If you couldn't, if you had, you know, if, if you couldn't get the part, if you don't know, somebody came along and bought up all the STM32s and you were going, shit, I need to ship my Kickstarter. Yep. It's like, well, you, you would have a go because the problem with this is that you can test it. Right. We've talked about this on the amp hour before. You can thoroughly test something like it is not a risk to use this unless, well, no, even then, no, there is almost, there's practically zero risk in doing this. Yeah. Dave versus spy versus spy. If you, if you, if you thoroughly test this, like you get this sucker in the thermal chamber, you test it over its thermal performance limits and voltage rail performance limits and, and clock oscillator limits and every other, you know, variation limit you can think of, you can be 100% sure that this chip's going to perform in your product.

Chris Gammell: Mm-hmm.

Dave Jones: So there's, so there's no problem there. So, so if I actually, so I wouldn't, I wouldn't necessarily, um, be, be against using this in my product, but I wouldn't design it in from the start. Right. I'd be designing in the ST one.

Speaker ?: Yeah.

Dave Jones: Yeah. And then only in a pinch would I think about using this one. But you certainly can.

Chris Gammell: Yeah.

Dave Jones: It's just a matter of testing, says the guy who comes from the test engineering background. Right. It's like, you know, this was my, you know, this is what I did every day. Right. It's just, you know, test stuff over thermal and, and other limits by even vibration and shock and all sorts of stuff. Right. To make sure it all, uh, worked. Yeah. I don't know.

Chris Gammell: I don't know if I would, I don't know if I would design something like this into my circuit, you know, and by that, I mean, I don't know if I would, I wouldn't design it. I don't know if I'd put a micro control in my circuit day. That's what I was really saying. You just really can't trust. You wouldn't put a micro in at all. You can't control. You can't trust micro controllers these days. Nah, nah. Dodgy as. Yeah. Uh, we should, we should, we should use that as our, uh, as our excuse to leave. Oh, we're 10 minutes past. Yeah. Yeah. Uh, I will be recording from the road next week with the Mighty Om and maybe Alvaro and a bunch of weirdos from Chaos Camp. So looking forward to that.

Dave Jones: You'll be in Germany.

Chris Gammell: Yeah. Yeah.

Dave Jones: And you're going to Iceland or something as well, are you? Or somewhere?

Chris Gammell: Yep. Yep. Stop it over there too. So that's just because you get. Why are you going to Iceland? Because the cheap airlines fly through there. So. Oh, right. Okay.

Dave Jones: Are you actually going to see some sites there? Yeah. Yeah.

Chris Gammell: Yeah. So I'm, this is like a 24 hour layover.

Dave Jones: Oh, is that all?

Chris Gammell: Yeah.

Dave Jones: Jeez. You can't see much in 24 hours.

Chris Gammell: Dude.

Dave Jones: Come on. It's Iceland.

Chris Gammell: Not a big country, Dave.

Dave Jones: Yeah, but come on. If there's any Iceland listeners. I've always wanted to go there.

Chris Gammell: Jeff and I will be there. Let us know.

Dave Jones: All right.

Chris Gammell: Catch you next time.

Dave Jones: Catch you next time.

Chris Gammell: No, I'll catch you next time. Yeah. See ya. Why do you always get to see Catch you next time? x x