#646 – Fan Fanboys

Chris Gammell: This is The Amp Hour Podcast. Released September 11th, 2023. Episode 646. Fan Fan Boys.

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: Should we change our intro? We've been doing this crap for like 13 years now. Oh, I'm not too fussed. No, no, not too fussed. Okay.

Chris Gammell: I mean, I could go first sometimes, but you push back on that.

Dave Jones: Absolutely. No, no. Age comes first. Age before beauty. Yeah, that's it. That's fine. Don't worry about it. Don't worry about it. Okay, right. I don't.

Chris Gammell: You know, I am older than I am when we started this, Dave.

Dave Jones: Significantly, but so am I. It's gone up in a linear proportion.

Chris Gammell: Yeah, that's how it goes.

Dave Jones: Although it doesn't feel like that. It feels, you know. No, no, of course. Logarithmic, but, you know. Right. Anyway.

Chris Gammell: Yeah. I am no longer a seller on Tindy.

Dave Jones: Oh.

Chris Gammell: I didn't know you were a seller on Tindy. I know, I know, I know. Selling what? I was selling this, like, tiny little 555 soldering kit. It was basically just for DEF CON, like, four or five years ago. And then I got surprised. I had been giving these things away at conferences, you know, just like, I was like, oh, I'm not going to sell any more of these. Right. I was like, I was like, I was like, I was like, I was like, I was like, I was like, oh my God, it's still listed. It's still there. It's still there. Yeah.

Dave Jones: I've, I, I have had that. Yeah. I've had that. Where have I, I've sold something obscure in the past and it's still on my website and people send it, you know? And it was like a service that like, I haven't thought, like it was, uh, Clickbank. It was, I think it's.

Chris Gammell: Like a payment service or something?

Dave Jones: It's like a payment service that was, you know, before credit cards, like online credit cards were like really, you know, a thing you have to use these obscure services. Yeah. Yeah. And I, I'm like, what Clickbank? You know, I get this notice from freaking, you know, like somebody's bought something that I had on the website from like, God, nobody's bought that in 10 years. I didn't even know that was still a thing. And yeah. Yeah. So you just annoyingly, um, so you just went, oh, holy crap.

Chris Gammell: I shipped the last two I had. You got a, you got a twofer. Oh, so you didn't ship them. Okay. Oh, a twofer. Officially no longer a seller. I keep thinking like, ah, you know, it's not that hard to like manufacture. Not, it's not that hard. It is not like, it seems like it's like something that could be worth doing, but then I just, I'm, I'm actually in awe of the people that are on like Tindy or equivalent. Oh yeah.

Dave Jones: Selling little things for five, 10 bucks and stuff like that. Oh yeah. It's amazing. You know? Yeah. I will refuse to sell anything on eBay if it's under like 50 bucks. It's just not worth my time.

Chris Gammell: Right.

Dave Jones: Like, like, honestly, like I get out the packing labels, right?

Chris Gammell: It's just like crazy.

Dave Jones: Like, I don't want to sound, you know, elitist or anything, but you know, like I bet that's something. That I did, right. That's something I used to do, but now my time is just, you know, the older I get, speaking of age, right? The older I get, the more valuable my time.

Chris Gammell: It's a more precious commodity.

Dave Jones: Yeah. Yeah. Yeah, exactly. Back, back when I was 20, I didn't, yeah, I was mailing out floppy disks, you know, in the mail.

Chris Gammell: Right.

Dave Jones: Actual floppy ones, five and a half inch, five and a quarter inch.

Chris Gammell: Right. That's great.

Dave Jones: You know, for 10 bucks a pop or whatever.

Chris Gammell: Do you guys have media mail over there? Is that a thing? Like, could you do that on media mail?

Dave Jones: Well, you can get a, yeah, I bought these cardboard mailers and you could get five and a quarter mailers. And, and the, uh, the, the three and a half inches are, we're just fine in there too. So I use those for, yeah. So you'd ship these hard cardboard.

Chris Gammell: It's a thing. It's a UPS, USPS thing. So the, the service here, it's like a lower rate. If you're just like, if you want to like ship a book, like super slow, but then I think like records and CDs as well.

Dave Jones: Yes. Right. Yeah. Amazon took advantage of that back in the old days. Didn't they? Like when they were first starting, I think. I don't remember. Amazon.

Chris Gammell: My main experience with it was like selling, buying and selling textbooks. I just, just for usage, not for like profit. Yeah. Yeah. Definitely not for profit. But yeah, there's, but then they, I noticed that they did have it for other, other things out there. And now I don't know who would ship.

Dave Jones: We didn't have specific rules for that, except that it was, it was, it got the same rate as a, so it's based on size, thickness and weight. So if it met that, then it could go via just a local stamp, you know, like a letter. Yeah. Yeah. Stamp. Yeah. Yeah. So it only costs, you know, 30 cents back in the day or something to post it. Now stamps a dollar or something. I don't know.

Chris Gammell: Wow.

Dave Jones: Yeah. Yeah. Yeah. So yeah, that was a thing. And yeah, I would ship by, you know, somebody sent me a money order, you know, so I had to go to the post office, line up at the post office. Just to cash my $15 money order, please give me my $15 in cash, please. You know? Money order. And then there was a $2 processing fee or something, you know?

Chris Gammell: Sure. Sure. Sure.

Dave Jones: Yeah.

Chris Gammell: The things we do in the inefficient younger days.

Dave Jones: I was, I was young and I was making money and it didn't, you know, it was fun and you know, it's like, yeah. But no, it's not something that I would do these days. I wouldn't, I wouldn't sell a $10 thing.

Chris Gammell: It did bring you to where you currently are, Dave. So you really can't get that.

Dave Jones: Oh yeah. No, no. Totally. Right. Yeah.

Chris Gammell: The snob for a maven that you are, you know, brought you here.

Dave Jones: Yeah. But no, I'm not going to sell something for 20 bucks on eBay. It's like, I would rather have it for free. I'd rather, you know, just wouldn't even bother with the shipping. Right. I have someone send you a, somebody turns up, picks it up. Self-addressed stamped envelope. Yeah. Right.

Dave Jones: Yeah.

Chris Gammell: Yeah. Classic.

Dave Jones: Yeah. Yeah. I used to buy things with stamps.

Chris Gammell: You'd buy them with stamps? Like in prison?

Dave Jones: You would actually send stamps as payment. Colin Mitchell talking electronics, who I've done a five-part interview series with. Sure. He was in Melbourne and he would accept, like he sold kits, right? Sold hundreds of thousands of kits. And he would accept stamps as payment because he was like shipping stuff every day. So, so you didn't have to pay him. All you had to do was go to the post office and you'd send him like a whole bunch of stamps. And that, and that was the payment for your item because it didn't matter. Cause it was all, yeah. It was cash or stamps. It's all the same thing. It's fungible basically. Yeah. Yeah. Yeah. Yeah. It's all fungible. Yeah, exactly.

Chris Gammell: Yeah. So yeah. That's, that's interesting. Yeah. Yeah. That's great. I mean, that is like literally like in like prison movies, you see people like.

Dave Jones: Right. Yeah. They have traded. Interacting with. No, they trade. Well stamps, stamps too.

Chris Gammell: Like books of stamps because you can't get stamps. Yeah.

Chris Gammell: And they are, they have actual value like internally and externally.

Dave Jones: Yeah, exactly. So. Right. Yep. Ah. So anyway. Yeah. Yeah. So you're, you're in the same thing. You wouldn't bother selling something for 10 bucks on Tinder. It's just, it's same thing, right? It's just not worth your time.

Chris Gammell: I mean, one thing I feel kind of bad, I mean, like it, it had some design, you know, this also kind of get to the, to the core of, of my like being. Right. So like, you know, it was this cute little. It's like literally five, five, five circuit. And I sold it for 20 bucks because of the trouble that it was. And people, some people paid for it. It was like a little add on thing. It was a little knickknack, whatever. There's a kit. Yeah.

Chris Gammell: I felt kind of bad about it. Like, obviously the material cost was not, you know, it's like, it's always that, that, that hardware brain thing, you know, like you're, you're like looking at you like, there's no more than like $3 in parts here. If you, you know, you amortize all the shipping costs, the PCBs and whatever. And like, maybe, maybe then you even, maybe you get up to five, six bucks with my time. But the real thing is you have to be like, shut up hardware brain, engage business brain. It's like, is someone willing to pay $20 for a knickknack? And yes, they are. Right. I mean, like, it's just, it's just one of those struggles that like those two parts of my brain really don't like each other. And you really have to shut down the hardware part of your brain if you want to stay in business. Right. I mean, like, not that I'm in business. It's just that it just reminded me of that tension of like, that would cost. It's what it's worth. That's, you know, for all the people that are considering doing this. Exactly.

Dave Jones: Yeah. Well, I've got like a fight, like, you know, and now I sell meters, right? I sell multi meters. And, you know, if I get the occasional one that's return failed, right? Do I bother even claiming that under warranty from Bryman or from Kane? Test? No, I don't bother. I just eat the cost. It's just not worth, you know, like, but now I'm actually getting quite a few. I don't know what it is. There's something in the water. There's at the, at the fab plants and I'm getting quite a significant number of meter failures these days. So they're actually building up. I'm staring at a pile right now. So I'm thinking it's getting to the point where I might actually have to actually claim them under warranty. It's, you know, the first time in eight years or whatever. Yeah. Yeah. Yeah. The actual factory warranty. Yeah. Yeah. Yeah. Because I just sell, buy and sell so many of them. It's just not worth claiming one, you know, as like a warranty in, you know, from the factory. It's not. Yeah.

Chris Gammell: So. So the thing that's, I think, different there though, like, so like you are selling a thing that has, well, actually, I don't even know, like your meters have comparable, they at least have like comparable things that are out there. Right. So you're selling a value add either because of your endorsement or, you know, you have.

Dave Jones: Because it's blue instead of red. Right. And it has my label on it. You can buy the exact, well, you can, some of the meters, you can buy the exact same meter somewhere else. Right. With the red Breiman logo. Yeah.

Chris Gammell: Right. And so that point, like you have, you have a supplier, obviously you're just, you know, you're charging the, almost like the retail difference there. And then like, but the price is basically set by the market, right? It's like, you know, the most you can get for the market. But I think it's, I think there's a little bit of a difference when it's like, you're selling something new for the first time. I feel like I'm always inclined to sell something lower than what I probably should. Right.

Dave Jones: Especially when it's like a market. Of course.

Chris Gammell: So like probably more like your, what's your current thing? The microcurrent thing, right? At the time it was not completely unique in the market, but it was at least standalone enough that you had some like market power there, you know, and you could set a higher price than the cost.

Dave Jones: I could have. Yeah. But I didn't. And, and I sold a kit version, which was, you know, required a lot of time and effort, but I figured some P you know, some people want a kit and it's good because they can save some money for people who don't have much money. They can build it cheaper, you know, et cetera. So I, so I went to that effort to actually, you know, actually, you know, stock and, and kit these up and sell them and support them. And you know, yeah. Right. So.

Chris Gammell: Right. But that was a choice because of, I'm just saying like, uh, when you are like entering a market as like a product maker, it's like a whole different thing than like as a, as a designer, I feel like.

Dave Jones: Oh yeah. No, totally. Yeah. But, but, but I suck as a businessman, you know, like I absolutely suck. Right.

Chris Gammell: I was selling $20 circuit boards on Tinder. I mean, I'm not the paragon of business here. Yeah.

Dave Jones: Yeah. Yeah. That's right.

Chris Gammell: Like, you know, but yeah, I think, I think just for people listening, you know, like learn from our mistakes. It's like the price is not like, there are no immutable rules. It's, you know, like we always talk about like 4X, the hardware cost is like, is like a guideline. But really the thing you should be charging is like, you want to like basically figure out what is the maximum you can, you know, if you look at like economics, you want to like maximize the price versus how many you can sell. And you should like modulate the price until you hit that, you know, you figure out that curve that, and then, and then you hit the maximum point in that curve. That's yeah. That's kind of.

Dave Jones: And that can change based on market dynamics. Like if it's a recession at the moment and everyone's, you know, feeling the pitch, your sales drop, you know, minor like going through the roof at the moment for some reason. I don't know why. I just, I just shipped 11 meters this morning overseas. They're sitting out the front waiting to be picked up.

Chris Gammell: Maybe exchange rate?

Dave Jones: Is that, that's just in one day.

Chris Gammell: Sorry. Y'all, y'alls have some cheap, cheap exchange rates right now.

Dave Jones: I don't, well, yeah, that's another bloody problem. The bargain based in Aussie buck. Oh, don't get me started on bloody Shopify. I'm trying to set up a Shopify store at the moment and, oh, the, it's just, no, no, just don't, don't, it's traumatic. Anyway. Yes. The exchange rate is a, like a real problem and trying to, yeah. And shipping rates. Sure. Oh God. Right. All that stuff. I'm going through that ordeal at the moment. I just did a rant video about that. Speaking about production bomb costs. I, I just did a, uh, tear down video of the new Rigol DHO. I watched it. You watched it. You did. I did. Yeah. Excellent. Right. And you know, that's.

Chris Gammell: That fan heat sink was super interesting. Like that.

Dave Jones: I am doing, I've got a live test running right now. I've actually disconnected the fan and it's currently at 70 degrees. Holy crap. So if the lab, if the fire alarm goes off.

Chris Gammell: No way.

Dave Jones: You know that my DHO 800 has caught a light and, um, I'm going to have to evacuate the building and I'll get charged 1800 bucks by the fire brigade. So yeah, I'm stress testing it with no fan at the moment.

Chris Gammell: What is it? So like, what is the heat sinking on?

Dave Jones: I saw that the, on the, it's on the, uh, it's on the FPGA. It's on the zinc FPGA. It's on the analog to digital, the 12 bit analog to digital converter, which is a custom ASIC jobby or custom Rigol jobby and the, uh, custom analog front end chips.

Chris Gammell: Okay. So all those things, but not the, the rock chip. I saw like you said it's running Android.

Dave Jones: That doesn't have a, uh. I hope to actually, is it the rock chip as well? It might be, it might be the rock chip processor as well. I'm not. I can't remember.

Chris Gammell: So they're just trying to draw all the heat away basically.

Dave Jones: Yep. Yeah. Yeah. And then when, and you measured the power as well, right? Basically all of the major parts on that board, they decided for those who can't, who haven't seen it, it's basically one big heat sink, which then has little machine protrusions, which go down on the top of each chip. So it's one giant heat sink that covers the whole board. Basically.

Chris Gammell: That's not machine. Is it? It looks like it's poor. It looks like.

Dave Jones: Well, it's machine. Well, I don't know. It's a cast whatever.

Chris Gammell: Magnesium. Yeah. It looks like cast magnesium or something like that. Right. Aluminum maybe. I don't know. It's, it's a cool shape though. It's an interesting, interesting concept.

Dave Jones: Yeah. It's nice. And yeah.

Chris Gammell: Are you going to put the heat sink back on and try and try and get that, that temperature when the heat sinks back on?

Dave Jones: Well, no, no, sorry. The heat sinks on at the moment. Just the fan is not on. Right.

Chris Gammell: Oh, wow. Okay.

Dave Jones: So, so the heat sinks, oh no, I wouldn't run it without the, I'm sure the dyes are, you know, so there's no fan. Right. So I'm currently running it with no fan. So it's, it's getting the heat sinks at 70. So I wonder what the dye temps are at. Woo. Anyway. So I am just testing it during this episode and I'll shoot a video right after we finish here.

Chris Gammell: Yeah.

Dave Jones: I already shot the before with the fan and it was a 40 degrees and now it's at 70. So oops.

Chris Gammell: So the fans doing some work, huh? Yeah.

Dave Jones: Fans doing the tiny little fan and everyone's talking about in the forum. Oh, people are obsessed with fans. This is why, uh, Roden Schwartz in their new scope, they put in the super duper silent, mega awesome fan in it. Like, it's just like, you, you, you don't even know it's got a fan. It's just, oh my God. It's like top of the range, you know, super silent, the design with rubber compliant mounts and everything. Get that sound out.

Chris Gammell: So I guess the scope is the kind of thing you're keeping on most of the time.

Dave Jones: You're keeping on for the whole day. Yeah, exactly. Yeah. You know? So anyway.

Chris Gammell: Yeah.

Dave Jones: Yeah.

Chris Gammell: Hmm. Are you going to try and get some like a CF, what is it? CFD? Computational fluid dynamics person. You're going to try and talk to someone like that.

Dave Jones: That'd be interesting. Yeah. Yeah.

Chris Gammell: Yeah. It would. But, um, yeah, just cause like, I always wonder about that. Like I remember, I guess the only time I really dealt with that was at Keithley. When I worked there, they had, they had like, they did some like CFD, you know, simulations. I've seen some, obviously some solid works and other, they all, all of the CAD programs want to talk about it, but I've never actually had any designs that have needed it because I, I usually don't like operate that close to the edge. Cause I'm not. Exactly.

Chris Gammell: I'm making $20 555 timer boards, Dave.

Dave Jones: Yeah. Yeah, exactly.

Chris Gammell: No heat sinks on those.

Dave Jones: You can do some basic calculations, what you need and you over engineer it a little bit and you put some air flow on it and Bob's your uncle. Right. And, and, and then you do some real world testing on it and you take temperature measurements. Yeah. We're well in the ballpark. We've got lots of margin. Right.

Chris Gammell: Please don't go above 40 C in your ambient and you'll be fine.

Dave Jones: Right. Oh boy. I, I, I just did some, oh boy, the fan boy, the heatsink fan boys. The fan fan boys did a video cause my, my computer failed right on my, my computer failed. My, the machine I'm using at the moment to render all my videos. It finally failed after a couple of years. It was shutting down when I was rendering my videos and it turns out that my heatsink compounded has failed.

Chris Gammell: Right.

Dave Jones: Yeah. The government shutting me down. No, no, no, no.

Chris Gammell: Actually your computer doesn't like it. It's just like, oh, Dave, come on again. Oh no. Oh, right. Again with a rant video. Yeah, exactly. Too many, too many GPUs and there you go.

Dave Jones: Right. It's LLM and it's like, yep.

Chris Gammell: There you go.

Dave Jones: So yeah. So it turns out, yeah, I had a combination of bad paste. It had just dried out over the years, you know? Yep. And so I, I repaste it and that fixed that. So it got better, but it still happened once. So I looked at it again and it was the way my fan was mounted. It, I, I, I got a 10 degree C differential in, in temperature just by relocating my fan to the other side of the heatsink.

Chris Gammell: Wow.

Chris Gammell: Okay.

Dave Jones: Because it's all about the inlet temperature. So it's currently mounted on the top of the case, which pulls cold air. I've got vents on the top of the case. So it pulls the ambient office air here in, instead of blasting it out and from the, and, and, and the GPU cards nearby. So it was, it was basically the inlet temperature of the fan. And a lot of people didn't believe me. I shot this video and that was my, you know, theory. And a lot of people didn't believe me. So I had to actually do another followup video proving that that was the case. And I got a 20, 10 degree C differential and everyone's going, well, okay, that's new. I haven't seen that before. Hold my beer.

Chris Gammell: Somebody's, somebody's wrong in the internet.

Dave Jones: Well, it's apparently cause that is the, the prevailing wisdom in the computer, you know, business, the Linus, you know, the Linus tech tips, you know, gamers, nexus industry. Is that fan placement doesn't make that much of a difference. You might get a couple of degree difference, but no, you've got to take individual circumstances into account. And I could physically feel with my hand at least a 10 degree C differential on the inlet temp based on the location of the fan. So I, so I, I went, it absolutely must make a differential if I change that fan to the other side. And sure enough, it, it did basic engineering for the wind, you know?

Dave Jones: Yeah. I could physically feel it with my hand. I knew it would make a large difference and it did. So yeah.

Chris Gammell: No more shutdowns.

Dave Jones: No more shutdowns yet. No, my, my temps now running even rendering flat out. It's under 65 degrees. I think so. Oh, nice. Yeah. So anyway, that was very interesting, but it actually required me to do a video testing it to prove it to people that it made that much larger difference. People just didn't believe it.

Chris Gammell: So anyway, I haven't built like a, uh, is it like a desktop rig?

Dave Jones: It's a, yeah, it's a home built desktop rig. Yeah.

Chris Gammell: I haven't done that in a long time. And I feel like I should like some people, like when they, so like when I'm screen sharing with coworkers, things like that, they're like, your stuff is compiling really slow. And I'm like, yeah, and they're like, well, and then they show me what they have on their desktop. And I'm like, holy crap. You know, like sometimes just the compilation times. And it really, it does add up, you know, like it really, if, I mean, if you're doing a lot of code, I suppose, but which I'm not. But it makes me wonder if I should have like a, you know, like a, what they call it, like a front and back end kind of thing. And that's not what it's called, like a server and bin client, bin client model, right? That kind of idea. Yeah.

Dave Jones: Yeah. Yeah. Yeah. Yeah. People tell me to do that. Oh, why don't you have a gigantic bigger render down in your bunker, you know, like have a giant, you know, you've got all these dumpster PCs, you can set them up as a cluster and you can do mass rendering. It's like, oh, bloody hell. No, no, that's just insane.

Chris Gammell: I just buy, I just buy a thin, a thin client style PC just for like taking a trade shows and stuff like that. Cause they're just insanely cheap, you know, and like the specs seem good. I don't know. It'll be here tomorrow. We'll see how it, see how it comes up. But like having another, another box to just kind of throw in a bag, you know, like instead of a laptop, it's like, it's like, yeah, let's see if let's, let's test my metric in Dave. Let's see.

Dave Jones: Right.

Chris Gammell: Hang on.

Dave Jones: My temperature's dropping. My temperature. I, I, I fear something terrible has happened as if a million. It cried out. Cried me out and that was suddenly silence. Hang on. I'm going to go check.

Chris Gammell: Hang on. All right. My God. It's going to shoot a quick video. Editor.

Dave Jones: There we go. Sorry. Just had to shoot a video.

Chris Gammell: Well, we made it, we made it, we made it into the video. We got to mention that.

Dave Jones: Yep. You did make it into the video. Everyone. You're welcome. Yep. Yeah. No, it's just, yeah. The scope has crashed and it's stuck on the boot screen and it doesn't smell that good. So I, I, I haven't killed it. I just disconnected the power. Yeah.

Dave Jones: It just, this tiny little fan, like as a little, you know, yeah, absolutely. You know, as small as they get, it makes a huge, huge difference. So it looks like this one's not going to be passively cooled. I'm afraid it looks like you need the fan in there to keep it. Yeah. Well, that was the, that was why I tested it, you know, stress test. So, yep. Hmm. It's still air is the worst. Like, you know, like even having a little bit of airflow, even from your air con, right? Even from your air con on your roof, that's just, you know, gently blowing air just around the room. At least there's some movement, which can go through the vents on your case. Right. And it's not.

Chris Gammell: Well, I know all about dead air, Dave, because you just walked away from your microphone. Right.

Dave Jones: Yes, exactly. Well, you know. Well, you know, what do you want me to do? Ah, anyway.

Chris Gammell: I don't know. You said you're going to wait until it started on fire. I didn't sound. I didn't hear anything about fire.

Dave Jones: Oh, well, I'm not going to. It's going to cost me 1800 bucks, dude.

Chris Gammell: That fire alarm goes off. Oh, fire alarm.

Chris Gammell: That's interesting. I don't know if that's the thing here. So if you set it off, it's $1,800. If I set it off or it just goes up. I guess it makes sense that there's like externalities. Yeah.

Dave Jones: Yeah. Every office has its own individual sensor, which goes back to a monitored control panel and the, and our building, i.e. me, the owners, we all pay for the fire brigade to, and the alarm automatically goes back to the local fire brigade. And as soon as the sensor goes off, boom, they straight here. So, and, and then, and then they come to the panel in the base of first thing that is come to the panel in the, in the foyer and they go, okay, which unit went off, who, who burnt the toast this time? And yeah. And you get a bill for 1800 bucks for calling out the fire brigade. Huh? So yeah. Yeah.

Chris Gammell: It's not a bad idea.

Dave Jones: It's just, no, no, it's a bit, but we deliberately pay for that service because if there is a real fire, we want them to come as quickly as possible. Right. So we have to sort of eat that cost. Yeah. Sort of like in, you know, hopefully a real fire doesn't happen, but if it does, we, you know, it, it basically they get notified like instantly. So it isn't like somebody has to call them up and it's got to go through the, you know, emergency services number and blah, blah, blah. Right. So, yeah. So if we didn't have that, then it would require someone to phone it in. Right. It'd require. Yeah. It's like, you know, the building's on fire and they turn up. No, it's not. But you know, so we choose to do that. I think it's, it might even be regulate. It might even be council regulations, perhaps for buildings over a certain size. Don't quote me on that, but that, that wouldn't surprise me. So yeah. Anyway, it's pricey. I know many who've had an expensive breakfast burning their toast in this building. Let me tell you.

Dave Jones: Yeah. Yeah. Yeah. Hmm.

Chris Gammell: Speaking of burning things up.

Chris Gammell: Speaking of burning things up. I am thinking about, this is a terrible segue. I am thinking about burning up some routing bits at a PCB manufacturer. And I was wondering about your input on this. Yeah.

Dave Jones: Standard routing bit, 2.4 millimeters.

Chris Gammell: Yeah. Have you ever done controlled depth routing at a, at a PCB house? Yes.

Dave Jones: Yes.

Chris Gammell: We have. What have you done it for?

Dave Jones: I, oh, it's basically for very form, low form factor stuff where you have to embed stuff into the PCB so they don't stick out in certain locations where you have to get a form factor thing. So, yep. Yep. You can, they will do anything you want.

Chris Gammell: Yeah. Yeah. Yeah. Yeah. I have a display and I'm thinking about having like a, it's basically like reverse mounted and right now it's held down with very copious amounts of hot glue. Right. And I'm thinking about not doing that and instead of having like seated in the back of like a PCB display. Yeah. PCB that I'm making. And I've never done controlled depth routing before. So it would be nice to have that there, but then I don't know, like, like ideas on like cost adders or anything like that. Like what the cost is.

Dave Jones: Well, I'm not sure what the current cost is, but there are two ways to do it. One, you don't actually have to do controlled depth routing after the fact, because controlled depth routing is done once all your layers are bonded together.

Dave Jones: So if you've got a multi-layer board, a four or an eight layer, they will manufacture those layers separately and they'll stick them together. Right. They'll actually adhesive bond the layers together. So you can actually, at, at the point of where you're doing, say you wanted the two outer layers or something and you wanted a control depth thing, they, they can actually fully route that board before they then stick them on. All together.

Dave Jones: That's a good idea. Yeah.

Dave Jones: And, and then you get like a sharper cutout. You get a perfectly sharp cutout. Whereas the controlled depth ones, depending on the routing bit they use, some can be like a, a physical, uh, drill with like, like a tapered drill, but others can be like a flat-ended bit at like a mill. A mill. A mill bit.

Dave Jones: Yeah. An actual mill bit. You know, but so there's several ways that they can do it.

Chris Gammell: Hmm.

Dave Jones: Okay. Yeah. No, it's generally, you will not get a, like, like a shared panel. They won't do that on like a low cost. Yeah.

Chris Gammell: I think that's the real thing that it comes down to is like, I'm gonna have to either get enough volume or yeah, I'm just gonna have to eat it on that sort of thing.

Dave Jones: Or you can get two different boards made and then you stick them together yourself. Oh, that's kind of interesting.

Chris Gammell: Yeah.

Dave Jones: Yeah. Hmm. You can certainly do it the El Cheapo way. Yeah.

Chris Gammell: And I guess you could like do like a little sandwich, like the sandwich.

Dave Jones: What we talked about the other week, the, uh, boom box Joe, Joe Grand's boom box. Right. I, I actually, well, after that, I watched the full video of him and I, and he gave a talk as well. There's a video of where he gave a talk about how he's manufacturing the board and everything. Mm-hmm . So I'll maybe I'll confine that and we can link that down below, but I watched that as

Dave Jones: And that was interesting. And I, but I couldn't help, but think, geez, I would have done it the real cheap way and actually just manufactured two separate boards and just stuck them together.

Speaker ?: Huh?

Dave Jones: Like he went the whole controlled depth route into, you know, do all these large cutouts and cut the individual leads in there and cut like, oh, you know. Oh, that's right.

Chris Gammell: Cause he did that. He did that to get like the diffuse effect, like through, like he was, he was backfiring through the board, right? Yeah.

Dave Jones: Yeah. He was doing all the advanced techniques and I'm going, oh, that's fantastic. But it, geez, it costs some coin.

Chris Gammell: Yeah. I think he was working with a, with a house too, with a, a, a PCB house that wanted to showcase that stuff too. Oh, did he?

Dave Jones: Okay. Right.

Chris Gammell: Yeah. And he did, I think on one of those videos, he showcased. One of the PCB, one of the us based manufacturers. Oh, that's right.

Dave Jones: Yes.

Dave Jones: And then he's up. Yes, that's right. He was showcasing the manufacturer. So they, they probably did it for next, but that would cost him a lot. And, and he did an eight color silk screen pass. So he did eight different passes on the silk screen, you know, and, and, and, and, and custom dye colors to boot. Right. So, so he actually got the Pantone match colors that he wanted and they individually, they mixed eight different custom Pantone color mixes for him for the silk screen. Holy crap. Right.

Chris Gammell: So that, I think, so that company, it was a showcase. Take on the, don't say no, just quote them a price. Right.

Dave Jones: Yeah, yeah, exactly. So yes, they were really showcasing their capabilities there. They were pushing their capabilities to the max, just to showcase it. But if you can go to them and get that same thing, but just be prepared to pay probably 4,000 bucks for a one board. Right.

Chris Gammell: I am, I am not kidding.

Dave Jones: Right. No, it's not uncommon for us to get a board. Granted we got it turned in 24 hours, but we were paying 2,000 bucks plus for a prototype board. US. Yeah.

Chris Gammell: I think, I think it is like, it's been, I've had some, like, I've been dealing with people that are, that are having some jarring experiences themselves moving from like, I've been talking to people that are like moving from Arduino to Zephyr and like, that is very jarring. Right. Yes. For a couple of reasons. And then like, I think moving from, I actually had this when I was doing, you know, I got quoted like a normal PCB process in the US recently. And I'm like, what? Huh? I, I can get these so cheap from China and whatever. Exactly. Yeah. You, you can, but that's not on the menu right now, Chris. Like, and like, and then it was like, oh, okay. Yeah. So that's just the price then.

Chris Gammell: You know, like that's just. Yeah. Yeah. Deal with it.

Dave Jones: You know? Okay. Yeah. And was it as a, a shared service or were you paying for your own panel?

Chris Gammell: No, it was, I was going to get the panel.

Dave Jones: Yeah. Oh yeah. Yeah. You were going to get there.

Dave Jones: That, that instantly, right. When I was a boy, there was no shared service. You, you had to pay 500 to $800 minimum and you got the whole panel. So you, you bloody well took advantage of that. You went, oh, I've got to buy a panel split. But actually at work, this was fun at work. You know, you, you, you knew you had to pay for the panel. So you throw all your own designs on there because you were going to waste the panel anyway. Yeah.

Chris Gammell: Right. Yeah. Yeah. Yeah. Yeah.

Dave Jones: Basically it, it was a do it yourself shared service, you know?

Chris Gammell: Yeah.

Dave Jones: Yeah. Yeah. Exactly. About the backend. People upload their boards and software automatically just joins them together and wax them in. And you know. Yep. Yep.

Chris Gammell: But. Did you happen to see, I guess it's. It's not in our list because we didn't add it. But Scotty from strange parts. He went back to jail. Oh, I haven't seen.

Dave Jones: I've seen that in the videos there. Yeah. There. Yeah. And he went, they've got 80 lines or something. It's insane. Yeah. Yeah. But I can imagine. Yeah. Yeah. It's enormous. Like I've, I've been to ones that have 16 lines or something, you know, but 80 like what?

Chris Gammell: Yeah. I mean, it just boggles the mind, you know, like just, yeah.

Dave Jones: For those who don't know a line when we're talking about a production line at a PCB assembly house, this is not a piece. This is not a bare board manufacturer.

Chris Gammell: No, no, this was a, sorry. I, this was the thing he went to was a flex PCB manufacturer. So it was flex.

Dave Jones: Oh, it was a flex PCB manufacturer. Oh, okay. Anyway. So line all mean the same thing actually.

Chris Gammell: Yeah, that's right.

Dave Jones: In, in both cases, in the case of an assembly house, which is assembling the components on your board with the pick and place machines, a line is. Literally line. It literally is a row of pick and place machines with the conveyor belt, but it's, but it's the, the auto stenciling machine, for example, the auto paste machine. And then the, or then the pick and place machine, then the reflow machine. And then the optical inspection machine. And then the bed of nails inspection machine. And they're all literally in a conveyor belt line. And your board just goes, travels on the conveyor belt. That's why if you're designing a panel, a PCB panel for any sort of production, you need the holes on the outer edge of the board. So it can go into the sprockets that then can shoot your board along the conveyor belt and, or, or, or can actually align your board into the sprigets for the bed of nails, you know, tester. So it can get accurate alignment and all that sort of jazz. So there are certain requirements, there are certain standards for the strip along the outer edge of your board, which has to go through the rails on the, on those lines. So anyway. Yep. Yep. That's beyond the scope of this. Beyond the scope of an amp hour episode.

Chris Gammell: I think it's still relevant though. Like it is like, you know, I think, you know, most, the most efficient processes are going to have, you know, just everything just flows through, right?

Dave Jones: Anything that stops the line, big, literally big red lights, strobe lights flash when, when, when the line stops. People start running. Yeah. Yeah. People start running. Yeah. Yeah.

Chris Gammell: I think the most, the most impressive piece of equipment that was on there is they said like, so then they're doing, I think it was, must've been photo resist. Instead of doing, instead of doing like exposure, like straight down through a, through a transparency, like how many of our PCBs are still made. But yeah, they have it like where they're, they're doing a line raster across. And as the board flows through at a constant rate, it's basically like a laser cutter, but it's just lasering off.

Dave Jones: Yes.

Chris Gammell: You know, just that, that one line. And as it passes through, yeah, it's just, it's nuts. I mean, that crazy.

Dave Jones: Okay. I definitely have to watch that video. Yeah.

Chris Gammell: Oh, it's, oh, it's so good. I mean, like JLC paid for it. Right. So like, you know, there's no hiding that fact. Right. Yeah.

Dave Jones: Oh yeah. They've invited me to go over there, you know, and they'll host me and they'll take me through the production, the whole facility.

Chris Gammell: I'm sure the scale would be, yeah. Fantastic. Yeah.

Dave Jones: It's absolutely mind blowing. Yep.

Chris Gammell: But yeah, that, that one machine really blew my mind. Yeah. Especially because like, you think about, I mean, like even there, there are these not needing custom tooling. Right. I think about like when you have tooling involved, the kind of the NRE cost goes up. Right. So you are getting a stencil made that some flat cost because someone has to go and cut out or etch out your stainless steel stencil. And that is like a fixed cost for your build.

Dave Jones: And for those who don't know, it isn't just a stencil. Then they have to frame it because these automated paste machines have to have a big frame. They have to have a big metal alloy frame on them going around the outside. So it doesn't warp and so it sits in the machine properly and you know, et cetera. Yeah. It costs money.

Chris Gammell: And, and so you have that kind of like fixed costs, but then you start to have tooling that has removed it. And, and so like in this, in the solder paste example, right, there are these like automated stenciling machines, but even those, they're like basically like inkjet printers for solder paste, which is awesome. Those still aren't that same like linear kind of way.

Dave Jones: They aren't as fast. They're usually for a prototype things. Usually for the high volume production stuff, they use, they have the big squeegee. It's an automated squeegee that goes in the, you know, it's, it's totally automated, but it's a big squeegee because you can do it all instantly. So yeah.

Chris Gammell: Yeah. Just thinking about like process, process, uh, standardization and stuff like that. So something like that, that laser machine was really, it really blew my mind. I mean, like, yeah, I had not seen anything like that. Of course that, that, that kind of stuff exists, but man, that was cool. Yep. So. Ah, yes. Good stuff. I don't have any need for flex PCBs, uh, and they're expensive. They're still even, even with like low cost. I don't think. Yeah. I mean, they are.

Dave Jones: Yeah, they are.

Chris Gammell: They're way cheaper than they used to be. Don't get me wrong. Oh yeah. No, totally. But like, you know, when you compare them to like FR4.

Dave Jones: Well, what is your idea of expensive?

Chris Gammell: Oh, that's a good question.

Dave Jones: Like, is it 50 bucks? Is it 500? Is it, you know, I, I did, I thought they were low cost flex services these days.

Chris Gammell: Oh, there are. Yeah. I mean, like, uh, but like unit cost.

Dave Jones: I got a flex PCB a couple of years back and it didn't cost much at all, but it was literally just like a little, uh, two layer.

Chris Gammell: Like one sided.

Dave Jones: And, and, and it was an inch by an inch. It was just like a, um, we were doing a mod. On the board. So it was like a little, so it was just like single chip sort of modded out to, you know, something else. It cost dollars or something. It wasn't, it wasn't much. It's really true. Yeah.

Chris Gammell: I think I, yeah, you're probably right. I mean, I think I was comparing it to equivalent FR for, you know, size and that's not a fair comparison, but I, last time I was using it, I was using on, I was working on a consulting project for a, a wearable that had like a bunch of LEDs on it basically. And so like a big matrix of LEDs and it was on flex and it had to like curvature, it had like curvature to it and it was not cheap, you know? Okay.

Dave Jones: All right.

Chris Gammell: But probably like five, six bucks for a PCB, but like in decent volume, you know, like a couple hundred volumes still. Now it wasn't, I didn't see the drop off either. Right.

Dave Jones: Okay.

Chris Gammell: So that was the other thing. Right. Yeah. Yeah.

Dave Jones: Hmm. And, and it depends on site. It depends on all sorts of things, whether or not it's just flex on its own, whether or not it's a rigid flex, whether it once again, it's got, you know, uh, you know, tiny micro V is on it and you know, all sorts of things.

Chris Gammell: Exactly. Yeah. Yeah. Like what the process nodes are. Yeah. Yeah. Totally. Yeah. I, I, I don't really have any need for that stuff. And currently and you know, someday maybe. All right. But not, not currently. I feel like you need to have like fancy, fancy plastics that need that have curves on them. And, and I currently don't do that.

Dave Jones: Got it. Well, I was going to, um, speak when I was talking about the scope before and we'll, I was actually going to get into the production assembly side of things because in that tear down, you no doubt saw that I was a bit flabbergasted that there's two chips missing in there. Right. Which is for the extra memory for the logic analyzer option, which is not even available in this model. Right. So it's not available in this model at all yet. They physically remove the chips obviously right to save cost. But then they left in all of the bypass caps and all of the termination resistors. And I'm not just talking about one bypass cap, I'm talking about 20 bypass caps and 20 termination resistors. It's like, why? And then somebody on the forum, of course, or in the comments, what?

Chris Gammell: Let me, let me guess. Let me guess first. Yep. Go. I think it actually relates to what we were talking about earlier with lines. I would guess that they go through two lines and they're completely, they're, they're separate, but you just put them through because the cost differential for. It's so little. Yeah.

Dave Jones: Yeah.

Chris Gammell: Yeah. It's so little. It's, you were right. It was not, it was not zero. It's not zero to have those caps on there, but is the cost difference. Like maybe it's even at the point I can imagine a high enough thing, high enough volume, volume is the word. Yes. Thank you. Yes. I do electronics. Yes. And that, that actually time. Even you ship it between buildings.

Dave Jones: Yeah. Time handling rack space, because if you've got two different models of boards, they have to be individual, they have to have their own individual rack space, their own individual, you know, handling and everything else. So yeah. So they handle the board, but that's not the thing that one of the people on the forum, or one of the people in the comments mentioned, they mentioned, oh no, it's for an EMC thing. They've got to have the termination resistors on there so that the lines don't radiate and stuff like that. They got to have the bypass caps on there. And went A, okay. Which is a reasonable comment, but A, the bypass caps and caps aren't going to matter a rat's ass. Okay. And B, the termination resistance. Yeah, of course. But if you're dumb enough to make a model that then radiates into deadlines. Like you, you couldn't even turn that off in firmware.

Chris Gammell: You have a firmware thing and you say, don't send stuff to the memory area.

Dave Jones: Don't send stuff to the memory area when we know the memory's not fitted. I mean, that's just ludicrous.

Dave Jones: So, yeah. So I think it must be, as you said, it must be a, just a cost trade-off handling, you know, like factory, you know, issue. Yeah.

Chris Gammell: So it's just easier for them to leave them on. Yeah.

Dave Jones: But of course you wouldn't want to populate the expensive chip. So you would leave that off. Yeah. Right. But the caps and, caps and the resistors.

Chris Gammell: Given our, our relatively recent bubblegum tap shoe history here, Dave, you don't actually know if they just couldn't get the chips. Right. Okay. Yes. That actually happened on a dev board I saw recently where they, all the documentation said, yeah, there's an accelerometer right there. Oh, they didn't fit it. Yeah.

Dave Jones: And it's just not there.

Chris Gammell: And it's like, oh, that, that is just, that is a thing from the shortage, I think. Yeah.

Dave Jones: Right. And it's like, you know, it's, it's real.

Chris Gammell: It's hard. It's hard to, to like get this across to people that, that haven't done a hardware. I don't have a good way to say it, but like the best I've come up with is you have to be perfect. You at least have to have everything in stock. One part will hold up an entire production line. I feel like that doesn't, that doesn't really get through to people, but like that idea, it'll just blow up. It's going to, it'll ruin your weekend. It'll ruin your month.

Dave Jones: You know, it can ruin your career.

Chris Gammell: Yeah. It'll ruin your career. Yeah. Yeah. Yeah.

Dave Jones: Yeah. Of course. Oh yeah. No, definitely. Oh, you know, your manager comes in. Oh, are these boards going to ship next month? Oh yeah. Yeah. No problem whatsoever. Yeah.

Dave Jones: Everything sorted. And then one part doesn't come through and you are, and you can't get an alternative for it because she didn't plan for it. Then you're screwed. Yeah.

Chris Gammell: Totally screwed.

Dave Jones: Yep. Yep.

Chris Gammell: Yep.

Dave Jones: Or sometimes you have to quickly do a bodge board or something to use a different part. Like a, a totally different footprint, you know, like, yeah. So, uh, yeah.

Chris Gammell: What a nightmare.

Dave Jones: Yeah.

Chris Gammell: Anyway, I wonder if, I wonder how many people do this. I wonder if, I wonder how many careers have ended because of the shortage.

Dave Jones: Yeah. Oh yeah.

Chris Gammell: I'm sure. You know, there's like the ones who are like, ah, you're fired. I do know people. And then there's. Yeah.

Dave Jones: I know several, I know several people who have, who made like, you know, they were selling boards on Tindy, they were doing stuff and they just went, no, I am done. I'm done. I cannot get these parts. I'm shutting down. I'm not going to be in this business anymore. I'm done.

Chris Gammell: Yeah. I was thinking how many people who are said, uh, you know, present company included who said, you know what, maybe I should do a hard or software for a couple of years.

Dave Jones: Right. Oh boy. Yep.

Chris Gammell: You know, Dave, the good old days, you know, like 2022, where you'd, you'd, you'd, you'd, you'd, you'd pet your wrist for, for an STM 32 part, you know? Right.

Dave Jones: Yeah, of course.

Speaker ?: Oh boy.

Dave Jones: And that, that was one of the reasons that I didn't go ahead with the microcurrent at the time. Sorry. The micro supply at the time, I've done a recent video on showing where the micro supply got to everything that, that was, that was one of the reasons, you know, is that COVID had just started and the supply chain crisis had just started. And I used two different STM 32 micros on there. It's like, you know, That's a rough go, man. That's a rough go.

Chris Gammell: I don't want any of this pain.

Dave Jones: You know, it's like, that was, yeah, that, that fed into the equation of why I went, eh, no, I've got no enthusiasm to do this. No, I just don't want the fight. No, no.

Dave Jones: Yep.

Chris Gammell: Well, and I was asking, so Scott, Scott Williams was back on a fellow Aussie of yours. He was on last week and I was asking him about like standard, standard parts and you know, like what they start with and they all make sense. But like, so he said like ESP 32 for wifi and RF 52 for Bluetooth. And I'm like nodding my head and he's like STM 32. Otherwise I'm like, Ooh, right. But that is totally a, like a good suggestion, right? Like up until two years ago, it was a great suggestion. And now it's probably a good suggestion again, you know, maybe, maybe deal with a GD 32 or a RV 20 32 or all the ones that are like cloning these things. Right. But, but the fact that there's clones is pretty good too. So.

Dave Jones: Yeah. Oh boy. Yeah. It just, it's, it's, and it could come back. Like all this drama could come back. I mean, China's in a spot of bother at the moment. And, and the, and the political tensions and everything happening, you know, I, I wouldn't rule out. Yeah. I would not be betting anything on China at the moment. I would not know if, if you are reliant upon China at the moment, you could come and gutter.

Chris Gammell: Yeah. So this is an interesting thing. Again, that didn't get posted to the, um, the subreddit because I'm lazy. Just came up today, but there was a, it was a news article about, uh, let me, let me find the, uh, Intel, it was on Tom's hardware. I'll send it to Dave here. At least Intel announces arm investment talks up risk, risk five. And I read that and I'm like, in my head, what I'm thinking is Coca-Cola announces Pepsi investment. Yeah. Yeah.

Dave Jones: Yeah.

Chris Gammell: That's just like the words, the words don't go together. Right. You know, like just like, what? And then you dig into a little bit and, and like, it's like, okay. Okay. So Intel foundry group actually talks arm investment and talks up risk five because the foundry group doesn't give a crap if you're using X86.

Dave Jones: They're just making as many chips as they can.

Chris Gammell: Right. We've talked about this on the show many times before.

Dave Jones: I'm just like, it's the fab. The, the, the foundry group is the fab and they will make your chips. If you're a big enough, uh, player, you got enough money. Who cares? Yeah.

Chris Gammell: Yeah. Bring it on, man. Yeah. Yeah. So yeah, it's, uh, quite, but it's just like that, that stuff is, uh, yeah, it's funny. It is interesting when Intel, like Intel and all these are like, uh, who was the other one risk five. There was another risk five thing where I think it was the couple of weeks ago when we were talking about it broad, no, Qualcomm, Qualcomm invest in risk five too. And it's just like, oh yeah. Yeah. Okay. All right. Yeah. Like, and some of it just feels like it's kind of like, well, we're at the cool kid table now, you know, like risk five is like the new hot thing. And we need to recruit around here too. But like, it just, you know, like these walled gardens based on, you know, they're building companies based on being walled gardens on their own thing. It's like, are you guys, you guys are here. Okay. All right. All right. It's that kind of party. All right.

Dave Jones: Oh boy. Yep. Can we circle back to the niche product market again? You know, the niche Tindy thing. Cause I just did, I, well, you know, it's not really related to any related to eBay actually. And in that I just read eBay last time too, right? Huh?

Chris Gammell: You talked about eBay our last episode, you and I recorded. Yes. You talked about buying power modules on there and things like that.

Dave Jones: Yep. Yep. Well, here it is again, right? I was repairing a, um, Agilent slash Keysight, uh, U1273, a meter, right? And it's got this OLED screen and this OLED screen so famously fails, right? In these meters that somebody presumably in China, somebody in China, they've gone to the effort, not only like, cause you can't really buy the OLED anymore, but you're really, even if you could, you really wouldn't bother because it's got some manufacturing defect that fails in all these meters. Apparently all, all these Agilent Keysight, well, you know, not all, but. Yeah. Proportion. Yeah. I could, I, there's enough of them that have failed out there. It's so infamous for it, but somebody went to the effort to reverse engineer the OLED protocol, which isn't hard if you've got the, uh, data sheet and everything else. Right. It's not, it's not massive, but you know, somebody went to the effort to redo this and then they used an STM 32 micro to drive a, to then interpret the signals coming from the meter, which expects to talk to one OLED. And then they translate the protocol to drive a different OLED screen, which you can actually buy. Wow. And yeah, they went to that effort and they sell lots of them on that. Well, they sell, you know, they sell hundreds of these things because there's, you know, hundreds or thousands. That's great. Yeah, exactly. Somebody went, somebody spent like a week actually designing this board. Right. Oh, and it's, okay. That's probably in China and they probably don't speak English. Who cares? Well, they found a niche, right? They just found this niche and went, I'm going to do this, dammit. And sure enough, you know, there's this one source for this board. Okay, here we go.

Chris Gammell: Here we go. How much they charge for that kit?

Dave Jones: Oh, it was expensive. It was 80 us bucks. All right.

Chris Gammell: You see, it was a lot of money. Right. It was probably like, it's like, it's like, it's like, you know, it's IP in there.

Dave Jones: The OLED screen is we, which I couldn't find info on. So it's probably like, I dunno, some, they got some bulk lot somewhere.

Chris Gammell: They know somebody. Yeah. Right.

Dave Jones: Yeah, exactly. Oh, they, they got them for 10 cents a pop because they were surplus or something. And they, yeah. And they designed this board, which just has an STM micro on it and a little, uh, convert and, and a little, uh, converted to boost the voltage up a little bit. Yeah. And yeah, it was 80 bucks us.

Chris Gammell: Wow. Yeah.

Dave Jones: But, but, but you're so desperate to fix your $500 meter that you'll pay the 80 bucks for the board.

Chris Gammell: Yep.

Dave Jones: Right. Perfectly priced.

Chris Gammell: Yeah, exactly. Yeah. I was looking at one of the e-ink screens that was on my desk today and I was shocked, shocked that it was a 2017, uh, date code. Right. And my co-worker said, well, of course it is. Like we're buying them one at a time. Like they're not like making a fresh batch for you, Chris. It was like, Oh yeah.

Dave Jones: Right. Right. Right. Oh boy.

Chris Gammell: Yeah. Yeah. I'm buying on Ali express, whatever. Like just buying onesies, two Z's, you know, like that's not, that's not how production works.

Speaker ?: Yeah.

Chris Gammell: I'm not, you know, and there's not enough volume that it's moving where yeah. Like 2017 is.

Dave Jones: Yeah. Yeah.

Chris Gammell: Cool.

Dave Jones: Six years ago. Right. Yeah. It's not uncommon to get old stock of stuff, you know? Yeah. Yeah. Well, even when I opened up that Rygol scope it, the, uh, the ASIC front end, I think had a date code from a couple of years back. So obviously when, when they designed that ASIC, they ordered a million of the bloody things and they're still using them. Right. Yeah. Right. What's interesting about, so the ASIC was 12 bit, but it's a 12 bit analog to digital converter. I think it's a custom Rygol or they might've teamed up with somebody to do it. Sure. Somebody mentioned that on the forum. Yeah. They teamed up with TI or AD or something, you know?

Chris Gammell: Well, that's what I'm wondering is like, so like many, many years ago, they, they weren't selling 12 bit scopes. Were they? No, no, no.

Dave Jones: This is a new thing. Well, it started 12 months ago. They, they did their first, uh, 12 bit scope, which was the 4,000 series. And then it went down to the 1,000 series.

Chris Gammell: Well then how does the date code match up though? But that's what I mean is like, if the date code is from a couple of years ago when they weren't selling 12 bit scopes, then it probably.

Dave Jones: Well, it takes a while to design those things. Sure.

Chris Gammell: No doubt. Yeah, no doubt.

Dave Jones: So they, yeah, they obviously made them. You know, they, they, they designed it a year. It would have been designed and made like a year before that scope was released. Right?

Chris Gammell: Sure. Well, so what was the date code? You said, I'm just ballparking.

Dave Jones: It was, I, I don't remember. It was like a year or two ago. Right? Oh, okay. It certainly was not a new part. It was not a new part. Right? So even though this is a brand new scope straight off the production line, it used this custom AC chip from a couple of years ago. So, you know.

Chris Gammell: Okay. That makes sense. That makes sense. Yeah. I, I talked to one of the designers on an Agilent scope at one point. Was it Agilent? Maybe it was a tech scope. I don't know, but they were using like really old Silicon, like, because they were using like super fancy processes too. And like, I admit, I forget, I must've met them back in my test and measurement days.

Dave Jones: Well, tech are famously still using the ASIC from 30 years ago.

Chris Gammell: Hmm. Oh really?

Dave Jones: Yeah. Well, I don't know. They had just stopped. Yeah. It was the, any, any tech scope that has 2K of sample memory. It is using the 30 year old ASIC in it.

Chris Gammell: Oh, interesting.

Dave Jones: It is. It is using, they milked that thing for, well, at least 25 years, I think they milked that ASIC.

Chris Gammell: Yeah. And like, would they, was it like the same thing where they batched up the chips and, or is it like, they just kept making the same. No, no.

Dave Jones: I think they're all old stock actually. Yeah. I've, I've talked to tech about this at one point and they said, oh yeah, we've still got a lot of stock of that chip. We've still got a lot of stock. I don't think anyone makes it anymore. It's like, they just ordered like, you know, maybe the fab that they were using, there was some like, it's not that all the chips are 30 years old or 20 years old or something, but maybe, you know, a decade ago, they, the fab just went, no, we don't have this process node anymore. Final run. Final run and tech went, okay, we'll order a million off them, please. You know, I don't know the exact number, but you know, yeah. Yeah. They said that there's a lot of those chips.

Chris Gammell: Yeah. At a fab, you know, like fabs always do like process node reductions and stuff like that too, where they might, they might obviate an old process or something like that. Yeah. But then, but then it wouldn't be worth it for tech to re up and like, no, no, no, it

Dave Jones: was not worth, it was not worth spending any engineering effort and costs to actually redo that silicon because they've got newer, more advanced designs now, but, but they still want to sell that scope because there's still a market for it. So you just manufacture the old chip again. Which, which scope was it?

Chris Gammell: Which scope?

Dave Jones: It started out as the TDS 210. The original TDS. Yeah. The original TDS. Oh, I hated those things.

Chris Gammell: Oh yeah. But that was groundbreaking, dude. I know. We've talked about it on the show. No, I, I don't, I don't actually harbor any ill feelings against it. It was just that like, I have negative associations with like my. Oh yeah. By modern standards. Yes. By modern standards. No, no, no. My frustrating early days of electronics where I didn't know anything was going on. Oh, okay. Right. I was using those things and I was just like, what am I even looking at here? Right. It's not a, it's not an actual product problem. Yeah.

Dave Jones: And then it ended up being used in their 2000 series scope as well. And that way they're 1000 series or something. Oh yeah. It was used in their modern scopes up until. I don't know if they're technically still shipping it. Don't quote me, but even as little as three, four years ago, they were still shipping scopes with that chip in it. Hmm. Yeah. Interesting. It's crazy. Yeah. And then it went into their, and then they remarketed it as the TBS, the tech basics series, which was designed to, which is tech's cheapest scope, which is like 300 bucks or something, which they sell to like labs, like a school labs and things. Uh huh.

Chris Gammell: Yeah.

Dave Jones: Yeah. TBS 1000. It's still on their website.

Dave Jones: I think you can buy a TBS 1000 series scope. Yeah. I think one gig sample per second. Yep. It's still the same. Uh, what's the, what's the sample memory? Yeah. Like, you know, it's probably 2K or 4K. Maybe they've used two chips in there. Yeah. Yeah. It probably wouldn't hold much of a candle. Oh, 20K. How have they done that? Yeah. 20K of sample memory. So that might be a variation on the, on, on that ASIC or something, but it's still, yeah, it's still that ancient thing and they're still milking it. You can still buy it. Yeah.

Chris Gammell: I'll happily sell it to you.

Dave Jones: Yeah. Hmm. And there's nothing wrong with that. If there's a market to sell into you, you just keep selling.

Chris Gammell: I mean, some people say that Tektronix tries to elongate the, uh, the usefulness of their silicon, much like, uh, pioneers do with a 555 circuit on a, on a, uh, gimmicky circuit board do, uh, well, hang on.

Dave Jones: Don't just sing my model.

Chris Gammell: That's what I'm trying to say. Yeah.

Dave Jones: They're copying my model. Okay. Well, they're copying the best. I mean, but don't just single out Tektronix. Hey, Keysight. I mean, remember the Keysight Megazune for ASIC is now 14 years old. Sure.

Chris Gammell: Sure.

Dave Jones: There has been no upgrade. They are still selling all of their current scopes. Use the Megazune for ASIC chip, the exact same ASIC chip with the exact same built-in four meg of memory. And there's nothing on the horizon. Yeah. 14 years. Like that was one of the first videos I did was, was the release of the Megazune for ASIC. It's one of the first videos I did on my channel and it was, yeah, I think it's 14 years now. So yeah. Yeah. It's nuts. Whereas I, I actually invent, I actually researched this at the, um, maybe five years ago when I went, well, where's the Megazune five coming out? And on average, they had released a new Megazune for ASIC every eight years. So they started out with the Megazune, then it was the Megazune two, then the Megazune three, and then the Megazune four. And then, so, so, so it went eight, eight, eight, eight, and now it's 14. So, I don't know, something's going on. Sounds like they reached the top, Dave.

Chris Gammell: They just can't, they can't, you can't improve perfection. Yeah.

Chris Gammell: Okay.

Dave Jones: Yeah.

Dave Jones: But once again, they still sell bucket loads of those scopes, right? And they're really great. And they're still the faster, one of the fastest things in the business. Right? Yeah. Yeah. Yeah. It's only very recently, 14 years later, that Roden Schwartz now beat him with four, uh, four million waveform updates per second versus a million. You know, it's like, it's only recently that somebody has actually beaten them.

Chris Gammell: So maybe they're taking longer because they're doing the ultra zoom instead of the Megazune. Maybe that's why.

Dave Jones: Right.

Chris Gammell: Right. You know, it's, you know, you know, that every time they do a spin. Ultra zoom is a trademark.

Dave Jones: I think that's a trademark of Rigol. Oh my God.

Chris Gammell: It's terrible.

Dave Jones: Which ironically, ironically, not supposed to tell you this, but Keyside D. Uh, in the early days. Uh, in the early days. Uh, in the early days.

Dave Jones: Uh, in the early days. Uh, in the early days. Uh, in the early days.

Chris Gammell: Uh, in the early days.

Dave Jones: Yeah, they re-bedged it. So they, their engineers helped out Rigol back when Rigol were a nobody.

Chris Gammell: And that's how the innovators dilemma goes, Dave.

Dave Jones: Yeah. Right. And I think Keysight might be regretting that association now because Rigol are doing their own custom Asics. They're one of the big players now. So yeah. I mean, that's the thing.

Chris Gammell: Every, the low end people always move up market, you know, I feel like there is no better place for that than car makers. I feel like anytime I see a low cost car, I look at him like in 20 years, that's if you survive 20 years, you are going to be a fancy car. Right. Because it's like, because it's just, you know, they, they have to be efficient at the bottom end. And, and then if they can survive, then of course they start hiring more marketing. They have better designers. They, you know, they get a better brand name, whatever. And yeah, that's, I guess car, car brands do a little more often though.

Dave Jones: Right. Anyway. Yeah. The scope makers are anyway. I think we do have a new revolution. I haven't done my review video yet. I haven't shot it yet, but I think, I suspect we have a new revolution in entry level scopes. It started out with the Tektronix 210 series. They were the first to pioneer a digital entry level bench real time sampling scope. And then it was the Rigol with the 1052E. That was the one that Keysight helped them out with. The 1052E. And that became the new benchmark standard. And it was the benchmark standard for seven, eight years, maybe until, you know, a couple of players like Siglin came along and eventually gave better bang per buck, but there was no real revolution there. And then Rigol did it again with the 1054Z, which four channel scopes became the new entry level standard. Now you're considered nuts if you buy a two channel entry level scope, right? It's just nuts, right?

Chris Gammell: You said that in your thing as well. Like in your review, you said, why would you even buy the two channel?

Dave Jones: Why would you even buy? Yeah. And why would you buy the two channel? They still make the two channel version, even if this new one.

Chris Gammell: What is it? What is the cost difference?

Dave Jones: The, it's a $60 difference between two channel and four channel and, and you get the two extra probes, right? I mean, sure. In, in volume, those probes cost five bucks to make, right? But you know, still, right? You know, and yeah, so it's a complete no brainer, right? You'd, and that started with the Rigol 1054Z and that's been, and once again, Siglin came out with a better bang per buck version, but it wasn't really revolutionary. And now, Siglin with the DH0800 series for the same entry level price of 389 Yankee bucks, you get 12 bit. Not only do you get four channels and 12 bit. So they've moved from four channels, 8 bit to four channels, 12 bit. Not only have you made that move, you also get HDMI out of standard. You get VESA mount of standard. You get USB-C power standard, whether you like that or not. That's a different question. You can do a whole episode arguing over that. And you get LAN, serial, decoding, everything built in standard and, and the, and, and it's touchscreen. Right? So, so the whole entry level market has changed again.

Chris Gammell: I guess we really can, we really can have it all, huh, Dave?

Dave Jones: Yeah. I mean, you know, yeah, there's a few shortcomings to it, I'm sure. But geez, you know, like they've, I think they've changed the, I think the entry level scope landscape has changed again. So this is the four, this is the fourth change that I see started with tech and then two Rigol and then now third Rigol change. And the others will have to copy.

Chris Gammell: Now there's absolutely no choice. You had mentioned that it's a Android. Like it's an Android OS. Yes. Yeah. Android. Is there any downside to that or no? I don't really know what the other ones would be.

Dave Jones: Not really. It's just meh. You know, I don't care. Like it doesn't, does it care? You can play doom on it, I guess, you know, right. And the other interesting thing is, is that you might've noticed it actually has an SD card in there containing the operating system. It's not built into the chip. And I've actually dumped that info from the SD card. So you can get that on my forum and people on the forum have actually very nerds, you know, about these sort of OS things. They actually verified. Yeah. The entire OS is on the SD card. So once again, that's a, that's a deliberate change to it being on the physical chip. Sold it on the board. So whether or not they did that for cost saving, whether or not they did that for flexibility, don't know, don't know. So yeah. Anyway, it's certainly an interesting change. Landscape shifted again. Thanks Rygo. This, this episode is not sponsored by Rygo.

Chris Gammell: It's not, not sponsored by Rygo. Yeah.

Dave Jones: But they just happened to be, you know, I just call it the way I see it. And yeah, I think it's changed again. So yeah, I'd sell your DS 1054 Z pretty quick. It might not be worth much on the second air market in a few months. Maybe. I don't know. Once again, I haven't actually reviewed it.

Chris Gammell: I gave it away when I moved. So yeah.

Dave Jones: The, I I've, I've still got a DS 1052e at home. Yeah. Yeah. Yeah. Yeah. It's still there, but yeah, no, you wouldn't use it these days, but technically I think you can still buy it. I think, I think you can still buy it and it costs more than the full channel 1054 Z. But like, like, I don't know. They've just got so much stock of these things and you know, I mean, they're bloody warehouse somewhere in an infinite warehouse that I don't know. They just, you can buy it apparently. It's just nuts. Oh my God. Anyway. Yep. Technology marches on.

Chris Gammell: And so do we. Our amp hour is up, Dave.

Dave Jones: It's up. Yes. We're gone.

Chris Gammell: I'm going to march on out of here.

Dave Jones: Fine. All right. Catch you next time. See you.