#536 – NFT Schematics

Chris Gammell: This is The Amp Hour Podcast. Released March 28th, 2021. Episode 536. NFT Schematics.

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: We're going to be rich, dude. We're going to be rich.

Chris Gammell: Are we? Yeah.

Dave Jones: We're going to, yep, yep, yep. We're going to sell our schematics. I'm going to sell my firstborn video. I'm going to, yep.

Chris Gammell: Yep. And how are you going to do this, Dave?

Dave Jones: I'm going to do it with non-fungible tokens. NFTs.

Chris Gammell: I do like the word fungible, but everything else about this, I hate. Everything else about this, I really hate.

Dave Jones: You brought it up, dude. You put it on the list. I did. You're guilty. It got zero votes, and we're starting off with it.

Chris Gammell: This all started as like this angry screed I had written out. And then I was like, yeah, I'll just post about it. Because I thought it would be funny to talk about it on here. But yeah. Where did you post it? On Twitter? No, on this Reddit post. I had this whole long thing about this.

Dave Jones: Oh, okay. Because you didn't post anything.

Chris Gammell: About how silly I think they are. Yeah. I just, I.

Dave Jones: So you wrote it all and then you deleted it. And just backspaced it. Yep. Exactly. It's quite a few people commenting on here. Yeah. Yeah.

Chris Gammell: So why don't you explain what these are, since you're in the whole crypto world. And I think it's ridiculous.

Dave Jones: Well, it's a, oh, it's good. If you want to get into the intricacies of it, but. No, I don't. It's hard. No. No. Basically, a non-fungible token is using a blockchain. It doesn't have to be any particular type of crypto, although some are more suited to it, to basically verify that you're the owner of something. And that was one of, it's not a new concept. That was one of the original issues for block, you know, one of the things that blockchain could do is that it's a permanent ledger. Right? So it records. Everyone knows that you're the owner of that Bitcoin. Yeah. And, but it doesn't have to be actual Bitcoins. It can be a products. It can be, you know, in this case, it can be digital artworks or something like that. So yeah, it's, it's just a way to verify that you own a digital asset. Well, you own an asset. It doesn't even have to be digital. It can be a physical asset. It can be like a Rembrandt painting or something. Right? A physical.

Chris Gammell: It could be like a deed, a deed to a house.

Dave Jones: It can be a deed to a house. It can be all sorts of things.

Chris Gammell: It's a paper that says you own the thing, right? Yeah.

Dave Jones: It says you own the thing. And a non-fungible token is, is basically that it's using the blockchain to verify ownership, but it's doing it for the, the, where all the hype is at the moment is digital assets. So you can sell tweets. You can be the owner of Elon Musk's first tweet.

Chris Gammell: Right? This is where I start to fall apart. Yeah. This is where asset asset is being used very loosely here. Yeah. I know asset being thing with value and everything that I've seen is very without value. Oh yeah.

Dave Jones: See, see, that's the thing because it's a digital thing. Like it could be a PDF or a tweet like everyone else or a picture, right? That artists are selling their digital, their digital JPEGs as.

Chris Gammell: And this is where it all started with this. What's the stupid guy's name? It's been going on for a long time. Binky. No. No. What's the guy's name? He's got a, he's got a stupid name. It is. Is it an artist? Beeple is the guy that sold, this is why it's in the news. Cause it, I mean, it started, he sold a painting, a digital painting for like $66 million. Yeah. I know.

Dave Jones: This is like, yeah, this is insane. Right. Whereas anyway, a digital painting, whereas you can own that exact same JPEG. It's identical in every way. You can own it for free, but somebody paid $67 million to be the verified owner of the original in quote marks. Right. Even though it's a digital thing that you can copy with 100% perfection. Like it's just, it's, yeah, it's insane. So it's relying upon the fact that down the road, somebody else will be even dumber than you are to pay more than you paid for it.

Chris Gammell: Yeah. Right. And so this is my next argument against it. It is the new, the nouveau Beanie Baby.

Dave Jones: Right. See, I was not into the whole Beanie Baby thing. I didn't, that wasn't here.

Chris Gammell: You think I was? You think, you think I was?

Dave Jones: Well, I don't know. You'd know, it was an American thing. I think you'd know more about it.

Chris Gammell: American thing. Yep. Yeah. No, I was right in there. Yeah.

Dave Jones: Yeah. It's just, and pet rocks and, you know, like, yeah. Well, not, not quite. Cause it's like, there, there are uses for this. It's not absolutely useless, right? There are real legit applications for non-functional tokens.

Chris Gammell: I mean, there was, there was uses for tulips in the Netherlands in the seventies.

Dave Jones: No, come on. You, you said it yourself, right? Deeds to houses and things like that. Right. To verify ownership.

Chris Gammell: You could have planted, you know. The, the, the, the famous example is the tulips in the Netherlands, right? Yep. They had this like crazy valuation that kept going up and up and up and you could have planted them. You would have been wasting a lot of money that you spent on. Right. Yep. Yep. Yeah. And then it just, I thought it was funny because like thinking about this, like for a schematic or for an electronics design, you know, it's the thing that actually, I mean, maybe the design itself doesn't have any hard value until you actually make the thing out of it, but you could say that the IP and the data around it has, has value.

Dave Jones: You could say like, well, what's the world's, what's the most famous schematic? Like a lot of people say it's the Apple one schematic. Right. Right. So in theory, if Apple came out and sold that as a non-fungible token, right. And to verify that you own the original, you know, drawing.

Chris Gammell: Probably wasn't a PDF though.

Dave Jones: Right.

Speaker ?: No.

Dave Jones: Well, no, it's, it's, it's, because that exists in paper form. Well, no, it was done. Oh no, that was drafted. Wasn't it? It was drafted. Yeah.

Chris Gammell: So like that photo you posted on Twitter this week of all those people on drafting tables.

Dave Jones: Oh yes. Yeah. Back when that was a day. Yeah. Well, that was the thing. Did, did you ever do, did you ever study technical drawing? Cause I, I never did this out. Right. Yeah. Yeah. My early engineering study was, yeah.

Chris Gammell: Right. I think I was just on the edge of it though.

Dave Jones: That, that doesn't surprise me.

Chris Gammell: I think like, um, I think some of the mechanical engineers that I went to school with. Yeah. Yeah. They still had like, they had a very traditional class. They were still doing like manual machining and stuff like that. So they weren't doing like CNC, but it was all about like measurements and, you know, drawings and, and stuff like that. And I think they made a hammer at the end of it. So.

Dave Jones: Right. Yeah. I, yeah, I can remember doing the class and you had to learn technical drawing. You had to learn the 3d projections and, you know, the orthogonal projections and, you know, things like that. And I still actually remember all that stuff. And, you know.

Chris Gammell: Does it have any relevance for you today?

Dave Jones: Well, people have commented on my whiteboards, how all that looks kind of like, you know, your angles are correct for doing your, you know, when I try and draw a three dimensional object on a, on the two dimensional whiteboard, people go, I've had comments that, Hey, that, that looks pretty good. David, the lines are a bit wonky, but it, but the angles sort of, you know, make it look, make it look good. And that comes from, you know, my, I'm not a, not an artist, but.

Chris Gammell: Okay. So take it, take a drafting, take a drafting class. And then many years later, someone might be like, Hey, Chris, nice whiteboard. Yeah. Okay. All right.

Dave Jones: Literally 30 years later.

Chris Gammell: Yeah. I still remember the basics of that, you know, I still. Worth it. I mean, does, does it have other benefits though? I mean, like, does it have like other benefits that you see or no?

Dave Jones: Oh, not really. Apart from getting concepts across as sketches and stuff. Like, you know, if I was doing anything serious, you do it in CAD now and CAD takes care of it all for you, you know?

Chris Gammell: Yeah.

Dave Jones: So it does the 3d modeling, you know, you do your two dimensional drawing and then it takes care of the three dimensional stuff and things like that. But, you know, it's learning how to dimension and do other things, you know? That's true. Yeah. Tolerances and things like that. We've mentioned, we've mentioned schematics, like, you know, package drawings, they screw up the dimensions and that they don't know how to dimension their package pin out their package dimensions properly in, in data sheets and stuff like that. I've had a few rants about that on people have on Twitter over the years.

Chris Gammell: Oh yeah. I mean, like, I feel like it's actually always drawn from the mechanical engineer's perspective and we want it from the CAD perspective.

Dave Jones: Yeah. We want it from the electronic. We want it from the usability.

Chris Gammell: Yeah. We want pitch and we want to know CAD width and we don't want to know about all these other crazy ass. And it's like, I'm never going to verify a SOIC.

Dave Jones: Yeah. And you don't want it to dimension all from one point either. Cause then you've got to add and subtract things and yeah.

Chris Gammell: I want it from the center. It's the centroid of the entire part too. You know, like all these things that are like, yeah, I'm sure we've talked about that on here before, but man, that's, that one stinks.

Dave Jones: Yeah. Anyway. Yeah. I think it's still a useful school, but yeah. Yeah. Leave it in comments down below, please. Do you, you know, if you're a recent graduate, I'm sure I'd be surprised if anyone's doing it. Surprised if there's a class.

Chris Gammell: I, it's interesting too. Sometimes like, like curricula, like kind of lags behind. Like I, I was actually, one of my buddies who's a classmate, he, he was asking me about like what we used in school and like, we were doing, so this was like 2002, 2003. And so we were learning on like Quartus two, which I think came out mid to late nineties, you know, like it was, but because the educational materials lagged behind, we were still doing like eight level design by drawing schematics instead of doing, we probably should have been doing Verilog. And you know, if anything, I learned VHDL on the job, I think from my coworkers. Yeah. Later when I started interning.

Dave Jones: So even into the late eighties, even in late eighties, early nineties, like we were still learning 6502 assembly language. And we had these antiquated in-house designed development boards, you know, and it was like, Oh God, no, come on. It's like, you know, we, we, we, we're still learning ancient micro processes when, you know, microcontrollers are the thing now. It's like, right.

Chris Gammell: Right. Yeah. I mean, some of it, I get it. Like it's tough to develop curriculum and stuff like that. And they don't want to update it all the time, but also, yeah, I want the newest things like, you know, like, and every time I bring this up to professors and I'm sure I've brought up on the show before, they're like, well, we can't keep up with all that stuff. And it's like, well, then what the hell am I paying you for? Right.

Dave Jones: And I like, I can remember, right. Well, this was technically, well, it was the late eighties. So we didn't have digital scopes like we did these days, you know, the little compact, you know, super powerful beast, but, but they did exist. And it was like, but we still had analog scopes, right. We weren't all on analog scopes, right. Labs were filled with analog scopes. And it was like the, the department, the engineering department had one digital scope, you know, when it came like, and it was, you know, I can remember when it came out and the lecturer had no idea how to use it. I had to show him how to use this digital scope. It's like, you know, he had no idea because like, it just got wheeled out. Like, you know, once in a blue moon, it was like the Holy Grail. You know, the, all the priests in white suits would, you know, stand around it and chant and, you know, oh boy. Yeah. Digital, but yeah, it's just when they were as common as mud, right. Well, they, you know, but not as powerful as they were now.

Chris Gammell: Yeah. But they were expensive.

Dave Jones: But they were expensive and they were, you know.

Chris Gammell: And there's no curriculum for it. I get it from the curriculum piece, but, but I feel like, you know, like some of the best things I really like when companies come in and start to like develop training and stuff like that too. Like that's another big investment, of course, but I know like TI does that. They have some educational training and things like that to try and like hook you young and get you using their parts and things like that. So that is, seems kind of symbiotic in that way, but it's, it requires a company of significant size in order to do that sort of thing. And then also depending where you're at a, you know, a program that they do that. But I just kind of, I would guess the thing I wish most would be like, just that the curricula would actually go out and point you at the things that you really need. There was actually just a tweet from one of the Google engineers was asking about like, why do students not ask for things that would make research faster or better, right? They're like less willing to, to spend money. And one thing that comes up for me is it was two things that came up for me. One is kind of just that general cheapskate mindset of being a student in the first place. But then the other is just not knowing about it. Like you don't know until you know, like I didn't know why. One scope is better than another scope because I was using it at the very beginning of my journey. And so it was just like, I don't, I don't know what all these other knobs do. I don't, I don't get it yet. And so I couldn't discern quality in the first place. And that kind of is what it comes down to for, unless you're like in the thick of it and like really struggling and being like, oh, how do I do this one thing? And then the other old engineer comes over and he turns the knob on the scope. He's like, oh, there, there it is. That's how you do it. You don't, you don't know you need that thing until you, until you struggle with it and Google around or ask the, you know, ask the experienced person or whatever. It's just really tough to get that, get your arms around that knowledge at first.

Dave Jones: Well, and it was harder back before the internet because, well, you didn't have the online community. You weren't exposed to an infinite number of things like you are now. It's like.

Chris Gammell: Yeah. Right, right. Yeah. Now you can Google for it. Yeah.

Dave Jones: Now you can go into your university lecture and you can go, well, why aren't we learning about, you know, ESP 32s and, and, and risk V and, you know, all this sort of stuff. And it's like, well, you, you, you just didn't know about this stuff pre-internet, you know, it's just.

Chris Gammell: Risk V.

Dave Jones: Risk V. Sorry. Risk V. Yeah. Risk V.

Chris Gammell: I feel like that's going to be a thing where I'm going to keep correcting until the end of the time. It's, it's cool though.

Dave Jones: I should know that. Cause like nobody ever says the, you know, the Fluke 87 V it's right. It's the Fluke 87 V, right? It's like, so yeah, I'll get that right eventually, but you can, you can probably see this not happen or happening a lot less with software that's automatically updated, like CAD software, for example, Altium in schools or something there you're not going to be likely using. You know, five or 10 year old copy of Altium. No, they're not locking it down to the old one. You're going to get the automatic. Protel 99. Yeah, exactly. You're going to keep getting the latest updates. Whereas people in industry, this is where the one place where students might actually be ahead of the industry. Cause the industry might be still stuck on Protel 99 SE. Right. And, and trust me, someone has worked it out to him. Like half the industry was stuck on, especially in China was stuck on 99 SE for 10 years. Like nobody updated cause it worked and it like, don't mess with my CAD tools. You know, they, they just work. So you're stuck with the old version. Whereas the, you know, you, you came out of your engineering course and you know, the latest version and you've got to go back to this old school one at some company because they don't want to upgrade because reasons, you know, they just don't want things to screw up.

Chris Gammell: Well, I mean, it's tough. Tool chains are just like, they get ingrained in companies and cultures and internal training and just, yeah, lots of things. I mean, think about all the, all the hooks from software.

Dave Jones: Oh yeah. We'd hear that all the time from the sale, Altium sales guys. We can't sell them the latest version. They're just so stuck on 99 SE and they won't budge the bastards. Even if we, even if we give it to them for free, they won't take it. You know, it's like. Yeah. That sounds right.

Chris Gammell: Yeah.

Dave Jones: Took a long time. So. I believe it. Yeah. Yeah.

Chris Gammell: You have to wait until, uh, until the machines give out and they can't install it anymore.

Dave Jones: Behind the times. Anyway, what else we got?

Chris Gammell: Oh, uh, you know, I, so I've been working with a mechanical engineer who I really like and he has been printing in resin and I am just amazed at the, like the, did you ever get a resin printer or no?

Dave Jones: I do have a, I technically, yes, I do have one. I have a, what are they called? The, the, the, the, the name escapes me now. The laser, you know. SLS.

Chris Gammell: SLS. Or SLA. SLA. SLA printer. Yes. Yeah. Yeah. Yeah. The SLS is the powder. SLA is the goop.

Dave Jones: Yeah. And then you've got to clean them with isopropyl. This was, I got it back when you couldn't get isopropyl because of the COVID thing. Oh yeah. This thing turned up and I didn't have any, I didn't have enough isopropyl to spare to, you know. Yeah. Yeah.

Chris Gammell: So technically I do have one, but I mean, I'm in, I'm jealous, I think just of the iteration kind of stuff, you know, that still doesn't exist for electronics in any appreciable way, but just like the quality of the prints that he can iterate on are just unbelievable. Like just the, yeah. Being able to do that over and over again, you know, he'll have three sets of prints in, you know, two days. Oh yeah. Right. And, and like be able to iterate like that. And I just, you know, I, I, sometimes I can't get parts in three days or at all really.

Dave Jones: I still remember being blown away. You can see the video. It's, we might have to link it in of where I opened the package of my first custom 121 GW multimeter. Right. And I didn't know what they were sending me. And they sent me this clear resin printed prototype thing. And it was like, wow, you know, in the case, it was two halves of the case and had the knob and it had like, you know, it was just absolutely incredible just to get a look and feel for the multimeter was just, it was absolutely amazing.

Chris Gammell: Yeah. And there's something about that. I mean, like you could look at a 3d prints all day, but man, holding in your hands. Yeah.

Dave Jones: Holding in your hand. Yeah. It's so very different.

Chris Gammell: Can't get over that. Yeah. Yeah.

Dave Jones: I was just blown away. So, yeah.

Chris Gammell: It's too bad. I mean, the, I don't know, like the boards are cheap and fast. And I think, I think the real thing is too, that like, you know, you've been able, people are going to say, oh, you've been able to get SLA prints forever, you know, next day, whatever fast. But like having that in-house is different. I think, you know, being able to just run it, run it, run it. And like not having the fewer like friction layers you have, the better you're going to better off. You're going to be in terms of iteration cycles. And it's like, you know, you and I have talked about picking places on here, of course, and board milling machines and just all these different things. And I guess I'm just, I'm just real jealous, you know, I just want, I just want a six layer board printer at home, you know, that like is like reliable and always works. Yeah.

Dave Jones: Yeah.

Chris Gammell: It just, yeah, it doesn't, doesn't go. Three, one.

Dave Jones: You know what this segues into? Oh, oh, DigiKey. DigiKey are now printing. Oh, this is the best thing since sliced bread. They're now printing the DigiKey part number and the values and stuff, the details on the back of the cut tape.

Chris Gammell: Yeah. Oh, this is. I got to see this early. Ben, Ben from DigiKey showed me this early and, or pulled me in to show it early with some of the other engineers there. And it's super cool. I, the first thing I thought of, what's the first thing you think about doing with this?

Dave Jones: I've often done this a lot. I would get out my permanent marker and I'd write on the back of the cut tape. You know, I physically had to do this. Now it's like, but I, so what's your question?

Chris Gammell: What I think my thing is the thing that I said is like, oh, I get so upset whenever I go to buy like a parts kit, you know, like a different values. Yep. And you know, they're just so expensive for what they are because someone somewhere is doing cut tape.

Dave Jones: Somebody has to manually cut them and label them and so on. And I've, I've, I've had those ones. I've bought those kits, those resistor kits, and they actually mark the value on the back of the tape. So exactly.

Chris Gammell: But that's what you were paying for. And then you were paying above and beyond, but now you can make a completely custom of your own. All you need is this little box. You go and print your, you know, you basically just order a bunch of values and then you just throw them in a box now. Right. It's, it is, they have in one fell swoop, they have obviated the buy, you know, get a resistor kit, get a whatever kit. Yes.

Dave Jones: You can buy your own resistor kit and it's done for you. Yeah. Yeah.

Chris Gammell: Yeah. I mean, yeah, you've always been able to do this and usually they're like vendor kits, but like, if you go on, if you go on a DigiKey or, or Mouser or whatever, and like you go to actually buy one of the kits from like a, you know, an AVX or a, you know, Kyocera or something, you know, whatever kits you want to get, they're pricey, you know, like they are.

Dave Jones: Oh yeah.

Chris Gammell: Cause really they're promotional things usually.

Dave Jones: No, you have to get them from the Wanhang Lo factory in China off AliExpress if you want cheap, you know.

Chris Gammell: Right. If you want, yeah. But I want cheap and quality.

Dave Jones: And you want, you want no one manufacturer. You want to know it's that capacitor. You don't want to know it's just some random, you know.

Chris Gammell: Right. Especially if I'm like going, like getting like in 0.1 picofarad increments or something like that in an 0.2.0.1 is like, okay, well, I mean, I could get those from wherever, but I think by the time they're off cuts from some lot, it's like, yeah, I guess I'm not going to go and measure 0.1 picofarad to verify every part. I want to just know that I have the actual part number. So anyways, that's what I thought of immediately. So make, make your own kits at home. That's kind of nice. I think, I think Ben said eventually they're going to actually mark, maybe they already do on the site when they are, when the labeling is available. Cause I think that would be really necessary so that you're not just like ordering random parts and being like, Oh, yep. None of these are labeled.

Dave Jones: That's a long standing thing that I've been meaning to do for years. I've been meaning to actually collate. We've got already actually got a spreadsheet of all these types of stuff. I've even got a sort of a kit. Anyway, I was thinking about selling a folder with like a starter pack with all the devices you know, like jelly bean parts, you know, it'd be a jelly bean starter kit. You know, it wouldn't include that passives.

Chris Gammell: Yeah. Cause it goes with your sleeves, the sleeves you sell, right? I mean like that's something that you already do.

Dave Jones: Yeah. So I was going to put them in the micro sleeves and yeah, I still sell the micro sleeve. I still got warehouse full of them. Micro sleeve. Yeah. It's great.

Chris Gammell: I should send you some. I know. It's a great product idea. I like it. It's just the name, Dave. I'm just like, what? I get it. What's wearing a micro sleeve? I know. It's just, it's your brand. It's fine. It's just.

Dave Jones: Okay. Oh boy.

Chris Gammell: You want a DMM? Micro DMM. You want a, you want a current measurement? Micro current. You want a supply? Micro supply.

Dave Jones: And don't forget the micro calc and the micro watch.

Chris Gammell: That's right. Of course. Yeah. Right. Right.

Dave Jones: So, you know, yeah, I used to sell my micro watch. You know, that was a kit. Used to be able to buy. People still ask for it. You know, I've got the webpage somewhere. I still. Oh yeah. I'm sure you do. Every couple of months I get an email saying, Hey, can I buy one of your micro watches? That's fine. No dude. I discontinued that like 10 years ago. Stupid ass me still has the webpage up, you know? Yep. And even web pages, like these old web pages, even if you put the information there, I do not. This is all the information I have. Please don't email. Don't ask. You know, I've got it on the webpage that says all this. They email you anyway saying, Hey, you know, do you have this? And, ah, boy. Yep. Yep. Anyway, just the other day I had to send somebody an old pic binary file for one of my video overlay projects that I've done a video on. It's like, he said, Oh, have you got the binary for that? I just need the binary. It's like, okay. I dug it out of the archives and sent it to him. Yep. Anyway. Ah, boy. Guess what's back.

Chris Gammell: What's back, Dave?

Dave Jones: Well, this is, this is today's video that I'm working on. The, uh, thermoelectric generators are back. Thermoelectric generators and the impracticality of them. It's like, yeah, there's this paper that, oh, here, I'll send you the link to the paper.

Chris Gammell: I mean, so like, well, there's the RTGs, right? Which is like what's on the Mars Rover and Mark Watney used in the Martian. Yep. To keep yourself warm. Most things are not operating from a radioactive level. So power levels are pretty important.

Dave Jones: There you go. I'll send you the link to the paper.

Chris Gammell: Is it a Delta T that determines the power output? Like what is the actual like linear element there?

Dave Jones: Yes. The maximum PowerPoint. Well, yeah. If you look at the maximum PowerPoint curves, I can actually send you the direct image to copy image address. There you go. I'll just send this to you. Makes for great radio. Yeah. Yeah. Yeah. We're just getting the same page here, folks.

Chris Gammell: Literally the same page. Yeah.

Dave Jones: There you go. Here's it. You can see the maximum PowerPoint curves and they're just like a bell shape. Oh, they're like, oh, sure. Just a typical curve. Right. And yes, you can get one volt out of these things. And if you read the paper, they made like the world's biggest, you know, the world's highest output voltage thermoelectric generator. This is actually really quite cool. It actually goes on your skin. It's actually flexible and it's actually a self-healing polymer. So when it bends and flexes on your skin, right? It doesn't actually break. It doesn't, you know, break down. Like it doesn't actually destroy itself. Right. Just through bending and stretching and stuff. So it's kind of like a really cool innovation. And, but they've, like the abstract says the world's best, highest output voltage thermoelectric generator. And then you, even in the air, you don't even have to go past the abstract to read. Oh, by the way, that's at no load. Right. One volt at no load. And at a 93 degree temperature differential. Right. Oh my gosh. Yeah, exactly.

Chris Gammell: So it's like you have ice on one end and you got a fever on the other.

Dave Jones: You're basically dead. You're, you know, so this will only work if you're standing on Mars kind of thing, you know, you'll, and then you'll only get 10 microwatts out of it. Right. And, and of course, 10 microwatts.

Chris Gammell: Well, it's microwatts on the Y axis.

Dave Jones: Microwatts, right. So, well, no, you haven't heard the best of it yet. So, so even at, so even under ideal conditions, right. 93 degree temperature differential, it'll give out 10 microwatts. No, 18 microwatts, right. Is the absolute maximum, but it'll only give you half a volt because that's where the maximum power point is on the curve. So you have to know, you know, it's the top of the curve is your maximum power points. I only give half a volt. Okay. So what do you get actually in a realistic scenario where it's on your skin at, you know, 37, 35 degrees or whatever, and then ambient room temperature where you get a hundred nanowatts. It's like, yeah, it's, so it's dropped from this magnificent 18 microwatts, which can barely power a Casio watch.

Chris Gammell: Maybe this is why people are asking for the micro watch. They're like, they're like really low power, right, Dave?

Dave Jones: Yep. Yeah. And, and you can't, and they're saying, oh, it'll power the internet of things and wearables and everything. No, no. The, the only wearable you can power, well, you can't, I'm, I'm calling it now. You can't power any wearable with this in a practical scenario.

Chris Gammell: I mean, maybe, maybe like a custom, custom ASIC, you know, like something. No, no, no. It's not even going to do it.

Dave Jones: You'd have to be operating at a hundred nanowatts.

Chris Gammell: I know.

Dave Jones: What are you going to do in a hundred nanowatts?

Chris Gammell: Well, you know, if you like, if you really accumulate the power over time, you build it up in a battery.

Dave Jones: Oh, but you got to build it up. No, no. That's just, you know, you got to energy harvest. You'd spend more.

Chris Gammell: I mean, that's how most of those things work though, too. You know, like when you see like the RF harvesting, things like that, they're started storing a charge and then they're just doing a single operation.

Dave Jones: Yeah, they are. But you'd spend, you'd, when, when you're DC to DC converter and your energy harvester takes more power than your actual device does, you're, you know, you're kind of screwed. Oh yeah. Yeah.

Chris Gammell: I think if you're starting from this point, you're kind of screwed. This is research, right? I mean, this isn't like.

Dave Jones: Well, yeah, it's research. But the thing is like, and, and one of my, uh, one of my favorite, uh, YouTube channels, they actually read this paper and they got all excited about it and everything.

Chris Gammell: Oh, is this the two minute, two minute papers?

Dave Jones: No, this is, um, Anton Petrov.

Chris Gammell: Oh, okay. Yep. Anton's good too. Yep.

Dave Jones: Hot to math channel. Anyway, my, my video is kind of like a response to him and he got all excited about it. You know, fair enough, because you know, if you read the abstract, it sounds fantastic. But when you actually go in and look at the actual data and do the back of the envelope calculations, you realize, nah, you can't power anything with this thing. And they, well, you know, anyway, it's really cool research, right?

Chris Gammell: But we would love to see if people do, I mean like, hell, go for it, man.

Dave Jones: Okay. If you can develop a, something that does a useful thing on a hundred nanowatts, go for it. Exactly. Yeah, sure. Right. And then you can put it on your skin. Yep. And then it doesn't work on a warm day, you know? So it only works if you're room temperature or under, or if you're in a cold climate.

Chris Gammell: This is a Chicago rated device in, from January to early March.

Dave Jones: That'll be on the spec sheet, will it? Yeah. Chicago rated. All right.

Chris Gammell: Chicago or above. Right.

Dave Jones: Right. Vancouver rated. Yeah. Right. Right. Yeah. Oh boy. Yeah. Anyway. Yeah. I just like, I, so I had to do a video on it because people just like, you know, read the top, the banner spec and they make all sorts of claims and you can't blame the researchers for getting all excited. Right. You know?

Chris Gammell: I mean, so in, in the, in the, uh, the realm of getting excited about specs and, uh, you know, and MPPT, I'm actually excited about this new part. I just sent you here, the BQ 25, seven, nine, eight. I'm using the BQ 25, eight, nine, five right now on the ABC board. And so, you know, it could take solar input could do this, you know, it's doing all this other is fast charge and a lot of that stuff. Uh, this one's kind of interesting cause it's basically, it can do higher charge output and higher voltage input. And so this is basically now like before I could only do a six volt cell because, you know, I, I saw the spec of like, Oh, 14 volt input on a BQ 25, eight, nine, five. Right. And I was like, I could do a 12 volt cell. It's like, eh, well, if you get it biased wrong, you basically, you could have more than 14 volts in the input and that would be bad times for all. Uh, so.

Dave Jones: So it goes down to three volts minimum. What's the startup? Does it still start up at three? Because that's the thing with, uh, that's the, that's the trap banner specs with these energy harvesting things. You got it totally right. Yeah. Banner specs will say, Oh, it works down to 20. Like some of them work down to like 20 millivolts, 20 or 30 millivolts. Yeah. Like energy harvesting. Energy harvesting chips. Right. Insanely low voltages. But the, but the trick, the trap for young players is that they don't start up at that voltage. You've got to have like, you know, 200, 300 millivolts. It's like, it's like a nurse shop. Right. Yeah. Yeah. Once it starts up. Yes. Then it's all running. It can go, it can drop back to 30 millivolts, but it can't, you know, yeah, it can't start up at that.

Chris Gammell: You need that, that's the charge you stored in the, in the super cap next door to get it to wake back up and do its thing. Yeah.

Dave Jones: Yeah. Don't get all excited, but yeah, cool. This is a little I squared C interface jobby.

Chris Gammell: And I like this family of chips too. I mean, I, I, I first found these cause I was, like I said, this is what I was doing for the, uh, for the ABC board, but I was, I like this because, you know, like having to switch in without having to think about the battery and it's just kind of doing power path stuff. That's cool. But this one actually now can do higher loads. If you're actually trying to do like multiple stacks of batteries, if you think about like having like a practical battery stack, like when you're like out in the, out in the real world. Yeah. Yeah. You know, having a single cell lithium ion is possible, of course, but this kind of thing is like, no, now you can actually start to get some, some more serious, uh, power in your battery pack by stacking cells and things like that. So. Right.

Dave Jones: And it's, it's actually got a, a tap point. Doesn't, no, no, no, it doesn't. You just put them in series. So it's not, it doesn't tap off each cell.

Chris Gammell: Yeah. Yeah. It doesn't do like, like individual monitoring or anything like that. So, um, yeah, I think.

Dave Jones: And it's fully programmable too. Is it? And is the output voltage fully programmable too? It says 2.5 volts to 19 volts. Is that like a fully programmable output power supply?

Chris Gammell: No, you'd have to follow this on with a, with a regulator as well. Oh, okay. So you'd have to put like a, a downstream buck if you're handling. Oh, got it. Okay. Right.

Dave Jones: So this isn't, doesn't have a built-in voltage regulator.

Chris Gammell: That's right. Yeah. Okay. Right. Yeah. So this is mostly a charge tip and then you'd put this out front of a, you know, this is basically your incoming, your power handling, that kind of stuff. Yep. Gotcha. But, uh, but, but because you can run off a, uh, panel, you know, you get a pretty wide range of panels. That's where it gets pretty cool.

Dave Jones: So if you're doing up to five amps too, that's pretty grunty.

Chris Gammell: Oh yeah. Yeah. Yeah. Yeah. I mean, cause all these things are like, like all these parts are doing.

Dave Jones: Oh, yeah. So do a hundred Watts.

Chris Gammell: Yeah. I mean, that's, so it's USB C, uh, spec. Right. I mean, that's really, usually when you see the five amp thing, that's usually what they're talking about is that you can switch in a, uh, you know, you can do PV panels, but then you can also switch in like a USB C and handle that kind of current. If, if you have a PD chip in front of it. So yeah, it's cool. Yeah.

Dave Jones: I like it. Yep. Is it, is it a newbie?

Chris Gammell: It's brand new. So I just got a eval board there. They sent me one. Right. And it's eval boards are just nuts these days. You know, just like the number of.

Dave Jones: Hang on. Hang on. I'm looking down the bottom of the page at the technical documentation. The first document is June, 2014. And the title, I kid you not. The title is summertime and the living is easy.

Chris Gammell: Really? I didn't see that.

Dave Jones: What? Yes. I'm clicking on the link now. Summertime. Yeah. It's a, it's a blog. It's a TI blog. Summertime and the living is easy. Song lyrics, George Gershwin.

Chris Gammell: When I think there's just like a referral talking about solar, I'm guessing.

Dave Jones: It's just a blog article. Okay. So, all right. Oh God.

Chris Gammell: That's what they linked to, I guess.

Dave Jones: Come on. Summertime. That's in their technical documentation. Yeah. Come on. Really?

Chris Gammell: Yeah. I was actually opening. I had to, I have, I have a junk box and I have just some like raw solar panels in there. I'm just like, where did I get these? Yeah. And so I'm going to try and like just plug it into this thing and see how it goes. We'll see. I mean, I'm going to measure the output first.

Dave Jones: Well, I, I got, I found my whole box of flexible solar panels the other day. You know, those, um, those printed on polymer sheet ones. I've got a box full of them, you know? So.

Chris Gammell: Yeah. What's, what's the output for those?

Dave Jones: Oh, it's, it's not much. They're, they're like 2%. If you're, if you think 20% efficiency of regular solar panels, ain't that great. These are about 2%. Yeah. They're not, they're not great. But the thing is they're flexible. You can roll them up. You can, you know, they're, uh, yeah. Yeah. They're really quite cool, but yeah. The efficiency is not, I don't even know what the state of the art is. I don't even know if it's hit like 5% yet or something like that.

Chris Gammell: Yeah. But if you're camping, like it's better than, you know, not having LED lamp for at night. Yeah.

Dave Jones: Well, I've got one of those folding, you know, camp solar panels, you know, big hundred watt job. It's sort of like folds out into four big panels and it's got little stands on it that, you know? Yeah. Yeah. So pretty good. Anyway, speaking of, uh, designing, uh, design choices, it's kind of, I'm, I'm stretching the segue a bit, but, uh, yeah, I. I wanted to talk, uh, we, uh, we discussed this, that maybe we'd talk about when to make design decisions. We're jumping to really complex design now when to make the decision whether or not to use an off the shelf, uh, operating system and, and, you know, you and, uh, drivers, you know, like to drive USB stuff and things like that. And to do MPEG encoding and, you know, codecs and, you know.

Chris Gammell: Yeah. I think the conversation was about like when to write stuff from scratch. Like, so.

Dave Jones: When to write stuff from scratch. And can we say where this comes from? Yeah, we probably can.

Chris Gammell: I think we shouldn't make any promises in case it falls through, but like, yeah, there's. Yeah.

Dave Jones: No, no, no. It's anyway, there might be something coming up with this on the Amp Hour, but anyway. Yeah. Black magic, uh, design, you know, I've been playing with their ATM switcher and you've been lusting over one of those.

Chris Gammell: I don't have any place to put it. So that helps a lot.

Dave Jones: Right. Oh, okay.

Speaker ?: All right.

Dave Jones: So, uh, yeah, I found out that they, uh, it's probably no secret that they, they actually write all their own stuff from scratch. They don't like, you know, I've done a teardown of this thing and it uses like a state that, you know, a $500 zinc processor chip with five cores on it. You think, oh yeah, they're running Linux. And you know, cause why, why would you, you know, why would you do it all from scratch? Wait, wait, wait. Actually, I didn't get this from our conversation.

Chris Gammell: So you're saying they wrote the operating system from scratch too? Oh yeah.

Dave Jones: They wrote the operating system from scratch apparently. Right. They wrote the operating, they've got their own operating system. They wrote their own MPEG codex. They wrote their own USB drivers, everything.

Chris Gammell: Wait, their, their own distribution or their own. You're going to have to clarify later, but.

Dave Jones: I'm, I will, this is what, this is what I've heard is that everything is written from scratch.

Chris Gammell: I could see. So here's where I draw the line personally. Like, yeah, okay. You're going to write your own drivers. That makes sense. You, you almost have to, you probably have to, unless you're working with a third party that already has pre-written drivers for something. That's fine. Encoders. I don't really know how to write an encoder. So I'm sure that there's, that's probably something that's in the software realm. And I sure, I'm sure that like, that makes sense to have, you know, some kind of IP there that you're not just like buying off the shelf from someone else, but an OS, like a full OS.

Dave Jones: Because I think a lot of these you have to license, you know, a lot of these, like you can't just like, yeah, you've got to actually license a lot of this stuff.

Chris Gammell: Well, and there's two levels too, right? There's like the, the software piece, but then if there's like hardware acceleration in the FPGA, you might buy an IP block. And right. So like, that's, that's something that Sammy was talking about a little bit last week on our show with Sammy from FNX and then Brian from QuickLogic also talked about that, you know, like that's just a big piece of being in the FPGA space of like, you can kind of buy your way towards pre-qualified IP. And it makes sense for some things, but, but then you are using the same, you know, you have the same specs. Oh yeah.

Dave Jones: We, we, we, we did this. We were working on a hugely like the five kilometer long underwater streamer and you know, it's got thousands of data acquisition, uh, you know, hydrophone sensors and everything. And it's all fiber optic connected. And we had to send stuff from the very end, five kilometers down all the way back to the, uh, the, uh, data recorder at the front end. So we have to packetize everything, you know, everything's, and it's got to be real time as well. So we, I can't remember the protocol. What was it? T T five or something. I, uh, there, there's like a serial, there's, you know, serial comms protocols. And we thought, oh, should we.

Chris Gammell: Real fast, real fast.

Dave Jones: Yeah. And, and it's all real time as well. So it does it, you know, so there's no late or there's deterministic latency and all that sort of jazz. And we thought, oh, should we write this ourselves? And then we went, no, let's just, no, no, let's just not dig around. We don't have the time or the resources. Let's just buy an off the shelf. The HDL call that just plugs in and does it. And it turns out that that kind of sort of worked as, as it turns out, we contacted, uh, a company that claimed to have had one of these calls that sounded like it did exactly what we want. Right. And, um, it was like, you know, $150,000 or $100,000 or something, but that's cheaper than actually developing yourself from scratch. Right. In man hours. Yeah.

Chris Gammell: I was going to say, you always, you always couched against like how many engineer hours it would be otherwise. Right. Yeah.

Dave Jones: Yeah, exactly. And we did the calculations and everyone was saying, yeah, all our VHDL guys, we go, yeah, no, let's just pay the money. Let's pay the 50 grand or whatever. And, and, and just get this call. And sure enough, like we, you know, paid our money and please, can we have our call please? And they went, oh yeah, we'll get working on that. And like, they hadn't written it. They had just.

Chris Gammell: Speck. Just a speck.

Dave Jones: They had to write it. Yeah. Yep. Just a speck. All they had was a speck and, and they went, shit, somebody bought it. It's like, it's like the alarm goes off. We got one. We got one. And they had to start working on bloody writing the thing. Unbelievable.

Chris Gammell: That's bad. That's bad. Yeah. So I think the thing is like, I think there is some benefit, you know, like in a, in a case like that, where you're doing a super specialized project or product rather, and you already have a competitive advantage in that you've, you've won the bid or you're an expert in the space, whatever, like, yeah, buying someone else's IP is not a big deal because you don't need to differentiate based on that IP in the case of like having the fastest video compression. Like there's tons of people doing video compression. So I could, I could imagine that there's reason to, to try and optimize that and try and do that for your spec hardware or whatever.

Dave Jones: But at what point in your development cycle, do you decide we're going to write all this from scratch? It's where we're going to take this R and D hit now, but trust me in 10, you know, if you're the CEO, trust me in 10 years time, this is going to pay off big time.

Chris Gammell: I would say for me, it would be rev two or three.

Dave Jones: No, right. Okay.

Chris Gammell: Right. Cause you still need to like prove that you can do the, everything else. I can, I can see like getting super bogged down with it. Well, we should, you know, we got a soup to nuts. We got to own everything here. It's like, okay, but like, let's get something in the marketplace, you know, like, right. Like duct tape and bubble gum, fine, whatever, but get a prototype out there. Get, you know, get, get to rev one without it.

Dave Jones: Oh, so you're talking about like further revisions of the product, get a product out there that runs Linux and all drivers. And then secretly in the background, you have a little skunk works team working for a couple of years and then magically outcomes all your own custom thing.

Chris Gammell: Right. And then you have something to like benchmark against and improve against. And you have like, yeah, right. You know, and I think most importantly is like having something that customers are already using and telling you about now, granted that doesn't always work because you know, people sometimes want to hold close stuff closer to the vest. That's fine. But like, even like internal users or people who are under NDA, just like anything that gives you some semblance of like being able to test it in the real world. It's, I think that's really important. And I think a lot of engineering organizations wait way too long on it. They're like, oh no, no, it's coming. It's coming. And then, you know, three years later, you're like, oh, cool. Well, thanks for telling me about that three years ago. What would you do? Oh boy.

Dave Jones: Please. If there's anyone out there has done this, like, or your company's decided to do this, you know, I'm going to write.

Chris Gammell: Well, will they tell us though? That's the real question.

Dave Jones: Well, no, maybe not. It's, you know, but it's not hard to sort of, you know, go and extract the, you know, log in via the debug kernel or whatever. And, you know, I try and extract these. You can see what stuff they're running. You know, it's not hard to sort of get in there and all they're not running Linux. They must be doing something, you know, secret squirrel.

Chris Gammell: You're saying it's not hard. Would you be able to do that? Because I wouldn't know how the hell to do that.

Dave Jones: Oh, no, I could go in there and, you know, I could look at what, you know, because there's like a boot code, you know, thing. There's usually like a UART output or something like that. And if it doesn't have all the requisite stuff that you'd normally see in a familiar OS.

Chris Gammell: Dave's putting on the hacker hoodie here, folks. That's what I'm hearing.

Dave Jones: Right. It's pretty basic stuff, hooking up a UART to a couple of pins. You know, it's not rocket science, right?

Chris Gammell: That's the thing. I never reverse engineer anything. Like, I just don't. I mean, like, I know that it's like a skill and I love watching people do it, but man, I just don't.

Dave Jones: Yeah, I don't do it. I just have a peek to, you know, just for shits and giggles.

Chris Gammell: You know, people like look at like firmware decompilation and stuff like that. I'm just like, wow, what am I looking at? It's just a pile of gobbled book.

Dave Jones: Yeah, yeah, yeah. Yeah, it's amazing. Yeah. I've made hats off to people who reverse engineer stuff from scratch. It's just crazy.

Chris Gammell: I just, the not doing Linux or not doing like some like office, shelf OS that I don't, I don't.

Dave Jones: Yeah, I know it. It doesn't.

Chris Gammell: But again, like, I think some of this is like, so some of it could just be like engineering culture internally, right? It's like, well, no, we have to do custom because we need to be super secretive or whatever. Okay, fine. But like, I feel like so many times it starts as, well, Marty started writing this thing on his own and then he kept going with it. And now we're, you know, we're deep in it and we just got to keep going. And then it's like, you just end up building the same thing over and over again.

Dave Jones: Oh, yeah. I can remember. I worked at a company in the late eighties who did security stuff, right? And they would develop, like, it was all Z80 based, but then they decided to, they were working on this new high end, you know, fancy pantsy product. And they were like developing their own, like obscure graphical operating system or something. I'm going like, what, like, why the heck are you doing this? And they were like designing their own serial cards and stuff to plug into, like, I'm just going, no, there's all these industrial, just buy this industrial PC and you can get like eight channel, 16 channel UART cards off the shelf that you can just plug in. Why are you designing all this? And they, you know, I told them about at the time, there were no Google links. So I actually bought in magazines and I physically bought it, you know, electronics design magazine, you know, the U S based one. I used to get that actually delivered here. And it was a big thing. Cause you couldn't, you know, you couldn't buy, you couldn't get it a news agent here. So I get, you know, subscribe to that. And I'd bring that in and show them ads for these boards. And they went, Oh, you can buy those. Can you? And I went, yeah, you've been able to do it for a long time. And they went, Oh, you know, but they were too far down. Like they were too far in to change it all. So they just went with their custom.

Chris Gammell: Sunk cost fallacy. I believe it's called by the economists. Oh yes.

Dave Jones: Yeah. Yeah.

Chris Gammell: Although most people usually skip the fallacy part. They're like, yeah, sunk cost, man. Yeah. Sunk cost.

Dave Jones: Sunk cost. No. Oh, but they, but, but they were there. They sort of had like a working product, almost ready to ship. It was like, you, you wouldn't. I don't know.

Chris Gammell: I don't know. Yeah. I think it's a tough thing. Like in any engineering organization, when do you decide to, to turn that crank and like, and switch to that new processor or whatever. Right. You know, like people talking about like, Oh, we could never switch our processor. And like, some of them are right. They're like, we don't have the code. We, we don't even know how, we don't even have the source code to look at in order to, to do a functional equivalent, you know, it's just like, Oh, okay. Well, yeah.

Dave Jones: Well, Kane slash UAI did that with the one 21 multimeter. I actually had, they went through like three, three prototypes. I've got three prototypes that use an old pick 32, uh, no pick a 24 F chip. Right. I had to choose. Uh, they, they originally thought they couldn't find one with LCD driver, but I found one for them. Look, here's the, here's the pick 24 F. And they went, yep. We're, we're all familiar with the pick and everything. And so we, you know, the prototype was practically done. They were working on it. Then all of a sudden I got an email saying, sorry, company's switching to arm. We're switching to arm. We're going to scrap, scrap the pick. We're going to rewrite this from scratch. And like, I was like, I've been working on this for nine months. It's like, you know, and they, sure enough, they just turned it around. Like three weeks later, I, I, three weeks later, I had another prototype in my hands with an ST 32 F something or other in it. And sort of like kind of sort of worky firmware.

Dave Jones: That's great.

Chris Gammell: It was amazing. Might also speak to the, uh, the complexity of the firmware on there too.

Dave Jones: No, well, yeah. Well, like they didn't have to write it from scratch. A lot of the C code they could port, of course, you know, but it's like, it's still.

Chris Gammell: That's still, yeah. That's non-trivial for sure. Yeah. Yeah.

Dave Jones: That's that, that is non-trivial. I was just stunned when they did that.

Chris Gammell: That's great. And I think that's, I mean, like, and that kind of thing is a good move, right? Like making that shift over.

Dave Jones: It's like, oh, of course I'm sure it paid off for them. Yeah. Yep. No, they, they just went, oh, look, we can't do like a lot of the stuff that we want to do with like potentially updating firmware over Bluetooth and doing other, you know, and SD card stuff and, and stuff we want to end projects. We want to do in the future, other products. Let's just move to arm now and take the hit and. Yeah. Yeah.

Chris Gammell: And that's, and that's, and that is usually like an overhead type of thing too. If you think about if you're an organization, it's like, yeah, you can't, I mean, that, that's not money. You're, you know, your, your client's not going to pay more for that. The feature of being on arm now. Right.

Dave Jones: Oh no, no, no, no, no. You, you have to sink that cost. You, you have to eat it. True development cost. So. Ah, the old pivot. Some companies have not pivot and died.

Chris Gammell: Oh yeah. Oh. You know. You beam. Different reasons, Dave. Different reasons.

Dave Jones: You know. Yeah.

Chris Gammell: As of a couple of days ago, you could go and you could put an entirely new OS on your device. Have you seen this news that came out?

Dave Jones: No.

Chris Gammell: Yeah. They just released. Tell me. An alternate. So the, the creators of Unix, right? The forebear of Linux and all of its variants. There was another OS that came out of Bell Labs called Plan 9 from Bell Labs.

Dave Jones: Sounds like Plan 9 from Bell Labs sounds like a, uh, thriller, like a spy thriller.

Chris Gammell: Uh, Plan 9 from Outer Space is actually what it's named after. Have you ever seen that movie?

Dave Jones: Oh, okay. That's where, that's, ah, right. That's in the back of my mind where I got that from. Right. Okay. Uh, I forget Wood.

Chris Gammell: Something Wood is the guy who made it. It is listed consistently as the worst movie ever, but, uh.

Dave Jones: Right. Really?

Chris Gammell: I used to do, uh, bad movie nights. My, my buddy's father used to collect like bad movies and like bad, like horror movies. And so we'd go over there and watch them.

Chris Gammell: Yeah. I have seen some doozies. Let me tell you. Uh, Plan 9 is very bad as well.

Dave Jones: Hang on. Haven't you taken me to task for watching B-grade crap on Amazon? Uh, I do believe you have.

Chris Gammell: This is Z-grade, Dave. I mean, this is, this is like specifically because it's bad. It's not because I'm bored and like, I'm trying to find something that's sort of.

Dave Jones: So, so this is below Amazon women from Mars. Oh, this is, you know.

Chris Gammell: Well, no, this is actually, I think this is the Amazon women from Mars kind of thing. I mean, like, yeah, this is, uh, rats was one movie that I watched that was terrible. There's another one called just space invaders. It was very generically a little plot spoiler here. The aliens killed people, not by injecting them with poison, but by their fingers had ethanol coming out of them. And that's how they killed the town drunk because he was already drunk and the ethanol coming out of their fingers, of course, uh, killed the drunk because he had alcohol poisoning. Then it was, oh, it was, it was formative. I mean, like all of my bad jokes come from a place and it's those movies. So got it. Yeah.

Dave Jones: Okay.

Chris Gammell: So where does this, so anyways, Plan 9 from outer space, very bad movie. That's also what this operating system was named after. Uh, it was never released. So we're talking Dennis Ritchie.

Chris Gammell: Yeah. Yeah.

Dave Jones: I mean, holy crap.

Chris Gammell: Yeah. And so they just put it on the internet. I don't think it's going to go anywhere, but it's really interesting that they just basically released it. I think it's a distributed operating system as well. It's not like.

Dave Jones: Have, have they upgraded it or has they been working on it or is it like the original stuff from the seventies or whatever or eighties?

Chris Gammell: I think it's probably like a virtual machine that's like living online. Right.

Dave Jones: Okay. Yeah.

Chris Gammell: The code or actually, sorry, it's just the code that's online. Maybe there's a way to use it. You can download it. I don't know. I'm just looking at this, but I will say that their, their little logo bunny is, uh, adorable. So, so that's fun.

Dave Jones: Yeah. Yeah. It's pretty cute. Little fat bunny. Yeah.

Chris Gammell: Anyways. Uh, yeah. So another, another, uh, operating system and like so late too, you know, it's just really interesting that it's like this kind of like, I think it's going to be like a pet project for a long time.

Dave Jones: They're just doing it for shits and giggles. I don't think they seriously expect anyone to take it up. They're just, you know, yeah.

Chris Gammell: But, uh, yeah, pretty, pretty fun. So you can put that on your, put that on your device if you really want to, you know, tell, tell black magic. They can, they can, uh, go and go and turn on a dime and put this on their device now with no hardware support.

Dave Jones: Better yet, put it on your resume. Yeah. Plan nine. Right. From Bell Labs. Oh yeah. Fluent in plan nine.

Chris Gammell: Oh, oh, oh, oh. Sorry. I missed this. You can get a bootable plan nine on Raspberry Pi. There you go.

Dave Jones: Oh, there you go. There you go. That's, that's modern. Have to try that one out. Yep.

Chris Gammell: Okay.

Dave Jones: Yep. Although no, that's out of date. It isn't no Raspberry Pi's out now. It has to be risk five, right?

Chris Gammell: Maybe. Yeah.

Dave Jones: Yeah. Well done. All right. We've got five minutes left. No one.

Chris Gammell: No, we've got 10 minutes left. 10 minutes left.

Dave Jones: What do we got? Jeez, we're racing through stuff today.

Chris Gammell: We are. We are. This was kind of interesting. There was a, so someone over on Hacker News had posted about their self-study plan. I guess we talked a bunch about student stuff today, but I thought this was interesting. So he's trying to get into like brain computer interface and stuff like that and wants to get into hardware side of things. And so has been posting about like, want us to have some projects, but then also like just the list of the books that they plan to like work through. And I kind of balked at it a little bit just because it's like so book heavy and the projects are like, oh, well, I'll figure out what projects are here. And I think, I think the main thing.

Dave Jones: No, I was just going to say like, it starts out. The first book is intro to electricity and magnetism. Yeah. I mean, that's like, oh, it gives me the heebie-jeebies, right?

Chris Gammell: Yeah. It's a university book. It's like a physics textbook, right?

Dave Jones: Yep.

Chris Gammell: And I think it's just, it's a, but it's a good, this is a great post. I'm glad this person posted it. I mean, honestly, it's, oh yeah. It's just, it's so tough in this scenario of like, do you dive into the books like this or do you have this other, you know? And so, you know, they list some early, early suggested projects and things like that. And so I thought you and I could kind of work through like a where to get started sort of thing. You know, what are some early projects? And then like, I think honestly resources too, like a lot of people like start with, oh, well you should start with art of electronics. I'm no, do not do that. No, no, no. So I don't know. I just, I just feel like, you know, this is an interesting kind of framing for getting started in electronics. I'm sure you've made videos about a lot of this stuff anyways, but.

Dave Jones: Oh, well, I've done a book. I've gone, I've done a textbook shootout of the different textbooks available, excluding art of electronics. Cause it's kind of like a special case. It's kind of like a, a go-to reference text rather than a learning book, so to speak. It's yeah. It's, it's just its own unique beast. It's like, but I did actually review like three or four or near identical textbooks. A lot of them are near identical. If you look at the arrangement of their chapters and everything, they cover like they're almost, you know, they practically copy each other, but they, of course, actually present the information in different ways. And some people will hate one way and they'll love another and vice versa. So it's, you know.

Chris Gammell: Yeah. I think, I mean, so given the person is looking to do like a brain computer interface, eventually, I would think, you know, I'd first go out. If I, if it was me telling this person what to do, I'd say like, first go out, find out what that looks like. What is the, you know, what's a standard kit you can buy? Maybe even go buy a kit to get started with it. Just, you have that in mind.

Dave Jones: That's what I would say. I would say, cause there's, I mean, geez, I was doing this. I was doing like ECG type interfaces back in the late eighties, early nineties. It's not, you know, these, these sorts of, you know, things, you know, you stick some electrodes on your brain. Yeah. Right. Right. Right. And, and, and you get some signals out, right.

Chris Gammell: And you just hack around with it. Yeah. And I think, I think that that, that, that gives a couple of things. One, it gives you the confidence to build, right? Two, it, you know, it, maybe it has a soldering component or some kind of assembly component, troubleshooting component. Like you always say the, I hope your project doesn't work. That kind of thing. Like I, yeah, that's great. Like you need that struggle upfront. And then, you know, the actually having a reference point.

Dave Jones: This is applications based. This is not learning electronics as such, is it? I think it is though. I think it is. It's learning electronics in pursuit of, but it's learning electronics in pursuit of one particular goal. Right. Sure.

Chris Gammell: But I think you need, I always say, I always think you need to start at some, you need to start somewhere. And I do not think it's a textbook. I do not think the first line you read is V equals IR. Let's talk about math. Like that's always my thing. Yep. You know, and that's how every book starts because they have to start there.

Dave Jones: I would agree. Mr. Contextual electronics. Yeah. Right. Right. Right. Yeah.

Chris Gammell: And so I think, you know, starting from a kit though is great. And so if they, people didn't have this and if they just needed like a default option, I'd probably point them at Ben Eater's course. Like I think that that is great too, because you get the computing, you get the, electronics, you have the community. Like I think Ben's stuff is just great as a centering force around like how does, and especially because a lot of people are coming in from the software side.

Dave Jones: So for those who don't know, he builds our breadboard computers. That's right.

Chris Gammell: And he has a, he has a kit now too. Yeah.

Dave Jones: Oh, okay. Cool. Right.

Chris Gammell: Yeah. So he has a YouTube channel. That's great. But then he also has a kit and there's a community around the kit as well.

Dave Jones: Nice.

Chris Gammell: So I just think that that's great. And then, but then still having that kind of like what you want to build in mind is really important. And then, you know, building out a board, maybe understanding that sort of thing. I think a lot of the hacker news kind of respondees also are, they have some very valid points that you need, you need a theoretical basis. And I agree with that, but I just feel like if you slam a textbook at someone's face, they fall down. You know, like it's just like, I don't know. Of course.

Dave Jones: Yeah. Yeah. Some people will get rid, there are a select few who will, that will excite them. Yeah, exactly. Exactly.

Chris Gammell: And some people like they start learning like that. And I think that's great. Yeah.

Dave Jones: Yeah. No, the, the, the vast majority will not like that approach. They'll just.

Chris Gammell: I think a lot of people listening to this, it's going to resonate more of like, let's go find a thing. But I think that leaping forward, this is how I think, you know, we talked about college curriculum too. Like I wish there was some, I think my favorite college courses that I've heard about have that end goal in mind. They have that like build out a thing in mind. Larry Sears, who has been on the show, he's taught at my alma mater. Like he worked out of the art of electronics, but he had things that you were trying to build. And then you struggled all the way along and you needed a lot of theoretical knowledge, but like, but having that end goal in mind is really important. I think.

Dave Jones: Yep. I totally agree. So, yeah, I think this is an excellent approach. Have a, actually you have an idea in mind of something you want to do or something you want to create and then learn the necessary tools to do that.

Chris Gammell: Right. I think, I think the important thing to know is that like electronics is just a slog sometimes, you know? And like, if you don't have that, if you don't have that like motivating end goal, I don't know.

Dave Jones: Oh yeah. It's just. And it's vast. Yeah, exactly. The thing about electronics is vast. I can guarantee you can be like the world's, you know, most ancient graybeard, right? You've worked on everything. I guarantee there are a million different things you don't know and you've never worked on.

Chris Gammell: Yeah. That's the best part. That's why we're here. There is no, yeah. There is no end. I know.

Dave Jones: There's no end. There's no end to it.

Chris Gammell: Get a, get a good, a good grasp on one thing. And it's like, Oh, okay. I'm going to go learn something else now.

Dave Jones: Something else is new.

Chris Gammell: You know, time for RF land, time for more firmware.

Dave Jones: 50 years, you could be working in the field and still you haven't, you know, touched countless things. It's just crazy. So yeah. Yeah. Yep. Anyway, that's yeah. I wish him luck. Yeah. I hope that works.

Chris Gammell: Yeah. I hope so too. I think that's a, it's, it's a, it's a noble, it's a noble goal regardless. And I think that having. And just writing it down helps. Yeah, exactly. And, and gathering people to your, you know, having some people back you and just like, Hey, how's your, how's your brain computer interface journey going? You know, that's really, that can really help, you know, like, and sometimes it won't work out. That's fine. But yeah. So. Right. Thanks to the author of this post.

Dave Jones: All right. Last, last dibs. Is there any last news? Is there any breaking news?

Chris Gammell: I printed a PCB vice.

Dave Jones: You do know you can buy them, right?

Chris Gammell: I do, but I have a printer and you know, you know how 3D printers go. Yeah.

Dave Jones: I like it. And you've got a hammer. Everything. That's right. Exactly. Yep.

Chris Gammell: I've actually been thinking about my Panavise. I've got like one of those huge Panavise ones, which, you know, they very nicely sent it to me.

Dave Jones: Yep. Yep.

Chris Gammell: I don't think there's any other piece of equipment on my bench that gets in the way more than that. Yeah, I know.

Dave Jones: Yeah. I've got a huge, I've got an even bigger one on a base plate with all these big flexible arms coming out of it.

Chris Gammell: I think that might be the one that I'm, that I have as well. Yeah. Yeah.

Dave Jones: Right. Yeah. Yeah. Yeah. It just gets in the way. Yeah. I mean, I've, I've used it a couple of times for various things, but otherwise, no, get that damn thing off the bench. You know, it's like, no, I want some small little tiny special purpose holder jig. Right.

Chris Gammell: I have a, a really big board that I'm working on right now. So it has been good for that. Just, but even still, like, I think that's tough. I think I like that.

Dave Jones: I really like the magnetic PCB things. Yeah.

Chris Gammell: Yeah. I was going to mention those too.

Dave Jones: I call them PCBites. I know. I know. Right. I can't say PC byte. It's supposed to be PC byte. No, I'm sorry. If you're going to capitalize PCB, no, do not capitalize PCB and then expect me to say PC byte.

Chris Gammell: I agree.

Dave Jones: Put a hyphen in there or lowercase B. Yeah. Damn it. I'm going to die on that hill.

Chris Gammell: I really like those as well. I don't use the probes as much as I use the, just the, the little holder thingies.

Dave Jones: Oh, the probes are very handy. Yeah. When you need them, you need them.

Chris Gammell: Totally. Yeah. Yeah. When you need to hold like four signals and you need to do that at a time. Yeah.

Dave Jones: Yeah. Yeah. You just need to hold things in place. Oh, geez. Yeah. Yeah. Yeah. Yeah. I think that's whoever came up with that.

Chris Gammell: Very, very clever idea. Yeah. Definitely hit the, hit the right markup.

Dave Jones: Yep.

Chris Gammell: But, but now I also have a, I actually did, I did a 3d printed jig for a board that I'm working on. So that was also a good use for my, some people do that like kind of naturally. They're like in that mode. I'm trying to get better at it. You know, having, having some jig building though, I'm getting jiggy with it as it were. Got it. That was terrible.

Dave Jones: I won't go down that path. Yep. Cool.

Chris Gammell: Anything else?

Dave Jones: Intel are expanding their fab plants. You know, it's, it's another year. It's another fab plant expansion.

Chris Gammell: So this one, okay. So this is a little, because I've talked about this a lot, I've always predicted that I think they'll split out the businesses eventually. I think this is the first step into that. Right. I think that eventually they'll be like, yeah, we have a foundry now, but also we're going to split out the chip business. So I, I just feel like that's, that's on the way, you know, it's really not going to impact anything. That would not surprise me either. And also what's it called? TSMC is also apparently building a plant. These are both in Arizona, which if people have never been in Arizona, it is, I've always wondered this, why, why, why the desert is so hot there? So hot, but maybe something with moisture or availability of power or something like that.

Dave Jones: But maybe, maybe lack of weather, you know, that like nine tenths of Australia is like that. Yeah.

Chris Gammell: But they're not building fabs there. So what, what gives?

Dave Jones: No, no, no, they're not.

Chris Gammell: I mean, I know some of it is like existing fab capabilities. So there's, you know, talent in the area, that sort of thing we've talked about the show before, but yeah, Arizona in summer. I don't know. Hard pass. I guess maybe the, the, because everything's so temperature controlled anyways, it's not a big deal. And they're consuming so much power anyways. That's fine. They'll just park it next to a power plant and burn a bunch of coal or gas or whatever. But.

Dave Jones: Well, there was an article, there was an article that I retweeted. I don't know if I'll be able to find it quickly enough in time here. It was about China started their own fab, like for this company, startup company that wanted to do their own fabs from scratch. And it just sits abandoned because they couldn't get the P I'm not sure if they couldn't get the people to run it or the money ran out, or it's a, it's a pretty sad state.

Chris Gammell: And they. Both are very possible. I'm guessing.

Dave Jones: Yeah. And they put like, you know, they already put, I think billions into it and they've got nothing to show for it after like five years. Like it's just gone nowhere. And it's like, yeah, sorry. It's not easy to spin up a fab.

Chris Gammell: Yep. Oh yeah. It's, it's a many, many year thing.

Dave Jones: And talking about the sunk cost fallacy. It's like, oh, we've, we've already spent two or 3 billion or, but we're going to need another 5 billion. Should we just, you know, take our $2 billion loss now, or should we keep going? Yeah.

Chris Gammell: I mean, so much of it is the, uh, is the infrastructure piece, right? That's like really what a fab is, is like piping and water processing and power and all the, all the guts of it. And that's about half the cost. And the other half is all the fancy ass machines you have to put in there. Really, really fancy ass machines and then getting fancier. So, uh, yeah.

Dave Jones: Yep. Hmm. But you can't just build it in some warehouse, you know, it's gotta be like class X clean room, you know, it's gotta be. Yeah.

Chris Gammell: I mean, the clean room is not as bad as it used to be. I mean, like, because they, they isolate now in the, the containers, but like, but yeah, still clean room and it has to be significantly large.

Dave Jones: So they, right. So the, so you're saying there's a seal containers that can cut already cleaned that can come into a dirtier sort of environment. And then they go into the machines, which are inherently clean inside and then it never. Right. Okay. Yeah.

Chris Gammell: So I think the fab that I used to work in, uh, which you mentioned is also they're expanding Samsung as well, maybe. Which also shut down. Oh, that's one thing we didn't bring up when the Texas thing happened a couple of weeks ago when all like the power went out.

Dave Jones: Oh yeah. One of them shut down. Yeah.

Chris Gammell: There was an article about Samsung shutdown.

Dave Jones: How, how, how to try and spin up a fab after a power. An XP shutdown.

Chris Gammell: You know, the first thing that happens when I've mentioned this on the show before the first thing that happens when you got an edge chamber and you got a whole bunch of like plasma floating above a, uh, floating above a, uh, uh, wafer. And then you shut the power off. Yeah. All that stuff that's floating drops, drops a whole bunch of crap. And yeah, you scrap a lot of wafers.

Dave Jones: So you have to run those. So you have to run those, uh, 24 seven. Do you? They like, like, just, do they have to keep power to the whole thing? No, no, no, no.

Chris Gammell: So it's like you put a wafer in and then you pump in a bunch of gas and then you turn on the RF generators and then that strikes a plasma. But then as like, imagine as, uh, what's a good way to think about it. It's kind of like a, like acidic cloud. That's kind of like the way to think about it. And so it's kind of like eating, eating silicon or polysilicon or, you know, solder or, uh, or resist material. And then it's kind of like floating up and around there. And then it's, you know, it's still moving through and it's getting sucked down into the, into the turbo pump below it. Right. And so it's all getting moved around, but it is, there is just stuff floating right in the plasma up above. And so then you turn the power off and then that acid cloud just is like plop. And so, but, but, you know, when you're working with such fine, fine detailed things, you've, you've pretty much anything that's in process, unless you can like possibly clean it off immediately, it's scrap. Oh yeah.

Dave Jones: And so it's yeah. Yeah. It's gone.

Chris Gammell: And then, you know, you can hopefully seal up the container.

Dave Jones: But I thought there wasn't much loss in like just losing a couple of wafers, like a couple of 300 mil wafers. I didn't think there was a, I thought the, the, the cost is added in every extra process. Well, I guess if you were right at the end of the, all your process steps, that's added all the value to it. Then, you know, if it happened early on in the process, then actually scrapping up. Yeah. That would definitely. Wafer is. Yeah.

Chris Gammell: I mean, that would, that would definitely suck. I think, I think there's that. And then I think like, but then you think about some of the stuff is also, so you have the ones that are in the chamber. Those are probably toast.

Dave Jones: Right.

Chris Gammell: You think about the ones that are in the machine, as long as nothing got out of the machine.

Dave Jones: How many would be in there at once? Like how, how many?

Chris Gammell: So if you had like the newer, newer machines are a lot bigger and like more expensive than even the ones I used to work on.

Dave Jones: Yeah.

Chris Gammell: Which was now 15 years ago.

Dave Jones: Right.

Chris Gammell: And so, but the ones I worked on, like had like three chambers of, so like three pods or foops that were like 25 wafers each. And then at any time in the machine, you know, you maybe, you might have five chambers and then everything else is kind of sitting in the foop waiting to get processed through.

Dave Jones: So, so we're talking hundreds of wafers here.

Chris Gammell: Oh yeah. I mean, at least, I mean, it depends how big the fab is too. Right. So.

Dave Jones: Yeah. Of course it could be thousands. It could be thousands. So if that power fails, wow.

Chris Gammell: So, so there's ones in the chamber and then there's also the ones that are in waiting. Right. So you think about like, I used to work in etch. So everything's already got photo resist on it. And if you leave the photo resist on too long, you have to like basically take it all back out and wash it off because it's corrosive as it can be corrosive.

Dave Jones: Oh, of course.

Chris Gammell: And so basically when a fab shuts down for a couple of weeks, anything that's like in process, I would assume that a lot of it just gets like tossed. I don't, I don't know how that would work. I never had to deal with any of that stuff. And a lot of this stuff is working on an old memory, but yeah, that was probably not cheap. And given that we're in a chip shortage too, you know, like I'm sure those chips are already spoken for. So now you're.

Dave Jones: We're already in a market where there's component shortages. So it's even impacting my own product. Oh, everyone. Yeah.

Chris Gammell: Oh, finally it hits you, huh? Dave. Everyone. Oh yeah. Yeah. Yeah. Yeah. Yeah. Finally. Yep. Yep. Yep.

Dave Jones: Yep. Yep. They can't make enough multimeters. Oh, we can, we can give you 60. That's all we can give you, you know, this, this month, you're going to have to wait another month or two, you know, before we can even make more than that. That's nothing, man. Yeah.

Chris Gammell: You know, I just keep seeing like car, car factory shuts down. I saw someone post the other day, like.

Dave Jones: Yep. Somebody, yeah. I retweeted it and I, yeah, somebody, yeah. They laid off their workers because well, they had component shortages.

Chris Gammell: Yep.

Dave Jones: So they lay off teams of people. Right.

Chris Gammell: And then you think about all the add-on stuff as well. I think it was Sam Snyder. Maybe. I don't remember who it was, but, but like just the add-on effects of it all as well. Just thinking about how supply chain kind of like goes down the line. So.

Dave Jones: Oh, it's crazy. Yeah. And, and then, then suppliers shut down because they're not getting orders from the auto factory and that was their 90% of their business.

Dave Jones: Totally.

Chris Gammell: And. Sucks. Sheesh. Sucks, sucks, sucks.

Dave Jones: Ugh. Yeah. Bloody manufacturing. Yeah. Yeah. Wouldn't, wouldn't want to be in it for quids.

Chris Gammell: Yeah. Whatever that means. What does that mean?

Dave Jones: Sorry. That's an Australian expression. A quid. Quids. Money. Quids. Oh, like the British people say. Quids. It's like old, old currency. Yeah. Yeah. Yeah. Yeah. Quid. Yep. Yep. So you have actually heard of it. There you go.

Chris Gammell: I know what quid is. Yeah. But you said I would never want to be in with the quids. So like, that sounds like a.

Dave Jones: Four quids. Wouldn't, wouldn't, wouldn't want to do it for quids. Is it, yeah. It doesn't matter how much they pay me. I'm not going to work in that industry. Aussie isms. Yep. Yep. Yep. Yep.

Chris Gammell: We got other links on the subreddit. People can check them out. We still have the survey open. If you have not done that, do that. It's closing soon. Yep. Do it. Do it.

Dave Jones: Yes, please. I'm reading the comments. All right. Talk to you soon. Catch you next time.

Chris Gammell: When you strike a big selling your latest whiz-bang schematic as a non-fungible token, be sure to remember us little people who suggested that stupid-ass idea here on The Amp Hour. You can share some coin and join the crowd at patreon.com slash theamp hour. For all you normal people, you can try that as well and join our Discord. Come discuss NFC schematics and other terrible ideas with us.

Speaker ?: Thank you. We'll be right back.