#110 – Armstrong, Camenzind & Museums - Outstanding Oneirophoros Obituaries

Chris Gammell: This is The Amp Hour Podcast, recorded August 26th, 2012. Episode 110, Outstanding Onero Foros Obituaries.

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

Chris Gammell: And I'm Chris Gammell of ChipReport.TV.

Dave Jones: ChipReport.TV, huh? You're up to episode number three.

Chris Gammell: I am, yeah. This last one was a doozy. I'm actually starting to loosen up a little bit.

Dave Jones: I saw it. And you even told us about it.

Chris Gammell: Yeah. I'm starting to loosen up. Yeah, I'm trying, yeah. You see my Buck Fitty thing? Buck what? Buck Fitty? That was my little fun editing thing.

Dave Jones: Gesundheit, no? I have no idea what we're talking about, Zoom.

Chris Gammell: Chip, yeah. Check out like about five or six minutes in.

Dave Jones: Oh, right, no, sorry, I didn't get that far.

Chris Gammell: Ah, yeah, okay.

Dave Jones: Yeah, no, my attention span's not that long.

Chris Gammell: Yeah.

Dave Jones: Sorry, dude.

Chris Gammell: That's okay.

Dave Jones: Man, Neil Armstrong died.

Chris Gammell: He did. He's not the only one. We lost a couple big ones this week. Who? Last week, too. Well, Hans Comins also passed away.

Dave Jones: Oh, of course, yeah. Oh, yeah, we didn't... Remember, we're two weeks out. We didn't report that because we had a guest last week. Yep.

Chris Gammell: Right. Yep. Very sad. Two big engineering heroes are gone.

Dave Jones: It's crazy. People don't know. Neil Armstrong was an engineer, as were all of the original Mercury 7 and Apollo astronauts. So were all engineers. You had to have a degree in some form of engineering to get into the space program back then.

Chris Gammell: Right. And he was also a test pilot.

Dave Jones: And he was also a test pilot. Of course, that was the other requirement. You had to be a test pilot. You had to be an engineer. You know. Right.

Chris Gammell: Well, at least to be the pilot. Right. Did you have to be a test pilot even if you weren't?

Dave Jones: All of the original Mercury 7 were all test pilots. But the others weren't. Like Harrison Schmidt, of course, got in because he was a geologist. I think most of the others were, though. Okay. But, yeah. Yeah, they were all test pilots slash engineers who went to the moon. Yes.

Chris Gammell: Badass geeks right there. That's it. Yeah.

Dave Jones: And they actually had pocket protectors back then.

Chris Gammell: Oh, yeah. Big time.

Dave Jones: Oh, I love it. Yeah, it's unfortunate. So, yeah, that was sad. It was. And, of course, Obama comes out. I'm going to have a whinge.

Chris Gammell: Of course you are.

Dave Jones: Oh, man. He comes out. You know, I actually tweeted this. This is what he'll come out with, you know. And, yeah, he'll come out, rah, rah, the spirit of adventure's not lost. And it's still, you know, spirit of adventure's still with us. I don't know how bullshit it is. How about funding the bloody thing for a start?

Chris Gammell: Yeah. You know? Yeah, man. I'd like to see that, too. But. Yeah. Dickhead. Typical politician. We've got to stick with Robotics for now. Yeah. Well, you know, politics. Yeah.

Speaker ?: Nah.

Dave Jones: Unbelievable. And I.

Chris Gammell: There's a piece of subtext on the XKCD comic about how many people are left on the moon and the actuary tables about how long they live. And they says something like, there's a range of planets throughout the universe who, due to economic conditions, decided to not explore any further than their own planet. Right. Or something like that.

Dave Jones: Nice.

Chris Gammell: Yeah. Yeah.

Dave Jones: That's pretty much what it is.

Chris Gammell: Right. And I tweeted.

Dave Jones: I did the math. You can put, I think it was 250 tons of people or shit or whatever it is. People. People. On the moon for the, every year for the cost of the U.S.'s maintaining their nukes. Go figure. You know, it's, yeah, it's expensive, but it's, considering what you, what you piss the rest of the money away on. It's, you know. Yep.

Chris Gammell: Yep. Yeah, it's, it's, it's too bad. I've, I've always said that if, you know, just like the original space program, if, you know, if things ever get kind of dicey with the U.S. and China, like in the Cold War kind of starts back up with China, of course, instead of Russia, it's like, that's kind of, sadly, the greatest hope for, you know, a bigger space program is China saying we're going to Mars. Yeah, yeah. And someone else saying, oh, we're going first.

Dave Jones: Yeah, that's right. Another space race. And I, you know, I've seen lots of space videos recently with all the, you know, the Mars Curiosity stuff. Right. Yeah. I've been watching many, many hours of space videos on YouTube, but I can't remember who said it. Someone will correct me, I'm, I'm sure. Maybe many people have, a few people have said it, but, you know, the kids have lost the, you know, they have nothing to dream about anymore. You know, if you want to be an engineer, somebody said, oh, you know, I can't sell to a kid. You know, look, become an engineer and you can eke out a 10% efficiency in this rocket motor. Right. You know, that's not going to get a kid excited and involved, but hey, you can design this rocket, which lands on Mars. You know, I mean, that is right. That's what's going to get the kids in.

Chris Gammell: Yeah.

Dave Jones: To engineering and science and everything else. Right.

Chris Gammell: Yeah. It's, it's tough. It's, you know, I saw there's this great quote, actually. It was a, I forget who it was. I think he's a, what was he? He was, oh, he founded a new big data company and I forget the guy's name though. It was, it was a interview on business insider and it said, a great tragedy is that the best minds of my generation are trying to get people to click, click more banner ads. Yeah. You know? And it's true. I mean, like. I know. It's very true. It's, it's sad, but I, you know, I don't, I don't disparage it quite as much. I mean, you're a little older than I am. I mean, you've got a little more catankered and thisness in you, right?

Dave Jones: Exactly.

Chris Gammell: And then, and that's to be expected. I mean, but I think there is a lot of exciting stuff going on. I mean, there's, I mean, hey, the curiosity, right? That, that alone. I mean, yeah, maybe you're not going to set foot yourself, but maybe you can control the robot that's, you know, blasting shit away with lasers.

Dave Jones: And that's what's going to get kids in. But it needs big, you know, pie in the sky projects like that to do it. Oh, I don't, I don't disagree in the least.

Chris Gammell: I think that kind of stuff is very necessary. I think, I think you're totally right about that.

Dave Jones: Yep.

Chris Gammell: But that's why the amp hour is here. Come on, Dave. We'll, we'll inspire for big, big projects, you know, like big, we got big projects, right?

Dave Jones: You too can be a podcaster someday, kids. Yeah, exactly. You too can eke out a full-time living from being a video blogger. Yay.

Chris Gammell: Someday you can be a cranky Aussie. Repeat after me. Good day, mate.

Dave Jones: Oh man, that's bad. Yeah, it was, right? It's not as bad as my American, so. Yeah, right? Oh boy, yeah. Yeah. Oh, the kids are just salivating at the thought of getting into our profession now, aren't they? Oh yeah. Oh yeah. Yeah. Big time. Oh boy. Anyway, I think it's tragic. Neil Armstrong, one of the true adventurers. Yeah.

Chris Gammell: Well, what about Hans though? Hans is, you know, Hans passed away too.

Dave Jones: An adventurer of analog.

Chris Gammell: That's right. Exactly.

Dave Jones: In the analog domain.

Chris Gammell: Yep. Five, five, fives. We'll live on forever though, you know. Jim, hear what happened? I'm sorry? Sorry.

Dave Jones: Did you hear what happened?

Chris Gammell: Like how he passed away? Yeah. Yeah.

Dave Jones: I don't think he wasn't like a hundred. You know, he was.

Chris Gammell: No, he was 78. It's an article in the e-Times just said he passed away in his sleep. I mean, I'm not sure if there was health problems or anything, but. Right. You know, I finally got the five, five, five contest back online. Oh, right. The site was down for a while.

Dave Jones: I visited every day, clearly. I know, right? Yeah.

Chris Gammell: It's a pretty regular location for a lot of people.

Dave Jones: Why is that? Is it just got up in the Google rankings or what? No, I'm kidding.

Chris Gammell: No one ever goes there.

Dave Jones: Okay.

Chris Gammell: When he passed away, I was trying to.

Dave Jones: I thought it might have been, you know.

Chris Gammell: No, no. I mean, I was just trying to put a tribute to him. Right. And, yeah.

Dave Jones: And even you had forgot your own site was down, right?

Chris Gammell: No, it was. I actually, I went to go update it and it broke. Right. You broke it. I have a theory that anything I do online is like poison. I mean, I can kill just about any server. And, yeah. So, I unfortunately brought that down when I was trying to put up a tribute. Nice work. Yeah. Oh, boy. Yeah. So, there's one up there now, though. And, uh, it's sad, you know. It's, uh. He actually wrote an interesting book, though. I didn't realize this. He put up. Yeah. I haven't read it yet. You can download it. No, not that one. Oh, right. Not the Designing Analog Circuits. The, uh. He did write one about Designing Analog Circuits. And I think that's the one you're talking about, right?

Dave Jones: Yes, I am. Yeah.

Chris Gammell: So, you can download that. Uh, he actually wrote a history book about, uh, forgotten minds or forgotten people in electronics. Yes. Yeah. So, he wrote about Tesla and he wrote about a bunch of other people. I haven't read it yet either. Never read what? But, uh. I, I mean, I might have to pick this up. It's a. It's an interesting idea. Yeah. You know, it's like small things throughout history that, that, um, that really affected a lot of stuff. I mean, him, the irony there being that him being one of them, right? I mean, he. Right. Yeah, yeah. If you don't know, he actually not only did the 555 timer, but he actually put the first, uh, he integrated the first Class D amplifier on chip. I mean, it existed previously. He did not invent the Class D, but he, uh, one of the first controllers he put on chip. And, uh, that's, that's pretty cool, too. I mean.

Dave Jones: And all the, uh, purists out there are booing.

Chris Gammell: And, uh, all the, uh, Class A audio files. Yeah, the Class A audio files. Yeah. Right, right. Yeah. I need to burn at least half my power in order for it to be good. That's it. Yeah. Yeah. No L's and C's in my circuit. Yeah.

Dave Jones: Or if they are, they go on play. Yeah.

Chris Gammell: Yeah. Yeah.

Dave Jones: And hand wound by nude virgins.

Chris Gammell: Yes. Oh, yes.

Dave Jones: Ah, boy. Everyone keeps asking me to do another rant on audio file stuff. There's a big thread on the forum. Ah, there's just so much snake oil shit out there. It's just, you can't possibly keep up.

Chris Gammell: No. Man. Yeah. Yeah. It's not worth it. Come on. It's like, do you want to yell? I mean, like, do you go down to find, you know, the mentally disturbed bum on the street and, you know, shout back at him? It's like, well, no. I mean, you could. Right. But eventually, though, you become the crazy person, so.

Dave Jones: Yeah, exactly. And some of these people have deep pockets and they have no problem suing, you know. Oh, really? Yeah. These people are not of sound mind, okay? They are not of reasonable mind. Right. They will sue your ass to protect their little scam. You, yep. They can keep it. So you have to be careful dealing with these people.

Speaker ?: Yeah.

Dave Jones: Especially the big players. They will, you know, the small players, they're, you know, they're not too much to worry about. But yeah, the biggies, absolutely.

Chris Gammell: Yeah.

Dave Jones: I don't know. But I'm not one to back down. I'm not one to be scared of shit like that anyway.

Chris Gammell: Yeah, you're just not wasting your time. That's just silliness, I think.

Dave Jones: I'm like number one or number two on Google if you type in audio file. Well, sorry, not Google. YouTube if you type in audio file, you know. Woo-hoo! All right. Number two with a bullet. So, yeah. Yeah. Pisses a lot of people off that I'm up there, I'm sure.

Chris Gammell: Yeah. Speaking of wasting time, we not only did I waste time and all of the time I spent working on the discussion server, the discuss.theamphour.com, that is now dead and gone.

Dave Jones: Well done.

Chris Gammell: Yeah, thanks. We actually moved our link board over to Reddit, which Dave hates. But if you go to reddit.com slash r slash theamphour, you can see all of our links that we're talking about this week and next week and, you know, kind of a running list and you can comment on them and everything.

Dave Jones: And you can upvote them. Oh, you can upvote them. Not that we'll talk to them. Not that we'll talk about them because it got upvoted. Right. But maybe we should. I should promote that, right? That's the idea. Oh, sure. If people upvote it, then we guarantee we'll talk about it. I don't know.

Chris Gammell: Sure. Yeah. Maybe. Because this is an off-the-cuff show. Or if you like it more.

Dave Jones: You know, this is just a...

Chris Gammell: Yeah, I don't know. I wasn't doing it by that. But it's just kind of a place to put links. But what I was getting to is a great way to waste time. You can see all of our links before the show even happens. Yeah. And then you can comment on it. And, yeah, that's a great way to waste time.

Dave Jones: As if there weren't already enough ways to waste time on the internet.

Chris Gammell: Yeah, that's right.

Dave Jones: Twitter, folks. Sign up today. Oh, boy. I swear I'm going to have to switch my computer off for X number of hours a day. Or switch my modem off, at least.

Chris Gammell: Switch your brain off. I can't do that. That's not possible. No, it's not. But maybe once the human brain is modeled, we could turn it off like that, right? You put a link on there about that.

Dave Jones: There is a great article in this month's IEEE magazine, which we like. We both get and we like. Yep. The Spinnaker chipset. Um, spelt exactly the same way as in the boat, Spinnaker. You know, the sail that goes out the front and catches all the wind and all that sort of stuff. I'm not sure what it stands for.

Chris Gammell: I thought that's the jib. Isn't the jib up front?

Dave Jones: Oh, I don't know.

Chris Gammell: Maybe it's another part of it. I only know a little bit about sailing. I've never owned a boat. I've only been out once or twice.

Dave Jones: So, okay.

Chris Gammell: Anyway, Spinnaker. Yes.

Dave Jones: Yes.

Chris Gammell: Spinnaker.

Dave Jones: No, I'm sure the Spinnaker is the big, is the big one out the front to catch the, they put it up when they want to catch the wind and go full speed downwind. I could be wrong.

Chris Gammell: All right, cool. I, I don't know.

Dave Jones: Do not take financial advice or sailing advice from engineers, folks.

Chris Gammell: Until they're very successful, of course, right? Once you've made it, then you have probably boats and lots of money to dole out the advice along with, right?

Dave Jones: Oh, boy. Yes, a boat is a hole to pour money into.

Chris Gammell: Oh, yeah. So, let's hear about the Spinnaker chip. It's a chip, right? You wrote about multiple, multiple chips. Now, is it on silicon or what?

Dave Jones: Well, yeah, this, it's a, the Spinnaker is a chip. They started in, what was it, you know, in 2005, they started out this. So, it's been going for quite some time. They got 12, sorry, 8 million bucks in funding from a UK government grant thingo. So, good to see the money actually, you know, spending money on something useful and to model the human brain. So, they developed this chip and this chip has, how many, it's got 18, where is it? Yes, 18 ARM 9 cores. Okay, that's a good thing about the ARM because you buy the IP and you can just drop it in, right? So, they used ARM 9 probably because, well, you know, it's easy to, the tools are easily available.

Chris Gammell: I'm guessing probably when they bought it, they probably bought it a couple of years ago and that was probably, you know, takes a while to get integrated.

Dave Jones: Yeah, it must have taken many years to actually design this chip and then get it made and, you know. So, it's not like you can suddenly change late in the design cycle. Oh, we'll use the latest ARM core.

Chris Gammell: Let's do everything over.

Dave Jones: Yeah, exactly. So, you have to lock it in a long time ago because it wouldn't have been a big group, you know. I'm sure it was a smallish group. Anyway, it's got 18 ARM 9 cores on it, each with its own local RAM and routing and it's got like a big routing core in the middle of the chip. You can go, there'll be a link on the site. You can go check out the internal structure of the chip. And then it's got SD RAM controller. So, each chip has 128 meg of SD RAM stacked on the chip. I'm not sure if it's stacked on die or whether or not, well, I think it probably is actually because it's, I don't see anything on the board. I think so, yeah. So, it must be, yeah, the SD RAM must-

Chris Gammell: Yeah, so they do like a flip chip or something.

Dave Jones: Yeah, they do flip chip or something. Anyway, it's all packaged in this one Spinnaker chip. And then they've got-

Chris Gammell: They put down a crap load of them.

Dave Jones: And then they put 48 of these chips per board. So, what's 48 times 18, right? So, that's how many processors you've got per board. And then the entire system that models the brain or a part of the synapse part of the brain, I don't know, right? Yeah. Don't take brain advice either from engineers.

Chris Gammell: Yeah. Especially not electrical engineers. Maybe like biomedical, right?

Dave Jones: Biomedical, yeah. Anyway, a total of 57,600 chips. So, hang on. Unbelievable. Calculator. Where we go? We're talking 57,000.

Chris Gammell: Was it HP-15 you got there? You got your- No, no.

Dave Jones: I've got a Casio FX-992S, which uses a ridiculous VPAM system. I hate VPAM, but it's the one I've got to hand you. I have no idea what that means. I think the other one's- Visually perfect algebraic model.

Chris Gammell: Sure. Sure.

Dave Jones: Yeah. Anyway. Times 57,600 times 48. No? Why do you say 48?

Chris Gammell: I don't get that. It's 18. Yeah. 57,000 times 18.

Dave Jones: Yeah. There's over 1 million. There's 1,036,000 cores in this system.

Chris Gammell: That's a lot.

Dave Jones: Yeah. But it's not much when you're modeling the human brain. You can only model a couple of neurons or a couple of synapses or something. I don't know. Yeah. Which is the amazing part of the brain. It just models one little tiny node. And scary.

Chris Gammell: Yeah. That's crazy, man.

Dave Jones: I know. But it's awesome. This type of engineering, I love it. It's great. I just love these big boards that have 50 big, huge chips on them.

Chris Gammell: All I can think is they're going to end up on eBay someday. Right. Like you or Mike's electric stuff are going to get one of these. Because eventually they will. I mean, they're so incredibly specific. Yeah. Exactly. And this being versus... Like a lot of supercomputers actually use FPGAs these days. And I think these still have FPGAs on the board if I'm looking at it right. I think they still route some signals with it. Yeah. Yeah. Definitely. But still, I mean, like that's a lot of good gravy. I mean, like that's a lot of stuff, right? I mean, like...

Dave Jones: Yeah. But you can't reuse these boards. They're like, you know, because they're purpose design. Right. Exactly. You can't hack them to do something. Well, you probably could. But by the time you've figured out how to do it, it'd be...

Chris Gammell: Well, what'd you need all this juice for too, right? Exactly.

Dave Jones: I don't know.

Chris Gammell: Did they say what the eventual goal was? I mean, like... I need a chance to read this whole article. Did they say they want to put someone's brain in there? I mean, are we going to have...

Dave Jones: No, no. They just model in certain parts of the brain and, you know. It's ongoing research. There's actually quite a few groups actually try to do this. So... Okay. Yeah. It isn't just one group. That's good. Yeah. I'm not sure what the ultimate goal is, but, you know, it's to figure out... Because they still don't know how the brain works. So that's why they... It's like a development platform to figure out how the brain works, you know. To see if they can develop the algorithms that work similar to a brain. It's not like somebody's already figured out and, oh, yeah, if we just put enough chips, it'll work exactly like the brain. You know. No. They still have no idea how the brain works. Well, they have some, but they need these... You know, it's a test bed. It's a hardware test bed to model all their algorithms and, yeah, try and come up with something that works very similar. So...

Chris Gammell: That's pretty cool. That's pretty cool.

Dave Jones: I like it.

Chris Gammell: Yeah. And they're not the only ones actually looking into this kind of stuff. Wired had an article this week actually about analog processors. Analog processors. I don't know if you saw that.

Dave Jones: No. They're making a comeback, are they? Analog processors. Well...

Chris Gammell: Didn't they go out in the 60s? I think it's all theoretical still. I mean, it's a DARPA thing. But basically, it's the idea is that eventually you're going to run into power and speed limitations and size, right? Yeah. And so eventually you just move to a different topology. I mean, you've stopped doing binary logic, you start moving to tertiary logic, and then eventually you can keep moving your way up.

Dave Jones: Yeah, but then you get to a point where, yes, it is analog, but then you can't... There's a reason why we went to digital. Away from analog is because it's predictable.

Speaker ?: Right?

Chris Gammell: Right. Yeah. Yeah. I mean, it has less noise problems. But I mean, you think... Exactly. Exactly. But they also talk about probabilistic computing, basically. You know, basically...

Dave Jones: 99.9% of your bits will get there. Yeah, you don't... Well, that's the idea. On average.

Chris Gammell: Basically, you don't need to do whatever... You don't need to get all the information all the time, especially at high speeds. That is the idea. Right. Yeah. And I mean, this is all very early stuff. I mean, this is still in, you know... They don't even have chips yet, I don't think, so... Right. But they're looking into it, and that's kind of cool. And I've said that for a long time, too. I mean, like, you think about where it can go. There's only so many knobs you can tweak, and eventually you're gonna... We're running out of the space one, right? I mean, and even... Yeah. You know, who did we talk to about that? Was it Oiven we talked to? I mean, basically, even when you keep shrinking transistors... Maybe it was Steve Liebson. I don't remember who it was. But as you keep shrinking transistors, you run into a whole lot of other problems. I mean, they're already running into some of those at 28 nanometers, which is what the current node they're at. You know, the next is 20, and then 14, and then 10. The problem... You know, like, it's still very tough, because as soon as your transistors start getting smaller, then your leakages go way up, so your power starts, you know, going way up, too. And, you know, it's just... It's just there's a whole range of problems, and then there's the whole running out of atoms. Eventually, you're going to just be, you know, you have to say, one atom on, one atom off, one atom on, one atom off.

Dave Jones: Yeah, but then you can go in other dimensions. You know, you go, well, the other dimension. You go, instead of 2D, you go 3D. Right? And it's... Yeah, that's true. There's ways around that. There's physical ways around that.

Chris Gammell: Yeah, but that's still all predicated on a smaller gate size and everything else, right? That's been what Moore's Law is built on, is the smaller gate size. And I have no doubt in my mind that they'll continue to innovate on that stuff, right? But eventually, they're going to not run out of...

Dave Jones: You must. Yeah, I mean, there is a physical barrier.

Chris Gammell: Yeah, yeah, yeah, yeah. But, I mean, it's not going to be a running out of room or running out of options thing. I think it's going to be a running out of economic decisions that make sense after a certain point, right? I mean, if it takes you... Right, yeah, of course. If it takes you a... You know, if it takes a 50 billion fab instead of a 10 billion fab, as it is these days, right? That's right. And eventually, you run into some kind of problem. I mean, still, you're still going to be making the analog stuff at 90 nanometers or whatever, but...

Dave Jones: The industry is self-regulating, kind of, in that way. You know, there's only so much people will pay for a chip, and there's only so much the investors will pay, and there's only so much, you know? Yep. Yep. So, yeah, and whoever comes up with the cleverest idea at the lowest cost will win.

Chris Gammell: Yep. Yep, definitely. That's what it comes down to. You know, and it's become such a big economic argument, too. You know, it really is, like, it kind of sucks, because it's like electronics running into economics. And that's kind of not outside the scope of the show, but outside the, you know, interesting factor, right? It kind of gets kind of boring. And, you know, we know that sourcing and, you know, part, what's it called, like, buying parts and everything, that's all still really important. And, but it's just kind of, it's kind of boring sometimes.

Dave Jones: Yeah, but it's, that's what engineering's all about. Engineering ain't landing shit on the moon all the time. It ain't landing shit on Mars. Yeah, yeah, that's right.

Chris Gammell: Right. It's not always sitting in the control room with a mohawk, right? Yeah, exactly. It's sometimes also sitting at your computer working through MATLAB simulations and punching your keyboard, right?

Dave Jones: Have you seen the video that they did where NASA and we know it? Where NASA and we know it isn't? Oh, yeah, that was, oh, that was lame. Yeah. Yeah, but it was, it was kind of funny. Yeah. I thought it was pretty funny.

Chris Gammell: Yeah. Well, we'll, of course, now put it in the show notes and everybody can see it. I'm easily a muso. Yeah, yeah. That's funny. It's, it's, the musician in me was just like, oh, come on, you didn't sync up the, like, the lyrics to the music. Come on.

Dave Jones: See, I know jack shit about that, so that's why I enjoyed it, you know? Yeah.

Chris Gammell: Yeah, see, once you get past that, it is a lot of fun.

Dave Jones: Right.

Chris Gammell: Speaking on the fab side of things, there's actually talk now of Apple buying into fabs and maybe having, so basically, Morris Chang, the guy from TSMC, mentioned the possibility of Apple having their very own fab. Wow. Right? That's how big they are. I mean, some people don't know.

Dave Jones: Well, they've crunched the numbers, right? Right, yeah. And they've gone, it's going to be cheaper. We've got, because they've got like $50 billion cash in the bank or something, right? Right.

Chris Gammell: Yeah, they did, I think. They probably still do, right? I mean, I don't remember.

Dave Jones: Yeah, I'm sure they do. I think they just had record profits or something, you know? So, I don't know. It's just, yeah, of course they can. You know? They can do that with petty change, you know?

Chris Gammell: Yeah, I mean, and they do that already, right? They come to a fab and they say, we want to buy your entire output, you know, all the wafers in a month. And so, I mean, I guess it does kind of go from there, but I don't understand the whole, like, so now they're talking about insourcing, right? Where Apple is actually in control of the fab and everything like that, too. I don't think that's going to happen. I think that's a load of crap, to be honest. I don't think, that's like the difference between being like, hey, buddy, do you want to have, like, four girlfriends and, you know, you can just kind of imply what you want to all of them or do you want to, you know, marry one of them and be tied to the fate of that? No, don't get me wrong. I'm married and I, shit, I'm getting myself in trouble here. But basically, that's the idea, right? I mean, they'd be tying themselves, you know, to that fab. So, if something went wrong, they couldn't just be like, all right, well, we're going somewhere else, you know? So, that's the difference. I don't think it's going to happen.

Dave Jones: You want to know what pisses me off? Apple just won their patent lawsuit against Samsung. I know.

Chris Gammell: I know.

Dave Jones: It just happened, what, yesterday or something? Yeah. And now they're talking about, yeah, the, you know, the Samsung phone may be banned in certain countries. As their tablet was here. I mean, they couldn't sell their tablet here in Australia while a court case was pending. They finally won it for their tab notebook or whatever it is.

Chris Gammell: Yeah.

Dave Jones: But, ah, man, it just pisses me off.

Chris Gammell: I know. It's just, it's a lot of lawyer shit, right? I mean, there's a reason Apple has a lot of money. But, man, yeah, it's, it's, it was just, you know what's annoying to me about all that? It was, it just flooded, it flooded the news stream, right? It was just like you couldn't get away from it. Exactly. And it's like, ah, I don't give a crap what, you know, like, oh, well, your UI has this button that's just like our UI that has that button. It's like, who cares? I know. Get out of my face. I don't want to hear this stuff. I just want to get past it and hear about, I want to, you know, if, if they're talking about new stuff, I don't know. I, I, what I, my hope is, is that it's a, it's a window for some smaller player to come in and, you know, eat their lunch, but I don't know if that's going to happen. And it's just, it's just such a big game these days that it's, it's a mess, right? It's like there's.

Dave Jones: And we're sick of hearing about it.

Chris Gammell: Yeah. It's, uh.

Dave Jones: Gee, no, you know. Did you hear that, uh, Microsoft were going to make like 3 million of their glass tablet things?

Chris Gammell: The Surface?

Dave Jones: Yeah, Surface, or whatever it is. Yeah, something like that. Even though nobody's actually used one yet, apparently. See, that's the thing. It's kind of like a vaporware and they've suddenly announced that they're going to make 3 million of them or something. It's jumping right in there.

Chris Gammell: It's just so weird how all these, these software companies are making hardware. It's just because they all have so much, you know, software has so much money, right? I mean, like, they have tons of money, right? Yeah. Hello, Val. Hello, friends at Val. Yeah. Right. It's the same thing, right? Yeah, yeah. It's, it's just, it's, they have money and now there's an economic reason for them to be making hardware and it's like, well, hope they hire some good hardware people like Valve did.

Dave Jones: Exactly. Valve may not necessarily manufacture their hardware. Right. But, yeah, they're doing it for the R&D side of things and, uh, yeah, they feel as though they have to. Yeah. It's an area that they have to at least be in from the design side of things.

Chris Gammell: Yeah, I know. It's like, I don't like that it's like, oh, well, we have to do hardware. It's like, no, want to do hardware. We want to do hardware. Hardware's cool. I don't know. Eh. And this whole, this whole economic side of this stuff is just messy. It's bullshit. Come on. Yeah. It's bullshit. Right.

Dave Jones: I know. We hate it.

Chris Gammell: Yeah. Can we talk about something different? Yeah, let's get into something fun. Um, how about the obfuscated sea contest? You, uh.

Dave Jones: That's not fun either. That's stupid. That's not fun.

Chris Gammell: Oh, I think that's fun as hell.

Dave Jones: No. To, uh, who can come up with the most obfuscated, you know, convoluted, unreadable line of code?

Chris Gammell: I think that's fun.

Dave Jones: No. Yeah. It's fun for a bunch of penguin dwellers. No way, man. It's firmware people too. I mean, come on.

Chris Gammell: Oh, no. Come on. Some of our listening audience would like this, I think.

Dave Jones: Oh, yeah. I, ah.

Chris Gammell: In terms of programming languages, you don't, there's only one step further that you can get closer to the hardware. So, I, I know that some of our listening audience would care.

Dave Jones: No.

Chris Gammell: Now, now, if you ever do this and you hand me a piece of hardware that has obfuscated sea on it, I'm going to punch you in the face.

Dave Jones: Because it's not practical. That's what I hate about it.

Chris Gammell: Right, right, right.

Dave Jones: And it's not even that fun because you read it and you go, oh, okay, I don't understand it. And, like, it's, there's nothing fun about it for me.

Chris Gammell: See, now, okay, here's the thing. Like, here's why I think it's fun. It's because me and my friend at work used to do obfuscated circuit contests. And it becomes a puzzle, right? So, I mean, I get it from the sea side of things. Yeah, yeah. It's, you know, like, you and I are pretty good at it. If you want to spend the time solving the bloody thing, yeah.

Dave Jones: Right. All right.

Chris Gammell: Exactly. And that is, that's the idea. It's like, it's a kind of an art form and people enjoy trying to dig through it and figure it out, right?

Dave Jones: Hmm. Yeah. And then you put dead end stuff in there and you put it in there.

Chris Gammell: Yeah, exactly. Yeah. You kind of, and, you know, they've done it on chips before too. They actually put in, like, fake, fake parts of a, you know, on an ASIC, they'll put in fake, fake circuits, basically. They don't do anything. They do that same thing for when, you know, people x-ray it and try and spend six months trying to figure out what it does. And it's, oh, that, that doesn't do anything.

Dave Jones: No, I just get, probably I'm a bit bitter because I just hate people who, you know, these smart-ass programmers, right, who code it in the most, you know, obscure, people actually do this in real programs, right? They, you know, they use, you know, instead of just making it all real simple with tons of comments and everything and making it real dumbass, no, they do the most convoluted way to do it just because they think and just because they can, right?

Chris Gammell: Right, it's like egos running into coding, right?

Dave Jones: Yeah, egos, yeah. Look at how good a programmer I am. So, oh, look, I'm going to put this all on one line. Well, you know, when you put it, when you run it through the compiler, it's going to assemble the same way as if you did it on five lines. You know, they're saving four text lines, but they're not gaining any speed or optimization. They're just making it less readable.

Chris Gammell: Right. Right? I don't think the obfuscated C contest is meant for production level runs.

Dave Jones: But I find that very common in regular C code. You know, I mean, probably, you know, I'm a bit pissed off because I'm not very good. You know, I can program in C, but I'm not, you know, computer science trained level in C, right? I'm just a hardware hack C programmer, right?

Speaker ?: Yeah.

Dave Jones: I've done it.

Speaker ?: I've done it.

Dave Jones: I've done it. I've done it even less than that. Exactly. Well, that's what most engineers, right? They've done a couple of classes on C, you know, and they can get by, right?

Chris Gammell: Right. And you just get someone else's program and you kind of fiddle with it until it works. Yeah.

Dave Jones: And you get a program which is written by some computer science freaking, you know, major or something. Ah, not in C. Come on.

Chris Gammell: You get a firmware guy who's really good with that stuff.

Dave Jones: I know. But yeah, exactly. But, you know, I get it all the time, right? I'll put my source code out there, right? And people will go, oh, this is shit. I'm going to rewrite the whole freaking thing because you've done it the wrong way. What? Right. It's readable. How have I done it the wrong way? You know? Right. I mean, you know, if, granted, if you've got a reason to do it, you know, like you have to optimize the speed or something, well, I would have put some effort into optimizing the speed of it, for example, or optimizing that loop out or something like that, you know? And, oh, no, they just, no, I'm just, yeah, I get it all the time.

Chris Gammell: Let me ask you, have you ever seen this where people do this on purpose at work in order to like for like job security or something like that? Oh, yeah, absolutely. Yeah. I've heard stories about that kind of thing. I've never seen it myself, but like even on the circuit side of things, they'll put it into like, they'll put like extra circuits or extra stuff on schematics. It's kind of, you know, the same thing with like an ASIC, right? They'll do that with, on a schematic, they'll do that for a board and they'll actually put in fake circuits there just for job security.

Dave Jones: Yeah, not so much, you know, that sort of deliberate, but I know it's more common for people to deliberately not do good documentation, for example. Really? You know, like their job is they're supposed to produce documentation, but they produce shit documentation. And that's the same as obfuscating your code, really. Yeah. Yeah. Effectively. Obfuscating your circuit, right? Yeah. Effectively. Yeah. So, especially in big complex systems, right? You know, if you don't have really good documentation, somebody coming in cannot understand it all, right? You could spend years trying to come up to speed on this thing. Oh, yeah. So, you know, that's another way to do it. It doesn't have to be through the code or the schematic level or whatever. It's just, you know, sheer laziness or sloppiness in terms of documentation.

Chris Gammell: Yeah. Well, and I've even seen, I've heard of and I've seen a little bit of like people putting like fake signal names or they'll put in like a coded, you know, like coded signal name so that only they could then decode it and they'll do like a find and replace. And, man, that's messy, man. That's like, I mean, like, that's like internal industrial espionage, right? I mean, like.

Dave Jones: Yeah, yeah. Exactly. That's crazy.

Chris Gammell: I don't know. Does that still happen? I would love to hear from our listening audience if that kind of, if they've seen that stuff still because, you know, I've heard of it, but it seemed like it was kind of like a older days kind of thing. And, you know, and then I compare it to like, you know, like how you and I kind of know some of the open source hardware folks and that's just the complete opposite of it, right? So just kind of that computing in my brain, it doesn't really, it doesn't make as much sense. Whereas maybe back in the day, you know, like it was a big thing or maybe it was just a, you know, a workplace thing and engineering egos kind of was dependent on those, right? I'm not sure how to really process it, but.

Dave Jones: Some people don't even deliberately do it as such. They just got some, you know, some weird way of programming, you know, they've just got, you know, they've got a weird way of doing VHDL. They've got a weird way of programming. They've got a weird way of laying out schematics. They've got, you know, it's just non-conventional really.

Speaker ?: And.

Chris Gammell: We talked to this guy over on the Engineering Commons podcast about software and he talked to a lot of scientists and he was kind of teaching them how to code better and everything. And he mentioned, he mentioned he met a guy that had 20,000 lines of code on this, this one project.

Dave Jones: Yeah.

Chris Gammell: And he's like, he's like trying to troubleshoot and everything. And then eventually he finds out that the guy only used the top 5,000 lines of code. Basically when he didn't, he didn't need it anymore. He would use that as his document revision. He wouldn't copy that out of the, you know, the active part of his program, then put it below. Right. And some of it is just people not, you know, never having good programming practices. Yep. Like, I mean, myself, but that's a pretty extreme example too of like having, you know, 20,000 lines of code where 15,000 aren't, aren't, aren't used.

Dave Jones: Hmm. And that's, yeah, that's one of the things that, uh, pisses me off. You know, people will use, will try and do it in one line, you know, as fewer lines of code as possible.

Chris Gammell: Yeah, exactly.

Dave Jones: And it's, you know, whereas, you know, there's no advantage to doing that. You know, don't be terse, be, you know. Be verbose. Unless it's, yeah, you know, be, be verbose. Unless there's some real legitimate, you know, hardcore requirement for you to have to do it, you know, some other way. Be verbose.

Chris Gammell: Assume your follow-on will be an idiot, right? Assume your follow-on will be an analog hardware guy who has no idea what's going on.

Dave Jones: Just, just simple stuff like, like, like, assigning, assigning a variable, do it on a separate line, you know. People know, they'll use the C, you know, thing where you can assign the variable first and then, you know, and no.

Chris Gammell: Yeah.

Dave Jones: Jeez. Yes, you can do that. Yes, C allows it. Yes, but does it make for readable code? No, I don't think so.

Chris Gammell: Yeah, just because you can doesn't mean you should, that kind of thing. Yeah, exactly. Right. Yeah. Right. I've been, I've been running into all this stuff. I've actually, I've actually been jumping back into, I don't program that much. I mean, I'm mostly doing hardware stuff. Yeah. But, uh, I've been wanting to, like, exercise some hardware and, you know, just, like, simple spy line kind of stuff. So, I've actually, I've been using the BeagleBone and I've been trying to get that up and running and I'm, I'm making a daughter board for that kind of thing. Right. And, man, it has been a while. Hope webs, huh? Yeah. Oh, so many. And it's, it's, you know, like, it's even, even just like, I'm like, oh, I want to get up and running and, you know, I want to make sure I know how to use this thing before I, uh, before I start, before I even lay out a board for it, right? Because if I can't do it, then. Yeah. Who, why would I, why would, you know, why, why have a board come in and then be completely clueless with what it is, you know, like be clipping wires to it and wiring it over to something else.

Dave Jones: Oh, it doesn't work. Yeah, exactly.

Chris Gammell: Yeah. Yeah. Yeah. It's, it's a lot more Linux based than I thought it was. I mean, like, I should, I should have known all this stuff, but, uh, it's very, very Linux based. I mean, basically it's running Angstrom, but.

Dave Jones: See, that, that means nothing to me, you know.

Chris Gammell: Oh, that's, that's a distribution of Linux. Right. So it's kind of like Ubuntu, but it's modified for like. Right. Smaller devices and everything. And I've been doing, you know, uh, on the bright side, I've been doing a little more Linux stuff trying to, when I've been fighting with our servers. So, uh, I have a little more, uh, knowledge about it, but it's, you know.

Dave Jones: I think if you have to go down to that level, you're doing it wrong. See, now what would you do then? So, like, say you just, say you just want to. I've never had to run, do anything Linux on my servers. And I've been operating my own servers for. Oh, you mean server side. 15 years.

Chris Gammell: Okay. No, I know what you mean. Okay. I thought you meant on the semiconductor side. No, no, no. On the server. So, if you're doing like a dev board though, how would you do that? Say, say you wanted to just try out a new, you know, a new spy ADC, right? Yep. How would you go about exercising that? How would, how would you actually, I mean, would you just build it and go?

Dave Jones: Well, these days I might actually use the, um, use something like the bus pirate, for example. Oh, okay. Which, you know, you hook the wires up and then you have a slightly, then you have a higher level, you know, programming language effectively that you can send commands to it and get commands back. Just like a ping kind of.

Chris Gammell: So, is that graphical then? I've seen your video but I don't remember.

Dave Jones: No, it's all text. No, it's all text based. So, you send it, you know, you type a text command into the serial command window. That's right.

Chris Gammell: Yeah, so like a hyperterminal or something. That's right. I didn't see that.

Dave Jones: Yeah, serial hyperterminal thing.

Chris Gammell: Okay, so you would just send certain commands like that. And that's the kind of, that's the same thing I want to do, you know? Like, basically the thing I like is I want to have a single platform where not only, you know, I want to build a little board. I want to be able to have, you know, like, you know, a laid out board and then I want to be able to rip off chip number one, put down chip number two, hook up signals, send it to the fab and have it back, right? Basically making like breakout boards that also happen to have, you know, spy and, you know, a serial link, you know, like just a simple link kind of stuff. And like I said, it's bus pirate, but I don't want to do it on a breadboard, you know? Like I can't, I can't wrap my head around doing a breadboard these days.

Dave Jones: Yeah, same, same here. I, like I groan if I have to sort of try and do it on a breadboard and a lot of surface, you know, a lot of it's going to be surface mount stuff too. You know, 0.5 millimeter pitch. Oh, yeah, you can get breakout boards, but I don't keep those in stock, you know? I've got to, oh, I might as well, you know, by the time I order that, I might as well order a board, you know?

Chris Gammell: And it's just kind of the new reality too, right? So, I mean, I think about all the analog stuff I work on. Basically, I mean, I need to be able to talk to chips these days, right? That is just the new reality. I mean, that's why I started Chip Report TV. I wanted to, basically, the idea is these days you're just, we're just chip integrators, right? I mean, that is the sad reality of it, you know? Pretty much, yeah. But it allows us to do so much more. Exactly. I'm a one-man band, right? I mean, well, once I figure out the beagle bond, I will be. But that's the idea, right? You need some way to talk to a chip. You need some kind of development platform because, you know, dip parts are just not realistic. I mean, people complain about that, but it's just, sorry, folks, it's just reality, you know? We are in a QFP world, you know? We have moved beyond the dip, unfortunately. And, you know, I could go and I could buy, you know, breakout boards from SparkFun, and they're good for that kind of stuff. Or, you know, there's lots of options for that kind of thing. But then, you know, even if I'm buying these chips, if I'm doing 20 mega samples, I'm not doing 20 mega samples on a frigging, you know, breadboard because, you know, those kind of, you'll start getting parasitics in there. I want to actually, worst case scenario, I'm doing copper clad. And best case scenario, I'm laying out a board and actually have a decent layout. Exactly. You know, yeah. And then you get all the back end stuff, too. Like, that's the other side of it, is that, you know, I'm developing with these chips, right? I need to not only, you know, check if they work, right? I assume they work. That's not always a great assumption. That's not always a great thing, right? Assuming that a chip works, you know, like errata sheets do exist. But on the same token, once I, you know, if I actually exercise the chip by throwing commands at it, then I'm already halfway there when I'm on the firmware side, I can just hand over this code that does work. And I'm sure you've run into that with your USB power supply and everything else, right?

Dave Jones: Oh, yeah. Tons of projects like that. Yeah. Well, you know, I will, if I've got an existing project that uses a similar interface, like it might be an SPI or an I2C or something like that, a lot of the times I'll take that and I'll, you know, I'll have my separate board with the new chip and I'll wire that into the old board because that's the platform that drives the, you know, and I've already got some code that can drive that and I extract all the other stuff I don't need and just leave, you know, the code to drive the SPI bus or something. Right. Yeah. That's really... And it starts from scratch and, you know, yeah.

Chris Gammell: That's smart. You know, the only downside to that is that you end up usually, you know, see people who are like, oh, well, I mean, if you're using 8051, it's not a big deal, right? It's like, oh, that's still around. But you don't really ever... I mean, like, you just end up using the same thing in projects. Like, I'm sure you've seen that. You've... Yeah. If you use the same chip, you're like, oh, well, it works, right? But then, you know, if you miss out on some of the newer stuff, if you have like an ARM chip that already works, you know, like, or the ARM chip that has new features or like an FPGA or something like that. Exactly. There's never really any incentive to move on to the next thing. So that's the only downside, really. But on the other hand, if it works, it works, right? Yeah, no, that's fine. Yeah. Yeah. Get it up and running. Get it out the door. Right. Just get it shipped. That's what's important.

Dave Jones: I agree. It's always a pain in the ass. That's why we've talked about this before, I'm sure, that a lot of the time it's just not worth prototyping and hacking something together to test. I just go straight to the production. Well, you know, I have a first stab at a production PCB.

Chris Gammell: Right. Right. Exactly. You expect it might fail, right?

Dave Jones: Yeah, you expect to do a couple of revisions, but that's fine. But at least I've got 95% of the guts of it there and I can program it and I can, you know, I don't.

Chris Gammell: Right. It's just much easier to do. And the only time you run into it is when you need a new digital platform, right? You know that you need to talk to it with a new processor because you need a new display. You want like a touchscreen now instead of like a serial kind of thing. So, yeah, that's... Yeah, man.

Dave Jones: I just jump straight in there and lay out the board and get a bang, you know? Right.

Chris Gammell: You save a little money that way too and you save time. Yeah.

Dave Jones: Possibly. Yeah. But, you know, I mean, often, you know, you get down the track, but usually you're going to get it, you know, 80 or 90% right.

Chris Gammell: You know, usually...

Speaker ?: Right.

Chris Gammell: A couple cuts, a couple jumps, right?

Dave Jones: Yeah, exactly. You're going to have to, you know, yeah, you're going to have to change your board, but it's usually not going to be a complete rip up and re-layout kind of thing, you know? And, you know, so that time, you know, 80% of that time, 90% of that time you invested in doing that board, well, it pays off at the end because then, you know, your final layout is, you know, 80, 90% done to begin with. And then your code's done because you haven't hacked code from here or there, you know, it's all on the final platform you're going to be using and, yeah. Yeah. Yeah. I don't know. I bet each to their own. I do see a lot of people still, a lot of people prefer, right, I'm going to, you know, go to, you know, hell on earth to build up, you know, to hack up this thing and try and get it working before I commit to my PCB. And, well, if that's what they want to do, fine.

Chris Gammell: Yeah. Well, I think you see that more with lower budget projects too because you need to get it right, right? I mean, like, I've worked somewhere before where it's like there wasn't money to expect failure, right? It was like, no, you have one shot. You have to get it right. And that's the difference, right? If you, I mean, I personally think that, you know, there's already a, there should be a budgeted failure in every project, right? I mean, you should expect, you know, at least one board turn, probably two. And, you know, you don't want to be sloppy about it, but engineers aren't going to be necessarily sloppy. And it's like, I think you, you get a lot, you get a lot of, uh, you know, if, if, especially because then you have engineers who are willing to try stuff out, right? If you know that your first board turn is going to fail, right? It's not going to be the final revision unless it's, you know, like you have a, you know, a 5% chance, then you're willing to throw in other circuits on there that could end up being your quote unquote innovation, right? If you say from the beginning, oh no, you have to get this right. What are you going to do? Right? I mean, you're going to put the exact same digital section, you're going to put all of the exact same stuff on there versus trying out something new and, you know, actually doing something cool, right? I mean, that's, that's the downside to it all.

Dave Jones: And often you're, uh, you know, the, the system you're working on is, uh, so big and complex. You can't just hack up little bits of it, you know? I mean, you just, you're going to go for that big final board, you know, you've got your, the product you're designing is one big final board. The only choice you really have is to go for one big final board. You don't have time to dick around and, uh, you know, and spend months and months developing all these little boards to test out all different various aspects. You're going to jump straight in there and, you know, and as I said, you know, 80%, 90% of the time you're going to be there. So yes, you'll do a few revs, but at least you have something to show off at the design review meeting, look, it's almost there, you know?

Chris Gammell: Yeah. Yeah.

Dave Jones: Whereas if you show off just a little, you know, a breadboard with a couple of link like that doesn't impress anyone, right?

Chris Gammell: No.

Dave Jones: A breadboard, oh, look, I managed to, you know, you go into your design review meeting with all the managers, look, I built the breadboard and I'm able to talk to this chip and they go, well, who cares? You know, it's not, you know, they, they don't have anything to scratch and sniff.

Chris Gammell: Right. Exactly. Yeah. They want like a, they want a feature, whereas you, you have a, uh, you know, a component working, right? Yeah, exactly.

Dave Jones: No, yeah.

Chris Gammell: No, they want to see, they want to see the, the pretty display. Right. And that's like, wow, you get that in the last 10% of the project. Yeah, exactly. Or after you've already released and you actually are iterating on the stuff you can update in firmware. It's like, no, sorry. The, uh, you know, getting that last PPM of, of error out of your design, no one gives a crap, right? Unless you can't sell it.

Dave Jones: Or they want the physical final form factor board to fit in the case so that, you know, the, so, you know, so that they have at least something to show off.

Chris Gammell: Right. You know? Yeah. That's a, that's a good, uh, that is a good lesson for young people. I'm learning that one. I've learned that over the past couple of years of the, you know, like the, yeah, the, the drawings, the drawings of, of your product matter just as much as the actual, you know, like CAD drawings of your product, right? The, the industrial designs can matter just as much.

Dave Jones: And often the circuit doesn't matter at all. Right. You know, nobody cares. Management don't care how you, what circuit you use to get this thing working. They want, they only care about the final form factor, right? So it's important to get that quickly up front so that you can show some progress. Hey, look, we're on the right track and people can play around with it.

Chris Gammell: Right.

Dave Jones: You know? Oh yeah.

Chris Gammell: I tweeted about that the other day. Just said, no, I mean, people don't give a crap. They don't care. You know, like if you, if you walk into a meeting with a range of different engineers, you walk into a meeting with a mechanical engineer, an industrial engineer, you know, like a software engineer and, you know, maybe one other EE, right? If you walk in and you say, oh, well look at, look at like how cool this, this A to D is, or look at how cool this display technology is. They're going to look at you. They're going to blink their eyes and they're going to say, well, does it work? I mean, like, does, you know, like, can I, can I control it externally? Is it reliable? They don't give a crap how it works. They just want it to work. That is exactly, that is the basis of, you know, high level electronics. They don't care how it works. They just want it to work. And it sucks. And it's nice when you have people that understand how it works. That's why you should cherish people that do understand the electronics because, yeah, because there's a lot more that don't give a crap. There's a reason that we are not a, we are never going to be a very popular podcast. I mean, we have a good chunk of people that do care how stuff works, but, you know, there's a lot more people that care about, you know, how the gadgets work, right? There's people that want to know how the phone works and what apps you can put on it more than, you know, what's inside the chip and, you know, how many transistors it has and, you know, like all the low current specs it has or whatever, right? I mean, like, no one gives a crap about that.

Dave Jones: And I've, you know, even for my own personal projects, right, I don't get a lot of satisfaction of breadboarding a chip up and talking to it. It's like, eh, you know, I mean, if you've seen my...

Chris Gammell: Well, you get the accomplishment, right?

Dave Jones: Well, yeah, you get the accomplishment, but I know I can do that. It's like no big deal, you know. I, even I want that final form factor. I mean, if you've seen my USB power supply video, right, I, you know, I hacked up a final board and just did it, you know, I didn't, because I wanted something that I just can look at and go, yeah, you know, look, I've got something real. You know, I've actually shown what looks like almost a finished product, right? And I did that, just bang, you know, I didn't even think about it. I just, you know, I was getting some boards made, so I lashed up a thing and I laid out a board and bang, I put it on the panel and, you know, I ended up with that board. Sure, it's not the final one. I'm going to completely change it, but at least I ended up with something that I could scratch and sniff and play with and turn the knobs. And get feedback on, right? You know, yeah, exactly.

Chris Gammell: Yeah, show other people. That's also important.

Dave Jones: And, you know, whereas if I had a circuit on a breadboard, it's like, no, boring.

Chris Gammell: Yeah, that is a little boring.

Dave Jones: That's even for my own personal project. I don't know, but everyone works different. I mean, you know, some people do get, you know, a big sense of, especially as a beginner, you know, you get a big sense of satisfaction for building your chip up on a breadboard and your lead flashes and you finally talk to it. Great, you know?

Chris Gammell: Oh, yeah, even before that, I mean, I remember some of my first boards, like when I got the boards made, I was just turning them over my hands nonstop. I couldn't, you know, it's just a good feeling, you know, like getting that board in your hands. It's like, I made this, man. That's cool. Yeah, I know. I like this. I will stick with electronics. I shall continue. Yeah.

Dave Jones: So is, with, that's an interesting question, with custom PCBs now being so cheap, the only reason not to do it is time. Really, you know, it takes X amount of time to, you know, X amount of a week or two to get your board, you know, manufactured and sent to you. But there's no real major cost reason anymore, you know, with the likes of the various low-cost PCB share services available.

Chris Gammell: Oh, yeah, yeah. So what do you mean by that, though?

Dave Jones: I mean, you mean that breadboarding is a way of the past? That breadboarding, is it, yeah. Is it going the way of the...

Chris Gammell: I don't think so. No, I don't think so. Because it's good for beginners, right? I mean, that's why you do it, right? So you can actually see that connection. I mean, making that jump into PCBs is tough, you know.

Dave Jones: Yeah, true. But once you've done it, do you go back...

Chris Gammell: Once you've done it once, though, yeah.

Dave Jones: I don't breadboard too much these days.

Chris Gammell: I mean, I'm doing most of my circuit designs at work and... Yeah. Yeah, I mean, I have a couple of things at home. It's not a problem. Yeah, right. Well, none of that, it's just having a breadboard at work is just... I don't know. Like, when you were at work, did you have breadboards around?

Dave Jones: I... We did have breadboards, but we rarely used them. Right. I mean, it's just a matter of... It's incredibly rare. Well, because these things often you're working on are not breadboardable. End of story. Right. You know? I mean, you have to lay out a board in order to plug it into a breadboard so you can access it. Yeah. Well, you may as well lay out your final board.

Chris Gammell: Right. No, yeah, exactly.

Dave Jones: No.

Chris Gammell: Or if you're having a service, you know, like sometimes you have, you know, people in your company that are doing lay out for you, you know, that can also... Yeah, that's another thing.

Dave Jones: You throw it over the wall and somebody else does it. Yeah, exactly. You know? Exactly. Wipe your hands of it. Yeah.

Chris Gammell: Interesting story I saw this week on Electronics Weekly. So it's LED, so it's not necessarily relevant to all electronics, but Osram, the LED maker...

Dave Jones: Yep.

Chris Gammell: They actually, they made pads on their LEDs so that they could cross reference against other LED manufacturers. And I love that idea. I love... So basically, they have... Maybe it was... Oh, no, they have standards. Sorry. So they basically gave away footprints so that you could second source their parts or competitor parts based on the difference in footprints, right? So you have, you know, Osram makes one type of LED with a footprint. And then, I don't know who else... Osram's who? Sylvania. That's Philips, right? So then Cree is the other one, right? Cree's another big one. That's right. So maybe Osram and Cree have different footprints, but they basically give you a footprint pattern that then you could put down so you can second source either one. And I love that idea. Yeah. I love... That's pretty cool. Yeah, exactly. So they give you that footprint and they've already done all the work for you. And basically, you just throw that down and then you can put down either one. I think that is a... I think we need that on every type of part, especially for... You know, and they designed their packaging footprint too to be more generic, right? Because if you have real specific, you know, like, oh, a little pin here, a big pin there, right? It starts to really throw off, you know, your design or your crossability, right? But say even you were going from like a, you know, a SOIC 8 to like an MSOP 8, right? I mean, like you could make a fancy footprint that could fit either one. And it's a little easier with LEDs because you only really need two pads. But I'd love to see that kind of thing for like higher pin count stuff. You know, I always want... I always... Whenever vendors ask, you know, like, oh, well, what kind of feedback do you have for us? Well, I want standard footprints. I want you guys to stop being d-bags about, you know, oh, well, this is our proprietary technology. No, I want to cross your part. Let your part play on price and let it play on how good it is.

Dave Jones: On performance and, yeah, everything else.

Chris Gammell: Yeah. Yeah, because it's only getting worse, right? I mean, we talked about a little bit before about not being able to, you know, breadboard something. I mean, hell, you can't even put down a second part on a tiny package layout these days because, you know, the footprints are also different, especially with QFMs.

Dave Jones: You know, and I find it every day when I'm laying out a board, I'll have to do a new footprint. Oh, yeah. That's when it's obvious. Because it's some manufacturers, you know, weird ass, you know, slightly different to, you know, oh, man.

Chris Gammell: Yeah. Now, there's a reason to like dip parts because it's a little more likely they'll be... It's at least a standard, you know, pin spacing. It might not be the same pin out, but it's at least the same pin spacing.

Dave Jones: We already have enough standard, you know, in terms of SO packages and, you know, but there's so many variations of the, you know, the small MSOP type. Yeah. You know, how many variations are there in just an eight pin, you know, dual in line surface mount, you know, package? I mean, there's just so many.

Chris Gammell: Right.

Dave Jones: Leadless, J-lead, you know, and the leadless carriers, there's probably three different types there in the leadless ones. And I said, oh, man, no. That's lame.

Chris Gammell: It is really lame.

Dave Jones: So I'm always laying out new footprints. Always. Yep. It's a pain in the ass.

Chris Gammell: It is. It is.

Dave Jones: Oh, man.

Chris Gammell: I saw someone actually sent this in. Speaking of footprints, it's actually a pretty cool video of like an old Motorola IC. Not an IC. It's a module, actually. They took old components, basically. And they basically created a module with like 100 mil spaced headers. And then they actually sealed the whole damn thing up. And then it was a pluggable module in like old school TVs. Like we're talking like tube style TVs. But they actually made modules for it. So it's pretty cool video of, you know, seeing how that goes in there. I mean, good luck trying to find one. But still. Yeah.

Dave Jones: Oh, it's all gunked up. I'm looking at the video now.

Chris Gammell: Yeah.

Dave Jones: It's all gunked up in brown stuff. What's?

Chris Gammell: Yeah. It's like sealed. And it's, you know, I'm sure they were doing it. That was probably security back in the day, too. Right? Right. Okay. Yeah. It's cool to see. I mean, like, so it's not a new concept to make, you know, make proprietary modules and stuff like that. I'm sure people complained about it back then, too. Right? Except back in the day, you were complaining that it wasn't a standard, you know, tube instead of a standard footprint. But still, it's cool to see that stuff. I mean, back in the day. Especially, like, the old TVs and stuff are cool, I think.

Dave Jones: But if you tried to open them up, you just open them up and they're pretty depressing, though. They might be cool in some aspects, but they're just old and crusty. Depressing? Yeah. Like, they just smell horrible and they, you know. Yeah.

Chris Gammell: That old burning dust smell, you know, like that. Yeah.

Dave Jones: They're just, ah. They're just, yeah.

Chris Gammell: Yeah.

Dave Jones: Let alone trying to desolder and re-solder them, you know. Oh, yeah. Have you ever tried to desolder old boards?

Chris Gammell: Yeah. Get your 100 watt iron out, right? Because they've got like a...

Dave Jones: And the smell.

Chris Gammell: Oh, the smell too. Oh, yeah.

Dave Jones: They've got a unique old, you know, old component smell to them.

Chris Gammell: 100% leaded solder and broken dreams, is that the smell? Right. Oh, boy. Yeah. Speaking of huge electronics, did you see Evil Mad Scientist? Did you see their post?

Dave Jones: No. What have they done?

Chris Gammell: They went to a swap meet and they found this old Bussman fuse. It's a liquid power fuse and it is just monstrous. So, you know, people know about fuses, right? Fuses are boring, right? I mean, they're just basically pieces of wire that, tiny pieces of wire that if you put too much current through them, they blow up, right?

Dave Jones: I'm looking at, this is almost a work of art.

Chris Gammell: I know, right? It's crazy. It's like, it looks like a baton it's so big. Yeah, yeah. But the specs are even crazier. I mean, that's the really crazy thing, right? 23 kilovolts at 15 amps. You want to get your calculator back out and figure out how much power that is? That's just like, that's like megawatts, right?

Dave Jones: The thing I'm looking at, though, is it's one big coil inside. How much inductance does this thing have? It's basically an inductor. Oh, yeah. It looks like it, right? In a tube of, yeah, it looks like an inductor.

Chris Gammell: Yeah.

Dave Jones: You know, wire wound, air spaced inductor in a tube full of oil.

Chris Gammell: Right.

Dave Jones: Like, holy crap.

Chris Gammell: Yeah. That's actually, so it's 350 kilowatts. That's how much power it can handle.

Dave Jones: And what's this used for?

Chris Gammell: I have no idea.

Dave Jones: Do they have a clue?

Chris Gammell: I'm guessing power. I mean, or arcing or something. I don't know. I didn't read. I was so amazed. I think they have a whole post on it, but it's pretty cool. I think they pulled, they didn't, no, they didn't get a chance to pull it apart, but it is crazy. So, if you should check it out. It's, I mean, yeah. Yeah. I've been looking at it.

Dave Jones: Oh, it's used on power poles and stuff by the looks of it, possibly.

Chris Gammell: Oh, yeah.

Dave Jones: Yeah.

Chris Gammell: Yeah.

Dave Jones: Huge ass stuff. Yeah. Big power engineering. Ugh, scares the crap out of me.

Chris Gammell: Yeah.

Dave Jones: Anything over 12-aust DC, high voltage, as far as I'm concerned. Really? Don't touch it. Yep. Yeah. Oh, and I'm a low power, low voltage man. Always have been. Yep.

Chris Gammell: Yeah. I've never had that benefit of that.

Dave Jones: Everyone asks me, oh, when are you going to do Tesla videos? I have no interest in that sort of stuff.

Chris Gammell: No? No? No, photonic induction? You're not going to follow on?

Dave Jones: No, no photonic induction. No, it scares the absolute crap out of me. I won't touch it. No? No?

Chris Gammell: Yeah.

Dave Jones: I hate working on live 240-volt stuff. I refuse to do it anytime, you know. 240 scares the crap out of me, too. It's better if it's possible.

Chris Gammell: 120, that just tingles. 240, that hurts.

Dave Jones: Nah. No thanks.

Chris Gammell: Yeah. Well, speaking of Tesla, did you see this? It's been a couple weeks. This has actually already been going on for a while. Yeah. Yeah. So, there's a society, basically, that has been trying to get a museum going for a while. But the backstory is, if people haven't heard this, I'm never sure how much our listeners have followed the news like I do. You know, it's tough to tell the difference. But, basically, the Wardenclyffe, or Wardenclyffe, or how do you say it? Basically, Tesla's old workshop, where he was going to transmit power into the ionosphere, then everybody could pull down free power from the ionosphere. Yeah. Which never worked. Yeah, exactly. But, basically, it was a brownfield because it was a film factory for a while after they sold it off. The tower's gone, but, basically, they got it all cleaned up. It's for sale for $1.6 million. And, basically, this foundation said if they get half the money, the state will kick in the other half, and then they can buy the property.

Dave Jones: How did they get that? How did they...

Chris Gammell: Like the grant or something? Yeah.

Dave Jones: Yeah. How did they get the government to agree to... I don't know.

Chris Gammell: I'm guessing it's probably... Going for the other half. Unloading... It's probably going to be unloading, like, a brownfield, right? I mean, I don't think I would want to hold on to a brownfield. Man. Yeah. So, oh, it's the Tesla Science Center. Okay. But, it's... So, basically, the guy... The Oatmeal. If people have never seen, it's, like, these really snarky, awesome comics. Yeah. So, Matt Inman, he got pulled in, I think, by Reddit. I think it was by Reddit. But, basically, he used his star power to... And, he had written a comic, basically, extolling the virtues of Tesla over Edison and calling Edison a bitch and everything. Which is a hilarious and true, you know, account of the differences between Tesla and Edison. And, we can link that in. It's actually quite good. But, basically, the whole idea was he was going to help with fundraising. And, as of today, it was a 45-day campaign on Indiegogo, which is, like, a Kickstarter. But, they allow charities. As of today, the $850,000 goal has been obliterated a while ago. We're 11 days in, and it's at $1.1 million. So, what happens there? The government, they don't get the government money? No, they will. They'll still get the government money. Oh, right. Any money over the $1.7 million for this campus will go towards actually building the museum.

Dave Jones: Oh, nice.

Chris Gammell: Yeah.

Dave Jones: So, the government will double it. So, anything they get, the government will double it. No, no, no.

Chris Gammell: The government's only kicking at $850,000. Oh, right. Okay. So, basically, they've raised $850,000 plus another $400,000, basically. All right. And... Oh, man. My math is terrible today. Another $250,000. And, basically, all the other stuff is going to go towards... You know, that was just for buying the land. And now they're going to actually be able to build buildings on it, you know, make exhibits and set up foundation... Or set up...

Dave Jones: I possibly think they don't realize what they're getting themselves in for here. Yeah. Because the government will want an organization. They'll want reports. They'll want... No, they do. They have all that stuff.

Chris Gammell: Oh, really? They have a 5013C. Yeah. It's a legit company. It's a legit charity already. Oh, okay. Oh, okay. Oh, okay. Right. This is for real. I mean, like, this all existed before Matt Inman got to it, basically. Right. So, but basically, you know, if people want to donate, you can, you know, get... Basically, it's just like any other, you know, Kickstarter campaign. You can get, like, stuff for it. You can get, you know, a photo of Tesla or you can get a bumper sticker. T-shirt or... T-shirt. Right, right, right. T-shirt sold out. You can get a baseball hat. But, you know, it's just a lot of fun stuff, right? I mean, but in the end, the cool thing is, you know, this might get Dave to the East Coast finally, right? I mean, if there's a Tesla museum, I'm sure you'd go to New York, right, Dave?

Dave Jones: I'm not a huge Tesla man. No. No? Oh. No, it's not that exciting. I'm sorry. I'm not going to fly all the way to bloody New York just for...

Chris Gammell: Ah, if you're out there for New York, just for... Make your fair New York one year. I'm sure you'd go to it.

Dave Jones: If I'm in the area, of course I would visit, yes. But, you know, I'm not going to go out there just for that.

Chris Gammell: Oh, just for that. No, I wouldn't expect that.

Dave Jones: Right.

Chris Gammell: But, yeah, I don't know. I think it's cool. I think it's great. I don't know. There's, yeah, there's always people saying, oh, we shouldn't do this. Yeah, man, man, government. You know, it's okay, whatever. At the end of the day, it's going to, you know, preserve a piece of historic property and it's going to be a museum. Yeah, no, I totally agree. Cool. Plus, Matt Inman's a funny guy. So, I enjoy it. I enjoy it a lot. But, so that's the oatmeal.com, I think. Or is it the oatmeal? Whatever. Yeah. We'll link it in. Our time's up. Is it? It is. Lost track. It's up. Well, people can find links and links that we didn't get to this week. That's a nice bonus there. And it's on a site that won't crash. At Reddit. Because it's not managed by you. Exactly. I am poison to all things internet. Reddit.com slash r slash the amp hour. And we'll put a link in for that too. And then Twitter and Facebook and everything else. Cool. All right, man. We'll see you next week with mystery guest number three.

Dave Jones: We think. Yeah.

Chris Gammell: See you then.

Dave Jones: Bye. Bye. Bye.

Speaker ?: Bye. Bye. Bye. Thank you.