AI and ChatGPT have been in the news about how it will change world views or will it be relegated, making sure NPCs in video games don’t repeat dialog?
How easy is it to make a retro gaming console? Stephen breaks down his design and build criteria that involves no custom PCBs.
Parker and Stephen chat about the recent Nintendo Switch hack and Stephen moving to Colorado.
Visit our Slack Channel and join the conversation in between episodes and please review us, wherever you listen (PodcastAddict, iTunes). It helps this show stay visible and helps new listeners find us.
Level shifters for the Gameboy VGA project. Uses the Ti SN74LVC8T245PWR.
Current state of the first project that was on the MEP. The Super Simple Power Supply.
Parker is an Electrical Engineer with backgrounds in Embedded System Design and Digital Signal Processing. He got his start in 2005 by hacking Nintendo consoles into portable gaming units. The following year he designed and produced an Atari 2600 video mod to allow the Atari to display a crisp, RF fuzz free picture on newer TVs. Over a thousand Atari video mods where produced by Parker from 2006 to 2011 and the mod is still made by other enthusiasts in the Atari community.
In 2006, Parker enrolled at The University of Texas at Austin as a Petroleum Engineer. After realizing electronics was his passion he switched majors in 2007 to Electrical and Computer Engineering. Following his previous background in making the Atari 2600 video mod, Parker decided to take more board layout classes and circuit design classes. Other areas of study include robotics, microcontroller theory and design, FPGA development with VHDL and Verilog, and image and signal processing with DSPs. In 2010, Parker won a Ti sponsored Launchpad programming and design contest that was held by the IEEE CS chapter at the University. Parker graduated with a BS in Electrical and Computer Engineering in the Spring of 2012.
In the Summer of 2012, Parker was hired on as an Electrical Engineer at Dynamic Perception to design and prototype new electronic products. Here, Parker learned about full product development cycles and honed his board layout skills. Seeing the difficulties in managing operations and FCC/CE compliance testing, Parker thought there had to be a better way for small electronic companies to get their product out in customer's hands.
Parker also runs the blog, longhornengineer.com, where he posts his personal projects, technical guides, and appnotes about board layout design and components.
Stephen Kraig began his electronics career by building musical oriented circuits in 2003. Stephen is an avid guitar player and, in his down time, manufactures audio electronics including guitar amplifiers, pedals, and pro audio gear. Stephen graduated with a BS in Electrical Engineering from Texas A&M University.
Special thanks to whixr over at Tymkrs for the intro and outro!
Hello and welcome to the macro fab engineering podcast. We're your hosts Parker Dolan and Steven Craig. This is episode 121. So before we start, like we do every week, we'll have the Twitter chat info stuff on May 25. This week, so it's same, same as every week, Friday at 1pm. Central time. Use the hashtag macro fab to join the conversation. Meetup info, which is actually this week, May 23, at 6pm at macro HQ in Houston, we'll have Brandon from particle and he is going to give a talk about particle and IoT and stuff like that. We're also going to be, you know, streaming it on Twitch for non Houstonians. It's going to be twitch.tv/macro fab, or just click the link below. And then we have the Houston hardware Happy Hour, which is the first Thursday of each month. It's this next one is June 7, at slug pokes so come by bring hacks, drink beer, drink coffee, eat food there. Stephen, unfortunately, will not be making it because he will be on his way to Colorado.
So actually, at that time, I will be in Colorado. Ah, okay. Yeah. So and I think Steven is going to try to put together a Denver Houston happy hour.
That's the plan right now. Yeah, I went to Denver, Houston. Happy Hour. Yeah, no. Hardware. Yeah, Denver hardware happy hour. So, you know, keep in touch with this podcast, because we'll have more information. Once I get settled in there. And I kind of figure out where's a good place to do it. I'm going to try to set up a Denver chapter.
And I'm hopefully we can like alternate the weeks. And so like, sometimes I can go up there. Yeah. And I'll be I will certainly be back here. Yeah, just go up there. Drink beer come back.
You know, flights are actually really cheap. Yeah. So for less than 100 bucks, you can go from Houston into downtown Denver. That includes the trip from the airport to downtown for less than 100 bucks. So yeah, and if you text Stephen, he can be your personal Uber? Sure. Find his number first. Yeah, you might find it on the bathroom stall at slow for a good time to call for an analog engineer. Yeah, I was about to be on the side of the road with a with a cardboard analog engineer hungry?
Feed burgers, well design boards for food. Okay, so see him? What do you been up to this past week,
you know, so with kind of like all the moving stuff going on, like I don't really have a lab available right now. So most of my stuff has been just like playing around on my computer or research kind of like what we were talking about last podcast where you just sit on the couch and research a lot of stuff. But I actually resurrected an old schematic that I was working on from like a year ago, for an guitar amplifier that I was just playing around with. I have like eight schematics that are just like kind of in flux with guitar amps that probably will never get built. It's more just like, I like drawing the schematic and like envisioning about what it will do kind of thing. And this is one of them that might actually come to fruition, we'll see. Maybe in the next couple months, I might do something on it. Regardless, I was just playing around with it. And I kind of came up with a specialty circuit like a like a utility circuit that plugs into the amplifier that I wanted to share. Just not because like there's anything super special going on, but because it might just be helpful to someone, it might be fun. So this is actually the switching circuit in the amplifier. So this amplifier has two channels that are switchable via either a switch on the front panel or a footswitch. Okay, so when you're playing your guitar, you can hit the footswitch with your foot, and the amplifier will rearrange its internals to sound different. So if you actually go to Mac fab.com/podcast, and look for Episode 121, in the podcast notes will have a PDF of this schematic so you can follow along if you want to. So this schematic I've got it kind of sectioned off where the top section is some transistor logic. And then the bottom section is some kind of digital logic and 555 stuff. So the top section is basically all it does is make sure that LEDs on the front panel switch properly to whatever channel so the looking at him like what this is interesting. Yeah, it's basically just like a whole bunch of high low transition LED on the foot switch to it. Yeah, so that the foot switch side is actually a separate box. I was confused about that in a little bit. And then I realized, Oh, she drew that ground symbol. And like, why is that there for footswitch and I'm like, Oh, you're just referencing that it's ground and it goes to the connector that's labeled footswitch one, right, right. Yeah, and The footswitch and footswitch one, that's my schematic symbol for our coax cable God and that's the coax that connects down to the coax only has two pins. Yeah, or two wires shield in center. Yeah, and, and so there's an LED in the footswitch. But if you if all you have is just two conductors, all you can do is pass current through it. And that's it. Yeah. So in order to switch channels, you just have to short those two, which, if you short those two, then it turns the LED off. But if I do that, then I need it to cascade and do all the other stuff down the line. So you know, this is this is something that just threw together. Actually, I haven't built this, but it's kind of simple. So I bet it would work. Regardless, what I kind of wanted to go over is the whole top section, basically, all it is, is if you see over on the right side of said signal in there's like a bus for signaling. That's that's the footswitch signal. So if if you're on one channel that is pulled low to zero, and if you're on the other channel, it goes up to five, gotcha. Really Well, what I wanted to show is what's underneath that is that kind of all that digital stuff. So what I created here was a circuit that was a dual edge detector. So if you see a transition from zero to five volts, or if you see a transition from five to zero, so going either direction, what will happen? positive edge or negative edge? That's right. Yeah. So you nine is an XOR gate, the exclusive OR gate, which the an exclusive OR will have an output high if the if the two inputs are opposite of each other. If the two inputs are the same, you know, one, one or 00, then it's low. So what happens in this like first section here with this XOR gate, one of the lines to this XOR gate has four inverters, which high low, high low? It basically just is a delay delay. Yeah, yeah. So what happens is when that signal in goes from zero to five, or five to 01, of the legs of the XOR gate, is slow, slightly slower than any other kind of gate technology. That is it's like, I'm gonna bet you nanoseconds. Yeah, it's gonna be like, it's probably CMOS. Is it TTL CMOS? Or what? TTL CMOS? Probably, those are usually like, this is all 7485 volt. Yep. So it's probably it's been a long time since those data sheets, but it's like, what 12 To 20 nanoseconds per gate, I believe, if I remember, right. In fact, ya know, the, these, this specific hex inverter that I chose here is 20 nanosecond. So there's about 80 nanoseconds worth of delay, right there. So what that does is whenever you get either transition, that XOR gate is going to go high for 80 nanoseconds and then go low. So one thing I can see here, is I want to see if your math is right on your delays, is what got your hex inverter and see how the gate speed because then we'll have a graph board and how the gate see speed changes with with temperature. Oh, yeah. And then see, if you're running into a condition where that input changes before that, that XOR gate can change.
That, that'd be interesting. I'll have to I'll have to add nanoseconds is enough for that. XOR gate or not.
I've seen this implementation with fewer than four M hex inverters. I put four, mainly because I had a six. You want to play it in Antarctica. That's true. Yeah. Like I hadn't I hadn't thought about doing this over temperature. We'll have to see that part. I mean, the reality is, this is going to be inside of a box that actually gets hot. Yeah, so probably, okay, you put it if you
put this thing in, like, in Antarctica and got it down to like absolute zero, it would never work because it just seems
like it's perfectly the hex inverters are invisible. Yeah. To time. Okay, so So yeah, so basically, this XOR gate, sends out a positive pulse anytime you switch channels. So then I throw that through another inverter to make that normally high, and then it sends out a low pulse, which triggers a 555 timer. So it's a signal in appears zero volts clean. Yep. What's odd overdrive? Not overly dirty. It's the same thing. Yeah, so there's a, they call it a dirt or a distortion or an overdrive channel. Yeah. So yeah, the OD is just shorthand for it. So yeah, shall I send that through into that 555 timer, and the five five timer if you see that pin, three in the output, it goes off to a mute pulse. So basically, what I have is a J FET. That's connected right at the output of my preamp. And this J FET. A J FET. Normally has really, really really high resistance between its pins until you send it the appropriate pulse based off of it's a P or an N style. Yep. And it basically goes, it acts as like a variable resistor with very low resistance. So this mute pulse will short out that J FET. And short out the exit of the end of the preamp. And basically, any pop that goes from all the relays and other things that are switching in my preamp, if there is a pop, it gets shorted to ground for a very short period of time. And that period of time is actually adjustable by the RS 17, in this schematic here, so that's, that's your, whatever the pulse is runtime, right? So yeah, so I have a I have a very specific pulse for the XOR gate. But I have a variable pulse for the five, five, so I can tweak it once I'm done. And actually, technically, I probably don't even need the 555 timer. Because what you can do is the for transition line gates, you could replace that with just an RC filter. Yeah, I was actually thinking like an RC filter with a trim pot. Yep. And you could tweak that. And then when it hits the threshold voltage of that XOR gate, that's for inverter, whatever, in reality, it would just be a trim pot where I just turn it until I don't hear a pop anymore. Yeah. But the reason why I went with this 555 timer is because it's just like an extra 20 cents. Yeah, and I don't know how long of a pulse I need. And the 555 timers just very, very explicit. So I just want with that. So regardless, this whole, the entirety of what you see here is just here to be able to accept a low to high transition or a high to low transition, and to mute my preamp for a very specific amount of time. In either way. Yeah. So I thought I might just share that because it's a it's a cool little tack on circuit that gives you a little bit of utility in here. It's interesting that this is your solution to that. And I would just stick an FM eight microcontroller on there. There's a there's a reason why. So why did you choose this? Uh, first of all, I like doing this. It's fun. Well, but so utility. I've done microcontrollers, multiple times in guitar amps. And I haven't done this. So like, Okay, I'm going backwards in technology, just out of fun. I've done the whole microcontroller thing multiple times, I've done I've done the thing where it's like, it's smart, it knows where it is, it knows what it's doing it, you know, it rejects bad things that you know, I even have some more like a microcontrollers in charge of the high voltage power where it'll like slowly bring things up. Or if you change states, it won't just like jump between two high voltages, it'll ramp down between them. I've done stuff like that, where there's a lot of smarts in it, but I was like, I just want to do something in this style. But when I was doing this, I actually had a question that came to mind. And I want, I want to ask you this question. Because this is a little bit. I don't know the answer to it. I have an idea, but I don't know the answer. So I think you could make a pretty decent argument that this is a goes into an analog circuit. And this is sort of analog II Yeah. Right. So sort of, but not because I'm using digital chips that are like, you know, gate level logic chips that go in this. So is this part digital? Is this mostly analog? Like, where's the boundary between that like, what, what would I have to do for this to be considered digital or so?
In my mind, all that digital means is there's thresholds. Okay, so when you hit a threshold, something happens got to lower something else.
And so if it can only exist in two states, yeah, then it's digital. Yeah. And actually looking at your your transistors up there, you using them as on offs? That this whole thing is digital, this whole thing is digital. Yeah. Okay. Because like, the notes, you're using two and three nano fours.
I'm like, Ah, the transistors are digital people use because we just use it on and off.
Yeah, it doesn't have much of it doesn't have much of a linear range to it. That's like the transistor, the default transistor Yeah, to
power. Like, I'm doing a high side p channel. MOSFET. Yep, that's all I used to turn it off and on from my controller, IPS, or like, on turn it off a 12 volt signal. I'm like, Well, I can't use a microcontroller pin, cuz it's only 3.3 volts. And you can't drive that p channel up high enough. So you just use a two and 394.
And if you need a PNP it's a 3906. Yes, yeah. And in fact, a lot of reason why I use 39 and fours and 39 sixes is because in the very changed well, they're cheap. But in the very first simulation software that I got the only two transistors that were available to read those two, so like, they're just sort of my defaults, like, I have like a BC 555 and 55756. I can't remember like those are more, I guess, analog II and more like amplifiers he like however you wanted to find them, but Like, if I'm doing switching stuff, and I just want a generic Jelly Bean PNP or a BJT, it's, it's going to be 3904. The thing though, is that if you put this together and put it inside your amp, people will be like, Oh, it's cool. If you just put a few FN 20 in there people like What the hell's that? Oh, yeah, it's gonna sound terrible. Yeah. No, it does nothing besides just look at two signals and output another signal. Yeah, that's right. That's right, exactly. And so the thing is, everything that you see on this page, there's no, there's no like, transition between anything. I mean, there's, there's, what I mean by that is, this exists, everything on this page between zero and five volts, nothing in between 05 volts, with some exceptions, like the LEDs hold the basis of the transistors at a slightly different voltage.
Actually, no, there is one analog section here. What's your, your discharge on your 5.5?
Uh, I mean, the discharge, I mean, you got a RC circuit there, I guess. It's ramping until it hits that it's cut off, and then it shorted out. That's what it does. So I've got this set up as what do they there's three ways to set up a 555. Timer. There's the a stable, the Moto stable. And the there's a third one, I can't remember. This is the one shot pulse slingshot poles. Yeah, yeah. So you have to reset the 555 for it to be able to do another pulse. But the reset happens as soon as your input pulse goes back north. So I trigger on the on the pulse down, which is, you know, from the XOR gate, and then it gets reset at nanoseconds later. So I mean, I've got this set up as a as a one shot. And the output supposed to be very sharp. I want it to be sharp, because I want the mute to be off and then off. Yeah, as fast as you can be. Because if you don't want to sawtooth on there, well, technically, technically, you could, but at some point you would actually hear you would hear that sound good. What back as the J FET. Kind of like somebody close. It's like a soft start thing. You have a walk. The Walk trimmer, yeah. Yeah. You know, I have actually tried that once, too, to make, because you don't want Okay, so you don't want the J FET. Opening too fast. Because if it opens too fast, it can send a pulse. I'll pop down the line. So you know that there is a little bit of like trimming available there. But but that's like an extreme case i I've actually never really run into that. And I've implemented something similar to this before with a PIC microcontroller actually. So I thought I might share that it's a fun little circuit for the slow ramp ramp, and you can just stick up cap on the on the input of that J FET. The gate, or it's equivalent to the gate. Yeah, yeah, you can. Yeah. So as long as you don't try to hammer too fast for the PWM or anything, you know, okay. I actually this brings up a little bit of a slightly separate topic. BJTs J fits moss FETs. All like the ones that I feel like I have, like memorized or BJTs. Like, I know how a BJT works on MOSFET as well. I know. J FET. J fits and J fits or depletion devices. Yep. But MOSFET. You can buy enhancement or depletion. Yep. So they work opposite of each other. And then you within each one of those, you have n and p. And it's one of those things where like, every time I think I remember it, I always like, oh, I need an enhancement and type of a MOSFET. And then I go on like, like, do I really need that? Or do I need a depletion stuff? Like, like, I'm always I don't know, it always just gets me confused. I always have to, like, look it up and be like, oh, yeah, this one does this, like a I think? Yeah, J fit is a normally on device, and you have to deplete it in order to turn it off. And so you have to have it at voltage basically. I don't know, it's goofy. Regardless.
That's a random day. BJTs are the only ones I was like, I got this exact opposite, like BJTs. I just don't play around with them enough. I'm like, driving saturation.
Yeah, actually, funny enough. I was designing some other circuits the other day. And that's one thing I really liked about BJTs is because you pretty much know what the base voltage is VBE it changes with current, it changes with temperature, but it's graphed out. And well yeah. But like in your calculations, you know, like the value to use. If you go look up practically any small signal MOSFET it's going to give you a VGS which is the turn on threshold voltage. And if you're lucky and might give you a typical value, most of the time they just give you here's what we think it will be max, you know, and it'll be like four volts. Yeah, but it might be two and
the trick is you just have to look at you Look, first of all signal stuff if you're doing like microcontroller stuff, because that's what's going to turn on around 3.3 volts. But then you look at that VGS. And then you look into they have a chart where they say, depending on the drain source voltage, what the VGS current can be, or like, at a certain VGS. What the current thing?
Yes. Yeah. So because you shouldn't have energy as well, there shouldn't IG you have a little bit noise leakage. But it should be in good ratably Tiny. I mean, it's basically your signals going into a small piece of glass. Inside the chair actually makes you start PWM. I mean, it actually makes a lot of problems. Well, it's a capacitor. Yeah. So yeah, if you're sending a an AC signal to it, you got different things. But if, if you're just switching it on, it should just
be it's a solid 100% duty cycle then yeah. I like I like how power to you is 100% PWM.
Ground is just zero PWM right, gotcha. Oh, we're nerds. Okay. Yeah, okay, so what you been up to par and I got it. Okay, I'm gonna pull it high. 200% duty cycle. Oh, okay. So you're gonna stop pulling it low. Every once in a while? Yeah, yeah. You just gonna always pull it high.
Okay, so I've been working on the Gameboy, a bit. I've been updating my blog, as I said, I would be so it's got two posts on there now.
I said I was gonna start posting more. You want a beer? Thank you We pointed that he spoke on the Gameboy design. I added level shifters to it because I had to convert the five volts on the Gameboy death 3.3 volts. I'm using the chip I like a lot, which is the SN seven, four LVC eight t 245. I think it's these.
So think about that. When you go to bed. Don't Gee, I'm like, I like that. Yeah, yeah. The chip is great. I think it's the T soft package I'm using I can't remember. I think it's P WR
is what T eyes extension is.
Does it have like a big pad underneath it? No. Okay. Now, these aren't like power. Power Level shifters or like just to do like digital logic. And this is a TTL stuff, right? Yeah. Well, depends. It's converting like, five volts. Level shift. Yeah. Yeah. So it's inherently, one side of it might be DTI. And actually it depends. souks like you, you give it. What that side voltages. So like the five volt side, it's getting five volts. And so it knows the that's what's getting. To choose your own adventure. Yeah, choose your own adventure level shifter. And on the other side, turn to page 73 For fun, so you can put whatever voltage you want, you can go up to 5.5 volts, I think it's what it is all the way down to like point eight. Oh, that's cool, or is it 1.2? Is it isolated? Or is it just straight pastor it's, it's not isolated, okay. And then it's an eight bit level shifter. And it's bi directional. So if you put a one on one pin and then put a zero, it'll shift it around. But thing is you have to use all eight one way 108 The other way, you can't just individually turn but you can dynamically change, you can change that. And that's got an output enable. So the output enable and the direction is controlled by the FPGA and I have those pulled up and down in the output enable. active low and so I haven't pulled up. Yeah, and then 100% I didn't have it I had the direction pins by default pulled up. No, they're pulled down so that it's always inputted into the FPGA so that way when FPGA boots up, because it takes time for it to configure itself that the Gameboy doesn't freak out. Oh, yeah, because I want it to be like the Gameboy doesn't know it's there while the FPGA is booting up. Gotcha. And so I'm using two of these because the there's like five or six lines I need from the LCD to do display. And then I wanted to read buttons and the buttons are matrix and so I needed to pull the entire matrix bus and I put that on a separate bus so that way, I can actually pull in the data and like I read a certain button combo, it the FPGA can actually drive the Gameboy to do a button combo. So you can like do like, you know, pre recorded button puts and stuff might not get that far in the process. But that's getting there.
Let's see, I'm going to do a, we'll talk about this a little bit the flat flex cable, because it's really hard to find this 21 pin point five millimeter pitch connector. And so I'm thinking I'm just going to make a flat flex cable that kind of just solid, flat flex PCB PCB. Yeah. And so one will just plug into the Gameboy the other one plugs into the screen board. And then those just get soldered onto the, onto the FPGA board. And so I'm going to document that on my blog of like, how I designed that, because I don't have any idea if someone just read some, I don't know where I'm gonna get built yet. So I got to find a manufacturer and be like, Okay, I'm going to use these guys, and then read how to do it and post all that how to do it, because I've never done it for checkout
the jlcpcb. Yeah, they, so I can't remember the name of the place. Easy, easy. EDA might have been them. They use JLC PCB for a while for their PCB services. And they've decided to break off and not do PCBs, but JLC offers really competitive price PCBs, and I think they do flat flags. And they're cheap.
Okay, they're real cheap. So I looked in OSHPark. Doesn't do flat flex yet. Yeah, purple flex, purple flex.
Hey, OSHPark. We got that trademark. Yeah. Flex. If you want that, you'll have to give us free PCBs for life.
I'd be fine with beer, or beer. And you can use flat flex TM, a purple flex to purple
flakes. Yeah. And then so what's left to do is that cable. And so I'm going to try to work on that cable this weekend, because it's Memorial Day weekend. So I have three days of like, I could do whatever I wanted to do. So I'm gonna do that.
And Parker spends it designing flat flex PCBs. That's right. It's gonna be like, sunny, beautiful, like 70 degrees out and be like, close the shade.
Close the shades and then you put shades on the screens too bright. Yeah, and but like weird electronic music in the background?
is gonna be like that. You play the music. That's the soundtrack human music from Rick and Morty.
Yeah, yeah. Great episode. The so what's left to do is that the VGA connector, so I got find one, design it, put it in, and then do the three, three 3.3 volt TTL to analog for the VGA. I'm going to do that both resistor network because that's what currently works on my FPGA and it's only like, what 12 bit color. So four bits per color. So the color is dependent upon the tolerance of your resistors well, you get like 1% But yeah, I mean, could you make it like, kind of like grainy and nasty if you use different Yeah, tolerance, or you just lowered the how many bits you're using? Hmm. Because like, I wonder if you got like really, really high tempo resistors and flex they Yeah, like you heat them up and like the screen would work or the colors would? I would Yeah, the colors would because the screen is fixed or the position. Here's your frequencies fixed because there's a digital signal but your colors, yes, that would be cool. There you go. It's like circuit bending for VGA with a heat gun
heat. And nothing I'm going to add is I want to add a serial terminal to it. Which I'm just gonna put like a ft 230 X FTDI chip on there, because I've never actually done that. On FPGA before I've done all the time, microcontroller worlds. I've never done that on FPGA. Oh, so
you just you've always done it through whatever other programming method Oh no, though. You always program them with whatever dongle like Altera uses USB blasters the brand. Can I ask a question? Why why is it called a dongle like why why is that the word like we get made fun of enough engineers like why did we have to go with the word dongle? Like ask 60s engineers? Yeah, thanks. Yeah. Okay, there we go. side tangent. Okay, continue. Yeah, so you played with your dongle to get the FPGAs working?
Yeah, so I've never implemented a serial terminal. Mainly I want for debug information. Because as I've been developing this, like you can input you can have a, like embedded signal analyzer like a like a digital logic analyzer, you can embed one in they have different names for what they are in their programs and stuff, but you can just add that in which is fine, except that it if they're always kind of clunky to use, and they're not perfect. But it's really the only real way to see what's going on inside the FPGA, but I'd like I want to just report like button states, I think it'd be better to be able to do that at run time, like as it's working through the serial terminal.
So I'm gonna try to implement a, you know, a, you know, a serial terminal TTL logic thing. And button reporter do like, you know, 9600 baud, stuff like that. Really simple eight, and one can go way faster than that I know, you know, real fast if I don't need to. And there's like, I've looked online really quickly before the podcasts and just like tons of like, I can just download something and make it work. But I think I'm actually gonna try to implement it on my own to see all the way down the gate level. Well, this gate level, yeah. Take this register is shifted out. Pretty much. Yeah, cool. I've implemented like level ship net level shows. I've implemented like, shift registers and stuff, like a 559595 740, C 595. I've implemented those into FPGAs. and stuff. And it's pretty easy. I don't expect that to be too difficult to do. It's just a never done before. So Courtright. Cool. Like, that's how the on my first pinball machine ever designed the reset vector, the screen ran off an FPGA, and it was that big LED matrix. And it actually the parallax propeller, which was driving all the code and like the audio, video, and, and the solenoids, and stuff, it actually looked at it, basically, I used my son for AC 5.5 driver, and just expanded it by how many ever bites it was that the display was, just, instead of doing eight bits at a time, it was doing like 1000, some odd. And so this mixin, expanded simple, you see 595. And so it works, you know, a slow bit works, I think, I think I mentioned this before, but there's a there's a company, I'm probably not at liberty to say their name, but I spoke to a handful of their engineers about some products that they design. And on this product, they design they had, there's all kinds of LEDs, like tons of LEDs across the front of it. And they they confided in me that every single one of those LEDs is on one bitstream from like just a string of five, nine fives. Yeah. And so when when they need to update one LED, they update, like 150 lines, and then just like hammer them all at once. And if you gotta if you've got a processor fast enough, you won't see it, I mean, it, you don't really have to update it that fast. And 150 bits is not that much real. It's not an you don't have to really worry about that because it doesn't actually flip to next register until you hit the latch line. So you can be really slow clocking out all that stuff. And then you just hit the latch, and then it flips them. All right, right. But But I guess like, if a user presses this button, well activates 100 and 22nd LED, I can see that it has to pump all of them out. And you know, actually, that's something Okay, so I don't know this about 595. I can assume, like if Okay, let's see, you know, the first 5595 in line. Yep. Let's say it's 111000. Yeah. If you give it that exact same code, does it do anything on its output? Or does it look at its output and say, Hey, you gave me the same code. I'm not changing. It doesn't do that. Okay. When you give it a string, and then you hit or string every year, you get a surprise. If you give it a bite, you're gonna bite it and hit latch. It's gonna change it's going to trip this register over, it's gonna shift its Register Now what is the what a change but if you give it the exact same bite that it currently has, yeah, Does anything happen on its outputs? Does it Yes, glitch?
No, well, no, what actually will happen is it will out so if you give it let's say you give it a a character, so
we're talking hex, okay, okay, you got an A, and so you have downline, you have more shift registers, but you clocked in and a again, so you won't put an eight bits and hit latch. It actually will push that a down the line to the next ones. And then you have whatever is down here and then a
right right yeah, um, cuz you can you can string them yes, in series, but Okay, so here's the thing. What if you send eight ones to a to it, you latch it and then you send it another eight ones and latch it? Does anything happen to it will not glitch does it? So I'm wondering, does the in input or does the does inside the 595 Does it have like an anti glitch thing or does it just like guarantee no glitch is fast enough? Just fast enough. I like that. Yeah. It doesn't like go down to zero and then back up to one quickly. Okay. Cool. I'm just guessing there, but it's probably fast. I've never seen that. So Just something I don't know, we need something that's fast enough to record that, like couple nanoseconds. Yeah, it's pretty fast. Cool. I bet you if we looked at the internal block diagram, it probably however, they designed the buffer that's there. It if you have a one on the output and you have a one on the input of this buffer, it probably doesn't look that output. That would make sense. Yeah. Yeah. In fact, I wonder if it actually, like, looks at feedback from its output. No, I wouldn't do that. No, I just, yeah. Okay. And that it doesn't have open collectors or open drains on the output. Doesn't it drives? He drives? Okay. Yeah. So it's a totem pole kind of thing. Yeah. Okay.
And then the mercury. Oh, yeah. This thing? Yeah. I've been working on the the brewery wine. I sent you the drawing.
Well, you posted on Twitter, too. Yeah. It's it's gorgeous. I love it. It looks like an appliance. Steiger. It's it's great. Yeah. If you haven't seen it, go to Parker's Twitter, which is at Longhorn engineer. And go check it, check out the image. It's great. It's fantastic. Because it's just like, it's it's hand drawn, you can tell it's just a guy who was like figuring it out as he goes. I'm like, I need a breaker there. I drew a box over the line breaker. But it's like, but it's one of those things where it's like, every engineer has done what you just did at one point in time where they're like, You know what the end result is? Yeah. And you kind of know the path in between it. But you still have to like define define it? Yes, I did. I drew that out. So I knew what kind of breakers I needed to buy, right? And what Wire Gauge I needed to spec because there's some parts of the circuit that has 50 amps, they can carry some that have 30 and then have one so I'm like, Okay, what kind of gauge I need. So actually, in what's funny, but and this is this is true of any schematic. All wires in every schematic are the same. Yes. But if you look at the end result, you're going to have some skinny wires, and you're gonna have some big monster fat. Look, I drew like numbers, what amperage? Each one was like, okay, these need to be 30. So yeah, and then I went online, and I used like, normal electrician
to code in quotes. What you spec? So like, the 50? Amp 246 gauge? I have no. Since it's technically chassis wiring, I can get away with like, a lot smaller. But I'm like, whatever. Let's just do what they put in walls, because
I have space for it. Right? Because if isn't chassis wiring or like the, quote, proper wiring, that's that that assumes a certain temperature rise, right? Yep. Based off of the maximum current flows through it. And then to code is almost like the current capacity double. Yeah, almost double is like a six gauge wire can handle like 96 amps. So when you put it in a wall, you assume no current flow. And so you have to spec something that can't dissipate heat, basically. You know, a buddy of mine was talking to me about it. Correct. But I'm guessing that's why I'm putting an asterisk right there. So this guy and his wife bought a house not that long ago. And they went the route where they bought like a really old house. It was built in like the 20s or something like that is a really cute house. It's a nice house and stuff. But he had been lamenting because he's having to fix the wiring in it. And he's like, I don't know where half the crap goes. He's like isolated one circuit. And I cut the breaker to it, and then went and tested it. And there were 20 volts on it. And he's just like, Where the hell is this coming from? And he said, his whole house is like that. I was like, Dude, you need to change that. Like now, before it burns down. We got we have these rules and regulations up for a reason. Well, that house was probably encode in the 20s. Yeah, yeah. Yeah. 20 volts 20. There we go. Yeah. Like, where is it leaking from that it's dumping 100 volts, you know? Yeah, that's
actually probably leaking from some kind of did this in the 70s or 80s, low voltage switching for relays?
Like, like, what through the mains or something? No. So you would it would hit a transformer, and it would drop down like 20 volts, and then it would go to a soft switch that you could press and that would turn a relay that would turn on the high voltage somewhere. Oh, okay. That was actually kind of popular in the 70s. I think 70s and 80s. Really? Yeah, that's insane. So there was a code. It doesn't exist anymore. I have not seen that in a newer house. And yet the house I rented in college had one circuit that was like that, but it's kind of weird that it's like back feeding.
So we had had an extra step down in the transformer for this soft Yeah. So you would run the 20 volt into your wall. That is So the switches were like a foot switch. There were like a little press button. Hmm. Yeah, that's weird. I don't know why they use 20 volts for that, because you could run 120 to it and have a switch that was designed for it. I don't know why they chose 20 volts.
It's almost like a doorbell circuit. Yeah. Because that's like 18 volts, I think, is normal circuits.
It's funny when you when an industry gets locked into something, and they're just like, we're not gonna change. I bet you. This is a huge, a doorbell transformer, which is 18 to 20 volts. Yeah. And that's what they used for it. But it's like, why? Well, maybe this guy's house was like the proto like test for that. And he has like the very first one of
oh, he's got the doorbell wires, like bridging aid or something that's quite a function,
or there's like a slightly conductive piece of wood that is dumping 100 volts across it, and he's actually getting 20 outlet. That's way more scary. Yeah. That's why I was like, Dude, you just need to kill the breaker rewire the whole damn house.
Everything. So yeah. So I've been up to, I guess, growing our Arvo. And so I think we already did your question, right?
Oh, yeah. Yeah, the question was the, where's the boundary between analog and digital? And I think that going back to the default is, if your circuit only exists in two states, then it's digital digital. There we go. What if you had three states? Because digital technically is three states. Well, you get you got your Schmitt triggers, which are they have a high and low transition. So that is, that is a tri state thing. Because then you have high Z, which is why I was talking about inputs, you actually have multiple seats on the output. Also, if you consider it that way. Okay, let's not open that guy. There we go. That was my question done.
If you have thresholding, because you can do technically, you could do a ADC that
has a threshold. So it says circuit analog, but your software is treating as digital.
Well, that's but that's like a software trip point. Yeah. So that would be that would be soft digital. Yeah. soft edge. Digi soft. Did you see him? If anyone you want to use that? Another six bit six pack? Yeah. Wait, wait, wait, wait. Let's just make sure we got this right. We have Digi soft and we have purple flex. Okay, we need at least one more by the end of the podcast. Yeah. All right. What's the next RFO? It's a Reddit article that was on the Electrical and Computer Engineering subreddit. It was some we've done this kind of before button. I think we could cover it in because it's the right time of the year is summer project ideas for a senior in computer engineering, huh?
So I would say if you have if you're a if you're an engineer, more, you're not an engineer yet. But if you're training to be one electrical engineer, and you have not designed a circuit board yet spend your summer designing circuit board.
That's a good idea. I mean, the thing is, it's senior in computer engineering. That's not electrical engineering. Okay, what but but so yet design but but a little bit more specific. Yes. designing a circuit board with an embedded system of some sort. Yeah, that would be that would make it like it make it easier. It says it is this is probably your first design. Make it an add on to an Arduino. Actually, okay, so yes, make it an add on to an Arduino but have a stretch goal and make it have something that's not Arduino on it? Like okay, in other words, make it make it something where you have to learn the non Arduino way in order to make it work with Arduino because then you learn both Yeah, next. Yeah. And in fact, have it have the same worlds but different ways of doing it have one AVR talk to an Arduino and like have them like pass information of some sort. Yeah, I can see that. I'll say just build a shield that plugs into an Arduino because you will learn so much from that. And then the next step I would say is incorporate Arduino onto your board just just consume it it's no longer designing the the microcontroller side and do this design number two
and my third suggestion is use macro fab to build it Yeah. Sucker plug right there. Yeah, that was that was a gut punch. If you like red PCBs, I know where you can get some
how many specific projects like an idea winning because one was generic just design a PCB. What about like a actual project? Hmm. Weather Station?
Well, yeah, weather weather station is that's good. Actually, summer is growing season for a lot of plants. So plants go into summer, and not like taters and stuff like that, okay. So make make a make a like, make mixing, make something that's like an auto watering system that like measures the humidity and not the humidity, like the dampness of soil, and will like trigger on your your watering system, something like that. And you could even do it or it's like, a, you know, it could, you could have a light sensor on it. So it knows, hey, I'm, you know, it's dark, if I run it now, or if I run it now then I'll get mushrooms. So don't do that like, or if you want mushrooms then do
so you know, there you go, there's, there's a, there's a useful thing to do. I really want to see that, like one like a sprinkler product. And it's got a mushroom button on it.
Mushroom mode. It's just it's not didn't say mushroom or anything. It's just a symbol of a mushroom. So it's a Yeah, with like a little LED that lights up once in the aged, do it. It's like the
bagel pin. Or you could you could even do something that like doesn't have like a specific use, but come up with a unique way of counting the number of cars that drive in front of your house on your on that road. And like do statistics throughout the entire summer to find all that out. Like if you did a project, it's actually really good for a computer engineer. Exactly. It's more software, that would be more software. And then if you if you wanted to do an interview and be like, Look, I created the hardware. And I analyzed all the data to show like what a good driving times around my home, you know, I don't know something like or even like, go a little bit further and make like nodes and have your neighborhood connected to nodes. And you can see like, what's like, what's the busiest street in my neighborhood? At what time throughout the whole summer or something like that? Yeah, I wonder if you can use, I'm thinking you can easily use like, you know, a camera and do image processing and figure out if there's a car. So I've done that before. That's actually that's not too hard of a problem on the on the hardware side, because that's a Raspberry Pi with a with a camera. And so that's a software problem. But on the hardware problem, you could probably try to do with accelerometer, and then have the device sit on concrete. I was almost thinking like, put it up against the curb. So it's perpendicular to the road. I'm measuring vibration. Oh, okay, I see what you're doing. I don't know, if you can get a mens accelerometer. It's sensitive enough to detect cars far enough away, I've actually done a decent amount of research on MEMS accelerometers. And for the most part, they're actually not very sensitive. Yeah, it's like it can tell if your phone is 90 degrees. And that's it. Yeah, I'm thinking some degrees, but I don't know about vibration can pick up something that light of a vibration? Well, I mean, my first job was a vibration analysis engineer. And we tried to get, I mean, a lot of a lot of times we were going down to vibration levels in the like, less than a millimeter, you know, worth of or very low G levels. And most MEMS were like more towards like crashing rocks kind of
sensitivity. Yeah. What about a piece? Oh, and then you measure the voltage off good. And she glued a piece down to the concrete? Yeah, I bet you could do that. Or that's a pretty sense, especially if you get a big one. You can probably get a good reading on that.
Yeah. What did you mean, like a really robust cable that was like a giant inductor, and you applied a voltage to that. So it had a magnetic field, or you had a field around it effectively. So when a car went over it, it disturbed the field. And then you could read that is that how the ones that they use they actually built make those? Actually, I
think that those are just a pressure tube. I think it's a pressure because I was think you just said that. I'm like, what if you just had a garden hose filled with water?
Just measure the pressure on pressure transducer? Yeah, so you ran over the pressure spike? Yeah, the the, the the difference is, if you did, if you did a magnetic whatever, then you would get one signature per car. But if you did a pressure tube, you'd have to identify how many wheels went over it. And what if Okay, what if a motorcycle went over it? And then what if trade Germans like big truck went over? You get different signatures, but they both give you a magnetic trail.
You could do a like like, like a funeral progression with your neighbor.
You can do is you can do like how many pulses in a random circus just goes through lot of the problem with the the transducer with the PISA to the whole thing. Yeah, like there's elephants Yeah, see, if you do a project, you have to think about if a circus is gonna go to your neighbor. And that might be where the strong point of doing visual identification, right?
Well and in in is really for computer guys than doing vision? You know, coding was Yeah, pretty good. Yeah. And especially since you can, like, you can see if it's an elephant or a car, right? Yeah. So so the vision, the vision option might be actually more difficult than, say, the the pressure tube because that would be really easy. Yeah. It's got different, it the pressure tube and or magnetic field or whatever, has hardware problems. But now law software problems. Whereas the vision is the exact opposite, you know, actually come to think about this is going through my mind right now, the pressure tube thing, you could actually probably do it without even a pressure sensor. Because you could have a diaphragm that would just complete a circuit by like, hammering it into well, no DC border. So actually, what you can do is you can do without any code is you can have a, like a shift register, if you know what you This is how I would do is you'd have the transducer, and then you'd have to figure out whatever, you know, drive a car over it and figure out how much voltage that transducer goes up to on its output. And then you have that go into an op amp and trigger a pulse when it goes above that threshold. And that pulls you either amplifier it or whatever you need to and put it into a counter. Yeah, and and then just read that counter. Yeah, eventually, it's a logical mechanical counter that you put into. There's like, I can't remember exactly what they're called. But they're basically counters that you give them pulses. And they just go up by one for like counting people going through gates and stuff like that. Well, like Yeah, like, there's like the ones that you hold in your hand. Yeah. And you click the button, they make versions that that instead of a thumb pressing it, it's a electrical pulse. Yeah. Or if you really get fancy, the pulse hits a solenoid and presses the button. But actually, if you want to get really, really fancy, then the energy of the car driving over it is enough to activate the solenoid itself. So it doesn't need a battery right. Now, that would be really fancy, right? There's a completely like, self contained system that doesn't hold any power. Or you're a mechanical engineer. Yeah. Oh, I'm sorry. Yeah, we're wrong discipline. Or find a mechanical engineer and do a double project. Yeah. Yeah. And then IoT. Right. And and the thing to do, right, yeah. And habit tweet, like 10 times a second. How does the circus of my neighborhood hashtag circus? That'd be great. So there's at least one weird project for you to do this summer. I, one minute we are not going to be doing that project. One last thing. Yeah. Do we have an SSPs update? Uh, you know, I saw on Twitter. Okay. So I saw on Twitter, someone was asking about that. In fact, I apologize. I don't remember your name at the moment. But you said that you had binged all of the the Mac fab podcast today, and that you're still waiting on an SSPs update. I was I was about to tweet back at you saying I I'm also waiting on an ESA. And one of the thoughts I had was pork and I were designing the SSPs sort of like as a two part thing. Yeah, I was doing a lot of the analog stuff. And Parker was doing a lot of the digital stuff. And we both had boards made. And the last time I even turned the SSPs on it did what I wanted it to do, like it actually produced and regulated a voltage. I have the thought I could potentially take it to Colorado and work on it a little bit and then bring it back to you and give it to you in a working state in terms of the analog stuff. Yeah, just a thought. That it's just one more heavy thing to lug up. Yeah, it's like 120 pounds. It's not, it's not like and I've got a lot of other 120 pound things to move. But that's because that one tweet was like, if I don't do that, there's like zero chance of it happened. Yeah, so um, because that's the thing is your side was working what was um, I had some transistors that I had flipped on my schematic and it was like a doofus move. So I flipped them back. And those ran just fine. Really all said and done. I do need to do another board spin on it just to make sure that those transistors are in the right orientation, but we didn't do a significant amount of load testing. And I hadn't seen well we didn't have the heatsink on it. We didn't have to make that right. We didn't have the heatsink on it and we didn't actually send it any digital codes because there's the way we have it is we have two separate digital analog converters that are in like A stack Yeah. And, and whatever code you send both of them, you can make it go positive. If you send one a code and negative send the other code, we didn't actually end up sending them any codes. But I did test the output using just an analog voltage from another power supply. And I was able to control the output. So technically, the analog side of the SSPs actually works. It does what it needs to do. It's, you know, there's a handful of things that I would need to test like, Okay, put it under load, you know, it, can it can it continuously dump into a load? What is it overshoot if you turn it on from zero to 120 volts at 10 amps? Like, yeah, what like, How bad does it overshoot? Like, there's a lot of that kind of testing to be done. But really, we're sort of at the point where Parker and I could marry this stuff together. Yep. But we would really need to resurrect it. And like, remember exactly, I think
I think we should, I think you should take it up, and then get the analog stuff done. And then chip back down, and
I'll finish up the digital one. Yeah, yeah. Because Because really, I can do all of the analog work without the digital stuff, because I know what your digital codes will result in. Yeah, it's going to be outputting. It's SPI. Yeah, it's yes. You're sending SPI to two chips. That's the only thing Oh, I apologize. There's one other thing that we hadn't done, we actually had a an output sniffer. We had a DAC on the output, that that read across a pretty high tolerance resisted via, so we could actually see what the output was. So I, if I remember, right, it's SPI to the DAX and it was I squared C to the from the law should do is I should design a little board that plugs into that, that you can just talk to it thoroughly over the computer, and then it will control the DAC. So you can say I want to output this. And it will do that. Great. So it'd be nice and simple. Because the main thing on the front end of the digital side is actually controlling all the switches and making sure that the tool is ever operating it can't fuck it up. Yeah, actually, you can spend more time making sure that happens. Exactly. Yeah. So but
if I just say, you know, you type in and hopefully you don't mess up when you press enter. We can do that.
Yeah, yeah, actually, really easy, because you have an entire digital board already designed. And that had like it had switches it. I think it had a screen on the screen. There's all testing systems. Yeah, right. Right. So like, actually, we we could technically work on this in parallel separately. Because I can get all my stuff done. And really actually what I need to do at this point, I need to run do another read spin of the board. As you only did half of it to you, right, you need to double it do a full spin of the whole energy. Energy. Energy. Cube. That's right. We got to the energy. Wow, that's old school. Yeah. Okay. Yeah. I need to redo an ENERGY CUBE board. Do the official one. Yeah. The the capacitor stack is done that already has all the rails on it. But I need to mount a copper block to it. Yes. So we need when we design when you design it to stack? Yeah, is the big thing on that is because the op amps need to be a certain space between them, because we're going to mount them back to back with the copper block in between recall, right. And you can't just mount the op amps to it because the op amps are live. The pad is Yeah. aluminum oxide, aluminum oxide with with ceramics. Yep, yep, yep. Junk on it. And so we had to design the spacer. And I don't know exactly how to do that yet. I know. We're probably gonna send it off the machine at you know, okay, so two options. Mouser has a lot of ceramic isolators. Yeah, that you can just buy. That's fine. We can do that. We can do that a block though. On other hand, well, at the same time, I do know a place in Houston that will mill custom aluminum oxide. They specialize in ceramics if we could. So in a bygone era, I actually designed a product that had an electrical isolator. That was like a very specific triangle. And you would mount that triangle to whatever machine you were wanting to mount your sensor to. And then you would mount your sensor on top of that, and they would electrically isolate them, but it would transmit vibration through and you would grease them up. So yeah, good couple you good. Good couple. Yeah. So if we wanted to get stupid with it, we could have like custom. Yeah, but that's a standard package and they make standard oxide packages. Yeah. And you know, given that, you know, I made that design three years ago. I'm gonna look through it one more time, because I do have some concerns about the the thing that's that really sucks is the SSPs is actually fairly efficient. If you want 120 volts, if you want a one volt at 10 amps, you had to get rid of 119 volts in order to get that. So it's not really It's okay. It's not efficient at the voltages that you would normally use as which kind of sucks. So like I, you know, it's fine, but it's not anything that anyone would ever, like, want to use on a regular basis. But we should finish it, we should write it. Yeah, yeah, cuz it was fun. I remember, like, enjoying it a lot. So I think I think what we need to do is figure out what the max load on those op amps are going to be temperature wise. Yeah, how much wattage we needed. Now, I want to temperature how much power we need to dissipate. And so we can design our waterblocks correctly. Yeah, cuz because the the thermal, and I think its performance is going to be tough. And I think it's pretty soon going to be is a design the copper contact area to be as thin as possible to get to the water with fins. Yeah, because we, so we had a rectangular block that we were going to put water through and like zigzag pattern, right? Like, kind of drill through it and have like, well, it was copper tubing through it, right?
No, no, it was, you would drill, the idea was because we were trying to build this on a drill press and not working too well.
But the copper block, say it's like two inches by eight inches. And it was like the same inch thick. And you would drill down all the way through here, the length of the length, and then drill down and then back through it. So you kind of would make a u and then tap the exit holes, and then three and put in brass insert. Oh yeah, I gotcha. And so that way, I remember no loop. Yeah, I that would work. I I want to see if I can design something and just get it machined. And see how expensive that is. Because I'd rather do that because it's a lot less work and work a lot better. What's your kid's gonna be making it at two pieces and then put an O ring? And so they sandwich together? Right? And and I tell you what, why don't you? Why don't you take a look at that and tell me kind of what the dimensions of the water block? Because then because that affects the energy on cube. Yep. And and all that jazz. And we already have like a whole like Monster radiator for this thing. And do we have a pump for it? We don't have any of that. We don't have any of the cooling stuff. I thought we had the radiator? No, don't have that. Okay, good. The Transformers in all the capacitors. formers are huge, huge. Like 20 pound transformers. Oh my gosh, yeah, yeah, there. This is gonna be really cool. When it's done. Yeah, it's gonna be awesome. I think we should just kick it back up and just say yes, it's finishing. Yeah. And the thing is, okay, so I'm moving into an apartment next week. And other than my full time job that I'm going to be working, I kind of don't have anything to do until I get into a house because all of my belongings are going to be in a U haul, or a storage facility somewhere. And so it's kind of one of those things where like, for the next two or three months, I'm gonna have a lot of like, after work, like looking at schematics, like designing mute circuits and things kind of, because I'm not going to have an electronics bench. So I have time to look at these schematics and kind of piece them together. Cool. Yep. So stay tuned. Next time, I guess for more SSPs updates. Oh, my God, we're actually doing more finishing. Yeah. And this is this is my last like in Houston. Podcast. So we're going to look at how things work. I may not be here next week because I in a truck driving. Yeah. But the week after I should be here. Well, it should be here in in audio. So
and we're not going to miss next week though. So right, right, right. There's there's actually a really exciting episode next week. It'll be really fun. Yeah.
Till next time, guys. Okay, that was the Mac fab engineering podcast. We are your host, Parker, Dolman and Steven Gregg. See you next time. Take it easy. We should have a really cool quote at the end like they do on embedded. I like that. Like, see you later guys. Take it easy, like really cool quote of some person that she's reading. Oh, I mean, that's operational quote. Yeah. What's inspirational hang in there. If this was a video we would put one of those dentist photos of a cat cat let's say like hanging from the Yeah, from the ceiling. Yeah. Thank you. Yes, you are a listener for downloading our show. If you have a cool idea, project or topic that you want Steven and I discuss, tweet us at macro fab or email us at podcast at macro calm. Also check out our Slack channel which we have continuing conversations about what we're talking about on the podcast. If you're not subscribed to that podcast yet, click that subscribe button wherever you listen so that way the latest episode shows up in your inbox right when it gets you know published
and make sure you review us because reviewing us makes us happy later everyone
like a year with all the city permitting and shit
Parker and Stephen chat about the recent Nintendo Switch hack and Stephen moving to Colorado.
AI and ChatGPT have been in the news about how it will change world views or will it be relegated, making sure NPCs in video games don’t repeat dialog?
How easy is it to make a retro gaming console? Stephen breaks down his design and build criteria that involves no custom PCBs.
