1 00:00:00,001 --> 00:00:05,440 Welcome to The Bootloader. I'm Tod Kurt. 2 00:00:05,440 --> 00:00:10,000 And I'm Paul Cutler. We have a special episode for you today as we welcome our first guest, 3 00:00:10,000 --> 00:00:13,000 Liz Clark of Adafruit Industries. Liz, welcome to the show. 4 00:00:13,000 --> 00:00:15,160 Hey, thanks so much for having me on. 5 00:00:15,160 --> 00:00:19,240 I'm glad you could be here. We've each brought two things to share. We'll chat about each 6 00:00:19,240 --> 00:00:23,480 one for a couple of minutes, but no more than five. Liz, why don't you start us off? 7 00:00:23,480 --> 00:00:30,440 All right. Yeah. So as Paul said, I work with Adafruit Industries. And so as a result, I 8 00:00:30,440 --> 00:00:36,240 have to confess my first choice is a little biased. It's the Adafruit It's a Snap iOS 9 00:00:36,240 --> 00:00:41,400 app. And I'm mildly obsessed with it. I've done a couple of Learn Guides with it recently. 10 00:00:41,400 --> 00:00:46,840 And this app is done by our iOS developer, Trevor. And it basically acts as a bridge 11 00:00:46,840 --> 00:00:53,440 between your iOS devices and your Adafruit I/O feeds. So if there's ever been something 12 00:00:53,440 --> 00:00:58,440 on your phone that you've wanted to log to Adafruit I/O or have Adafruit I/O control 13 00:00:58,440 --> 00:01:05,000 something on your phone, now you can do that. And the way that you build it up is using 14 00:01:05,000 --> 00:01:10,140 Apple shortcuts. And if folks don't know, in Apple shortcuts, you can kind of make these 15 00:01:10,140 --> 00:01:17,240 like mini scripts using kind of these little widgets. So you can kind of you can even use 16 00:01:17,240 --> 00:01:23,840 scripting things like if statements, you can do Base64 encoding, all that kind of stuff. 17 00:01:23,840 --> 00:01:30,160 So there's a lot of functionality. Recently, I've done two guides using it. The first was 18 00:01:30,160 --> 00:01:38,240 I used a shortcut to send my iOS photos up to Adafruit I/O using Base64 encoding. And 19 00:01:38,240 --> 00:01:45,280 then I'm using a Qualia S3, which has an ESP32 S3 on it, with a 720 by 720 round display 20 00:01:45,280 --> 00:01:50,880 to take the data down from the Adafruit I/O feed, decode it, and then display it on the 21 00:01:50,880 --> 00:01:56,260 display. I just think that's really awesome because otherwise you'd usually have to like 22 00:01:56,260 --> 00:02:00,620 create all the bitmaps or JPEGs ahead of time so it'd be able to be displayed. But this 23 00:02:00,620 --> 00:02:06,040 you can kind of do it on the fly just by like running the shortcut, sending it to the I/O 24 00:02:06,040 --> 00:02:11,520 feed if it's a snap. So I'm really into that. And then the other one that I did is using 25 00:02:11,520 --> 00:02:17,440 Apple HomeKit using off-the-shelf smart home devices. I used a temperature and humidity 26 00:02:17,440 --> 00:02:23,120 sensor and a light sensor. And I was able to get that data into Adafruit I/O and display 27 00:02:23,120 --> 00:02:30,240 it on a dashboard. And then the final thing is I got a light strip that's a matter-controlled 28 00:02:30,240 --> 00:02:36,720 light strip. And I was able to set up kind of a feedback loop where if I change the value 29 00:02:36,720 --> 00:02:44,440 on a feed in Adafruit I/O, it would send a text to my phone. And then I was able to use 30 00:02:44,440 --> 00:02:50,480 a text automation with shortcuts so that it would then control the light strip. So basically 31 00:02:50,480 --> 00:02:55,960 I'm able to change the value on the dashboard and control the light scene, which I think 32 00:02:55,960 --> 00:03:01,960 is really cool. And then I was able to write some CircuitPython code to use a feather TFT 33 00:03:01,960 --> 00:03:06,160 with some buttons. So I'm controlling the light strip from the feather using this kind 34 00:03:06,160 --> 00:03:08,000 of bridge in between. 35 00:03:08,000 --> 00:03:09,280 Oh, very cool. 36 00:03:09,280 --> 00:03:13,040 Yeah. So I'm really excited about it to snap. I think Trevor's done a really awesome job 37 00:03:13,040 --> 00:03:16,880 with it. He also has his own guides. He's done a health status board where he's able 38 00:03:16,880 --> 00:03:22,360 to log all his workouts for the week. And he's also done a weather project. You can 39 00:03:22,360 --> 00:03:28,320 send the weather from your phone for your location up to Adafruit I/O as well. And with 40 00:03:28,320 --> 00:03:32,600 weather projects, you often have to have an API key and all that. But this, you avoid 41 00:03:32,600 --> 00:03:35,840 all that. You're just sending the data straight to it. And then with CircuitPython, you can 42 00:03:35,840 --> 00:03:40,400 just read the feed directly and display it. 43 00:03:40,400 --> 00:03:45,360 I liked his project using the health kit with the health statistics. It showed how many 44 00:03:45,360 --> 00:03:49,360 steps he had taken for the day. And I could totally see putting something like that on 45 00:03:49,360 --> 00:03:53,120 my desk to guilt me into getting the 10,000 steps a day that I need. 46 00:03:53,120 --> 00:03:54,120 Absolutely. Yeah. 47 00:03:54,120 --> 00:03:55,120 Totally. 48 00:03:55,120 --> 00:03:56,120 Yeah. 49 00:03:56,120 --> 00:03:59,400 The whole shortcuts thing, I really want to get into it because Apple has forever had 50 00:03:59,400 --> 00:04:03,560 this on the desktop, this AppleScript language, which is this weird English-like language 51 00:04:03,560 --> 00:04:09,160 that you can use to have programs talk to each other, which is fascinating. And so I 52 00:04:09,160 --> 00:04:12,560 think shortcuts is built on that. 53 00:04:12,560 --> 00:04:14,080 I think so. Yeah. 54 00:04:14,080 --> 00:04:19,840 And so by Trevor making an app that's just another shortcuts participant, everything 55 00:04:19,840 --> 00:04:24,040 else that speaks shortcuts, which is every other app on the phone or on the iPad, can 56 00:04:24,040 --> 00:04:29,880 now participate. And now you can do what you did, like send photos up trivially using MQTT 57 00:04:29,880 --> 00:04:30,880 or whatever. 58 00:04:30,880 --> 00:04:31,880 Yeah. 59 00:04:31,880 --> 00:04:33,480 It's just so crazy. 60 00:04:33,480 --> 00:04:39,040 Yeah. And I'd never used shortcuts before or anything like that. And once you get into 61 00:04:39,040 --> 00:04:44,880 it, it kind of blows your mind because it's all the stuff you can automate. And every 62 00:04:44,880 --> 00:04:48,800 app does seem to have its own functions that are available too. 63 00:04:48,800 --> 00:04:54,040 Yeah. I think Apple almost mandates that you have certain shortcut capability. You can 64 00:04:54,040 --> 00:04:58,240 open a file, save a file or something. When you submit an app to the App Store, it's like, 65 00:04:58,240 --> 00:05:01,720 "Hey, you need to implement this to become part of the ecosystem." 66 00:05:01,720 --> 00:05:02,720 Okay. Gotcha. 67 00:05:02,720 --> 00:05:09,000 Well, that was a good first one. Tod, what do you have for your first one this episode? 68 00:05:09,000 --> 00:05:14,800 Even though I was a big machine learning person back in the day, I've been very down on most 69 00:05:14,800 --> 00:05:21,480 of the modern LLM "AI" in quotes, if you can see my fingers making air quotes, AI stuff. 70 00:05:21,480 --> 00:05:24,960 But there are certain cases where I'm a big fan of it. And Raspberry Pi just introduced 71 00:05:24,960 --> 00:05:30,840 a new camera module for their Raspberry Pi little computers. It's called the AI Camera. 72 00:05:30,840 --> 00:05:35,840 And it plugs into the normal CSI port of Raspberry Pi, just like any other camera you'd get for 73 00:05:35,840 --> 00:05:40,120 the Raspberry Pi. But it's got a little bit of extra special sauce. And some of that extra 74 00:05:40,120 --> 00:05:46,040 special sauce is that it's got an embedded RP2040 microcontroller with 16 megabytes of 75 00:05:46,040 --> 00:05:54,440 flash. And what that's doing is it's snooping on the camera data and you can use it to have 76 00:05:54,440 --> 00:05:59,920 a pre-trained AI model loaded up into the camera module. And so you just get down the 77 00:05:59,920 --> 00:06:07,840 video feed and then the machine learning results. It's like synced to the camera over I squared 78 00:06:07,840 --> 00:06:13,300 C over the single cable. And so this like offloads all the hard work of doing machine 79 00:06:13,300 --> 00:06:17,600 learning stuff on the Pi and puts it in this little camera module, which is pretty cool. 80 00:06:17,600 --> 00:06:23,880 And philosophically, I love these where the models are running on device at the edge, 81 00:06:23,880 --> 00:06:28,480 rather than where they have to make a network request out to the world that is charging 82 00:06:28,480 --> 00:06:35,240 you some dollars per request. And it might go down. And so it's pretty interesting that 83 00:06:35,240 --> 00:06:40,040 they're able to get the functionality and stick it in the camera module. And so you 84 00:06:40,040 --> 00:06:43,000 just talk to it like a normal camera. And then if you want to talk to it like a normal 85 00:06:43,000 --> 00:06:48,760 camera plus some other stuff, you can load up some machine learning model into it. Like 86 00:06:48,760 --> 00:06:53,360 the two demos they give are a pose estimation. So it can just look at the video and kind 87 00:06:53,360 --> 00:06:57,520 of see how a human is positioned in 3-space, like where their arm is, where their head 88 00:06:57,520 --> 00:07:02,880 is. And one of the other examples is sort of the hello world of vision machine learning, 89 00:07:02,880 --> 00:07:06,360 which is the object detection. So you like hold up a coffee cup to it. It'll draw a rectangle 90 00:07:06,360 --> 00:07:11,080 around it and say cup. And so these are just the demos they give out. They've got a whole 91 00:07:11,080 --> 00:07:14,760 repo with a bunch of other examples. And the show notes, there'll be links to all this. 92 00:07:14,760 --> 00:07:19,800 There's a pretty good Hackster article that shows them trying to use it and getting started 93 00:07:19,800 --> 00:07:26,520 with it. But there's also a GitHub repo, a really nice getting started guide on the Raspberry 94 00:07:26,520 --> 00:07:30,960 Pi documentation website. And Adafruit has these in stock or they had them in stock. 95 00:07:30,960 --> 00:07:37,280 I think they're out of stock currently. I was surprised to see that they're $70 more 96 00:07:37,280 --> 00:07:41,280 than the cost of the Pi itself. But when you actually look at all the functionality, it 97 00:07:41,280 --> 00:07:44,000 kind of starts to make sense. Yeah. I mean, I think one of the reasons why it's so much 98 00:07:44,000 --> 00:07:47,640 more expensive, because you're thinking, oh, if it's just essentially a Raspberry Pi Pico 99 00:07:47,640 --> 00:07:51,680 and a webcam module smashed together, why is that like double the price? Well, it's 100 00:07:51,680 --> 00:07:56,800 like they're actually using a really good camera module. They're using this Sony FX100 101 00:07:56,800 --> 00:08:02,280 module. Because with this machine learning stuff, it's really garbage in, garbage out. 102 00:08:02,280 --> 00:08:06,960 So having a really good image sensor to get the best quality image you can before you 103 00:08:06,960 --> 00:08:11,720 feed that through your model is really important. So they have a really good sensor on the front 104 00:08:11,720 --> 00:08:16,200 end, better than the most sensors. So I was thinking of getting this module mostly just 105 00:08:16,200 --> 00:08:20,880 because it's a good camera. And then playing with AI stuff, because I don't really have 106 00:08:20,880 --> 00:08:24,240 a use for doing this sort of stuff. It might be fun to maybe turn on lights when I walk 107 00:08:24,240 --> 00:08:29,080 in the room or something. I don't know. But yeah, so that's out now. And it's cool because 108 00:08:29,080 --> 00:08:33,560 I know when I've tried to do machine learning stuff on the Pi, just running models on the 109 00:08:33,560 --> 00:08:43,600 Pi is kind of a pain. Yes. So I saw that all go by when it came out, and I didn't look 110 00:08:43,600 --> 00:08:47,920 too into it because I assumed it was just a camera. I didn't realize it also had the 111 00:08:47,920 --> 00:08:53,080 RP2040 on there. So that's really cool. And 16 megabytes of flash. 112 00:08:53,080 --> 00:08:58,120 Yeah. And if you're looking for a similar idea to have it be much cheaper, is for the 113 00:08:58,120 --> 00:09:04,040 last two years now, there's been this little $10 board called the Person Sensor by Useful 114 00:09:04,040 --> 00:09:08,360 Sensors. You can get them at SparkFun. They have them in stock right now. Me and Paul 115 00:09:08,360 --> 00:09:12,680 talked about this back in episode three of the bootloader. And this is around Halloween 116 00:09:12,680 --> 00:09:18,640 time as well. So I made at that time a little CircuitPython eyeball that would track you 117 00:09:18,640 --> 00:09:22,680 around the room as it recognized your face. And so it's much simpler. It's running on 118 00:09:22,680 --> 00:09:30,360 a very cheap STM32 with a really cheap little webcam sensor. But it's $10, and it can recognize 119 00:09:30,360 --> 00:09:34,760 a face to tell you if a face is looking at it, which is really useful for a lot of cases. 120 00:09:34,760 --> 00:09:39,640 Like if you've got some sort of UI that you want to save power and have some of it powered 121 00:09:39,640 --> 00:09:43,680 down when no one's paying attention to it. But if you walk up to it, it'll light up. 122 00:09:43,680 --> 00:09:48,840 It'll turn on. It'll say, "Hey, I'm here. Press this button to go." So I think having 123 00:09:48,840 --> 00:09:53,400 people detection without having to know who the person is, is really useful. And both 124 00:09:53,400 --> 00:10:00,680 of these cameras can do that. And if you just need that, spend $10 instead of $70. 125 00:10:00,680 --> 00:10:04,400 So yeah, so that's my first one. How about you, Paul? What's your first one for this 126 00:10:04,400 --> 00:10:06,920 week? My first one is the Bumpin' Sticker by Guy 127 00:10:06,920 --> 00:10:11,720 DuPont. I love all of Guy's projects, probably because so many of them have a musical take 128 00:10:11,720 --> 00:10:16,000 to them. I had him on the CircuitPython show a couple of years ago, and he also did a maker 129 00:10:16,000 --> 00:10:20,380 chat with Liz this past August on CircuitPython Day. And I'll link to those in the show notes 130 00:10:20,380 --> 00:10:24,080 as well. If you haven't come across Guy's projects before, you should definitely check 131 00:10:24,080 --> 00:10:28,240 them out. The Bumpin' Sticker project is a take on the bumper stickers you might have 132 00:10:28,240 --> 00:10:33,000 seen on some cars that say, "Keep honking. I'm listening to insert your favorite band 133 00:10:33,000 --> 00:10:37,800 here." Except with the Bumpin' Sticker, Guy is sharing what he's listening to in real 134 00:10:37,800 --> 00:10:43,360 time. He picked up an HDMI screen from AliExpress for about $60 and hooked it up with a Raspberry 135 00:10:43,360 --> 00:10:48,040 Pi. He used his Last.fm account, which tracks everything you listen to, which is hooked 136 00:10:48,040 --> 00:10:52,520 up to his Spotify account. This lets you use almost any streaming service with the project 137 00:10:52,520 --> 00:10:56,680 as long as you have Last.fm, so it doesn't matter if you're using YouTube Music, Apple 138 00:10:56,680 --> 00:11:02,200 Music, Spotify, et cetera. He then wrote a couple of TypeScript programs using Valtown 139 00:11:02,200 --> 00:11:07,720 that scrapes his Last.fm page. And then lastly, he added a Particle IoT board with cellular 140 00:11:07,720 --> 00:11:11,800 service. That board gets the data of what he's listening to, I think via JSON from 141 00:11:11,800 --> 00:11:16,440 the Valtown TypeScript program, passes it to the Raspberry Pi, which then generates 142 00:11:16,440 --> 00:11:21,440 the image and displays it on the HDMI screen as a bumper sticker on the back of his car. 143 00:11:21,440 --> 00:11:25,240 He also shows how he wired it for power from the car's wiring and how he mounted all of 144 00:11:25,240 --> 00:11:28,920 it and even weatherized it. All in all, it's a really cool project. 145 00:11:28,920 --> 00:11:35,480 I loved that project. Just how all the pieces came together. I also think the video is an 146 00:11:35,480 --> 00:11:40,200 excellent explainer on how it all worked. The video is so good. 147 00:11:40,200 --> 00:11:47,000 Yeah, so good. It's excellent. Yeah, he does such a great job. And I think 148 00:11:47,000 --> 00:11:50,680 in the time that he's released that in the last couple of weeks, he's already released 149 00:11:50,680 --> 00:11:54,120 two new projects just this week that I'm trying to catch up on. 150 00:11:54,120 --> 00:12:00,200 He even has a new Mastodon bot account that I won't spoil, but I saw him post up today. 151 00:12:00,200 --> 00:12:04,840 Yes, that was the one that I saw this morning. So we'll link to that in the show notes as 152 00:12:04,840 --> 00:12:06,600 well. Excellent, I haven't heard of this. There's 153 00:12:06,600 --> 00:12:11,680 two aspects of his project that I really enjoyed as a nerd. One is the HDMI display he used 154 00:12:11,680 --> 00:12:17,400 wasn't like your standard 4x3 or 16x9 display. It's like a 4 to 1 format. It's like this 155 00:12:17,400 --> 00:12:22,640 really long rectangle. I think it's meant to be like a smart rear view mirror or something, 156 00:12:22,640 --> 00:12:27,600 but it's like it's the size of a bumper shape of a bumper sticker rather than like, obviously 157 00:12:27,600 --> 00:12:34,960 a screen. And so to have it change when it's on the back of his car and it changes, you're 158 00:12:34,960 --> 00:12:37,960 like, wait, that wasn't a bumper sticker. That's something else. 159 00:12:37,960 --> 00:12:41,360 Yeah. I can't imagine driving around Boston and 160 00:12:41,360 --> 00:12:44,320 actually encountering that and trying to figure out how that was actually done. 161 00:12:44,320 --> 00:12:47,440 And the other is it. Oh, yeah, sorry. Go ahead, Liz. 162 00:12:47,440 --> 00:12:50,080 Speaking as someone who's around Boston, that probably wouldn't even be the weirdest thing 163 00:12:50,080 --> 00:12:56,800 you see on the road that day. And the other thing is he's using this thing 164 00:12:56,800 --> 00:13:05,800 called Valtown, which is a like their little slogan is, if GitHub Gists could run and AWS 165 00:13:05,800 --> 00:13:10,200 Lambda was fun. It's like there's all these services out there that will run code for 166 00:13:10,200 --> 00:13:15,040 you and charge you and all that kind of stuff. But Valtown is free for like trying it out 167 00:13:15,040 --> 00:13:19,840 and doing some like open source stuff. And it is just like you have a function and you 168 00:13:19,840 --> 00:13:24,720 say, go run this function when this happens, like when this URL comes in. And that's pretty 169 00:13:24,720 --> 00:13:29,800 cool. I've been wanting to have a project that uses this because it looks like actually 170 00:13:29,800 --> 00:13:34,440 kind of interesting and fun to play with. Time to dust off your JavaScript and TypeScript 171 00:13:34,440 --> 00:13:36,480 skills. Yeah, totally. 172 00:13:36,480 --> 00:13:39,860 All right, Liz, what's your second one for us? 173 00:13:39,860 --> 00:13:44,440 So my second one's a little bit of a left turn from my previous one, but it's the nothing 174 00:13:44,440 --> 00:13:51,480 to see/hear wireless fuzz pedal. So I follow quite a few guitar pedal folks on Instagram 175 00:13:51,480 --> 00:13:57,400 and newest post from different noises last week is a wireless fuzz, quote unquote, that 176 00:13:57,400 --> 00:14:02,320 completes the circuit by transmitting it via an infrared LED. So the circuit converts the 177 00:14:02,320 --> 00:14:07,080 signal from guitar level to headphone audio level, and then again into light that's sent 178 00:14:07,080 --> 00:14:12,800 out through an infrared LED into a transmitter. And what I love about different noises is 179 00:14:12,800 --> 00:14:17,400 he includes a schematic for all his custom pedals. So if you scroll through his feed, 180 00:14:17,400 --> 00:14:23,160 it's just schematics of pedals you could breadboard together real quick. And he includes some 181 00:14:23,160 --> 00:14:28,720 video explaining the circuit and also proving that when he blocks the IR light, there is 182 00:14:28,720 --> 00:14:32,520 no sound. And when he lifts it back up, you have it. And he said he got it to think about 183 00:14:32,520 --> 00:14:40,200 50 feet between the LED and the transmitter. So it's still like a kind of a standard fuzz 184 00:14:40,200 --> 00:14:44,200 circuit. It didn't seem to affect the sound of it. Of course, fuzz is always a little 185 00:14:44,200 --> 00:14:50,920 low fidelity to begin with. But I just think it's so cool. I love when people like figure 186 00:14:50,920 --> 00:14:57,280 out weird ways to transmit signals, you know, whether it's like an audio jack for a soldering 187 00:14:57,280 --> 00:15:04,840 iron or, you know, anything like that. So I thought that was really cool. And in general, 188 00:15:04,840 --> 00:15:06,240 his work's always really interesting. 189 00:15:06,240 --> 00:15:12,880 I was looking at his Instagram page and I love, like, there's this whole community of 190 00:15:12,880 --> 00:15:17,840 pedal builders and DIY electronic builders that work on these strip boards. And they 191 00:15:17,840 --> 00:15:23,200 have these like way of describing how the circuit is both like how you would build it 192 00:15:23,200 --> 00:15:26,640 by just saying like, this is where the parts would go on the strip board. And it's just 193 00:15:26,640 --> 00:15:32,460 really beautiful because it's like the strip board is like a horizontal rows of copper 194 00:15:32,460 --> 00:15:35,720 are connected and that's how the holes are. So it's sort of like breadboards, but, you 195 00:15:35,720 --> 00:15:41,960 know, it's a whole 20 pin row is connected. And so you can like make these pretty complicated 196 00:15:41,960 --> 00:15:47,440 circuits without really much wiring. You just have to be really clever on where your components 197 00:15:47,440 --> 00:15:52,360 straddle the rows and stuff. And so it's really fun to look at them just without even knowing 198 00:15:52,360 --> 00:15:57,240 what they're doing. It's like little cityscapes. 199 00:15:57,240 --> 00:16:01,960 And I love the DIY guitar pedal community, similar to the maker community. They're always 200 00:16:01,960 --> 00:16:07,680 sharing. They're really welcoming. And it's just folks having fun experimenting. 201 00:16:07,680 --> 00:16:09,760 Very cool. Tod, what's your second one for us? 202 00:16:09,760 --> 00:16:14,640 All right. Well, so I'm kind of sniping one of your topics a little bit here, Paul, because 203 00:16:14,640 --> 00:16:21,440 you're the one who's got the Bambu printer. So my 3D printer, it's a Prusa Mark 2.5 S 204 00:16:21,440 --> 00:16:28,840 that I've had for, I don't know, six or so years. And it works great. But, you know, 205 00:16:28,840 --> 00:16:35,400 in the intervening years, 3D printers have gotten much better, like surprisingly to me, 206 00:16:35,400 --> 00:16:43,960 but like the Mark 4 S Prusa and the BambuLab, they both literally print twice as fast as 207 00:16:43,960 --> 00:16:49,040 my printer. Like I loaded up profiles for both and like sliced it and see how long, 208 00:16:49,040 --> 00:16:54,080 like what the slicer said it was going to take. I'm like, oh, really? Come on. So I've 209 00:16:54,080 --> 00:16:58,040 been, I've been looking around. I've been trying to figure out, should I go Prusa again? 210 00:16:58,040 --> 00:17:03,840 Because open source, you know, but I really like having the enclosed frame. And so I'd 211 00:17:03,840 --> 00:17:09,280 love to, so like the BambuLab X1 is really appealing to me because of that. And so I 212 00:17:09,280 --> 00:17:15,040 stumbled upon that there is an open source firmware for the Bambu, which kind of blew 213 00:17:15,040 --> 00:17:22,140 my mind. And it's sort of like, it's a, it's a thing that, that modifies the application 214 00:17:22,140 --> 00:17:29,120 processor board of the X1. So the X1 apparently works kind of like a, like my Prusa with Octoprint, 215 00:17:29,120 --> 00:17:33,440 where there's the Octoprint Raspberry Pi that kind of controls the user interface. And then 216 00:17:33,440 --> 00:17:38,480 there's the motion control system that is controlling the actual motors and the extrusion 217 00:17:38,480 --> 00:17:43,380 of plastic and stuff. And the Bambu is set up very similarly. And this open source firmware 218 00:17:43,380 --> 00:17:49,120 called X1 Plus only runs on the application processor and only modifies it slightly so 219 00:17:49,120 --> 00:17:53,640 that it will run an alternate firmware off of the micro SD card. So if you want to run, 220 00:17:53,640 --> 00:17:57,620 if you want to go back to running stock Bambu, you still can. But if you want to run this 221 00:17:57,620 --> 00:18:01,720 new open source version, you can try it out. And this new open source firmware is just 222 00:18:01,720 --> 00:18:06,640 basically a different user interface. It doesn't affect the motion control aspect of the printer. 223 00:18:06,640 --> 00:18:09,840 So it's still going to be fast and good and all that kind of stuff. And so, so I've been 224 00:18:09,840 --> 00:18:15,080 like, huh, maybe this is the path I take for a new printer is, is I kind of do both closed 225 00:18:15,080 --> 00:18:19,960 source and open source at the same time. And the reason why I discovered this is because 226 00:18:19,960 --> 00:18:26,160 the people behind the X1 Plus firmware also have come out with a crowd supply campaign 227 00:18:26,160 --> 00:18:31,500 for an expansion board that will let you add like more cameras or lights or other sort 228 00:18:31,500 --> 00:18:37,680 of sensors and actuators that you can trigger via your G code or whatever. And so it's not, 229 00:18:37,680 --> 00:18:41,240 it's not required to run the firmware, but it's like a, if you have the firmware, you 230 00:18:41,240 --> 00:18:46,000 can use this cool extra little plugin board and do a bunch of other, other interesting 231 00:18:46,000 --> 00:18:48,840 stuff. So that's why, that's what I'm kind of thinking right now. It's like, should I, 232 00:18:48,840 --> 00:18:51,920 should I get a new printer? And should it be one of these? Should it be a Bambu with 233 00:18:51,920 --> 00:18:55,720 this new crazy open source firmware, which is very intriguing. I don't know if it'll 234 00:18:55,720 --> 00:18:59,800 run on your printer. Cause I, Paul, cause I think yours is a P, a P1, is that right? 235 00:18:59,800 --> 00:19:03,240 I have a P1P and yeah, I did look at it. That was the first thing I looked for and nope, 236 00:19:03,240 --> 00:19:08,200 this is just for the X1 Carbon. So the higher end one, I was too cheap to get the open source 237 00:19:08,200 --> 00:19:15,080 version I guess. But I'm, I'm interested to hear that, that you're thinking about joining 238 00:19:15,080 --> 00:19:18,480 the dark side and going with a closed source printer, even though you could argue that 239 00:19:18,480 --> 00:19:22,880 the new Prusas aren't very open source to begin with either. And this is almost a luster 240 00:19:22,880 --> 00:19:26,600 of two evils. Yeah. That's the thing is I've, I've been doing open source printers since 241 00:19:26,600 --> 00:19:31,600 like the very first MakerBot. And so yeah, it's, it's the fact that Prusa has been going 242 00:19:31,600 --> 00:19:36,360 more closed sources has even made me think of going to other printers that are not, not 243 00:19:36,360 --> 00:19:41,480 open because, because of that aspect. And so yeah, it's, it's still, it's still in the 244 00:19:41,480 --> 00:19:46,320 thought processes for me. One of the things that's, that's forestalled my choice is there 245 00:19:46,320 --> 00:19:53,000 is a service called CraftCloud, which is a website you can go to and submit, upload an 246 00:19:53,000 --> 00:20:00,160 STL and it will essentially request quotes from hundreds of 3D printer companies across 247 00:20:00,160 --> 00:20:05,080 the U.S. and then it'll, it'll sort them by price and you can, and other features that 248 00:20:05,080 --> 00:20:09,520 you care about. And then you can order like one or, or in my case, I ordered like a, like 249 00:20:09,520 --> 00:20:15,640 a small batch of like 10 and 20 of a little synthesizer enclosure. And the quality was 250 00:20:15,640 --> 00:20:20,840 great. It was from some company in Georgia and, and you know, they shipped within like, 251 00:20:20,840 --> 00:20:25,520 they printed in two days and shipped within, I mean, it got here to me in like two days. 252 00:20:25,520 --> 00:20:30,120 So it was like about a, like less than a week of time, physical time. Yeah, it's great. 253 00:20:30,120 --> 00:20:34,320 So I'm like, well, geez, maybe I just don't get a fancy new printer. Maybe I just use 254 00:20:34,320 --> 00:20:36,600 this service for some of the things I care about. 255 00:20:36,600 --> 00:20:42,120 Well, that is a consideration. How many times have you heard someone buy a 3D printer and 256 00:20:42,120 --> 00:20:45,200 a year later, it's just sitting there collecting dust too. 257 00:20:45,200 --> 00:20:50,120 Exactly. And, and, and, and the costs are, the costs are right at the, at the, right 258 00:20:50,120 --> 00:20:55,880 at the perfect place of where I'm like, ah, you know, if it was a little cheaper, I just 259 00:20:55,880 --> 00:20:58,720 would use this all the time. But if it was a little more expensive, I just would buy 260 00:20:58,720 --> 00:21:00,880 the 3D printer. 261 00:21:00,880 --> 00:21:05,800 I would recommend that if you're looking at printing ABS, but especially ASA, which isn't 262 00:21:05,800 --> 00:21:11,080 nearly as bad as ABS to get the X1 carbon. If you're only going to stick with PLA and 263 00:21:11,080 --> 00:21:16,600 PETG, any other printers are good enough to print PLA these days and get the speed increases 264 00:21:16,600 --> 00:21:19,480 that you're talking about. But that enclosure really makes a difference, especially for 265 00:21:19,480 --> 00:21:23,400 ASA being the common one and ABS probably a little more uncommon. 266 00:21:23,400 --> 00:21:29,120 Yeah. Yeah. I'm not really into those fancy plastics, but. 267 00:21:29,120 --> 00:21:34,400 I had the same printer that you had, Tod, and like three or four years ago, I updated 268 00:21:34,400 --> 00:21:41,680 to a Creality Ender and it was kind of amazing to see just the, the jump in performance. 269 00:21:41,680 --> 00:21:42,680 Yeah. 270 00:21:42,680 --> 00:21:43,680 I think you would definitely appreciate it. 271 00:21:43,680 --> 00:21:44,680 Yeah, totally. 272 00:21:44,680 --> 00:21:47,680 And I will say, Noe is considering the Bambu X1 right now. 273 00:21:47,680 --> 00:21:52,240 Yeah. Yeah. It's so good. I mean, the thing is, is that, is that the, the Mark II, my 274 00:21:52,240 --> 00:21:58,000 Mark II 5S, it has been an appliance where like I've got Octoprint running and I will 275 00:21:58,000 --> 00:22:03,440 just send a print job to it knowing that like, oh yeah, I cleaned the bed three days ago 276 00:22:03,440 --> 00:22:07,280 and that's clean enough for the thing I'm doing. And I just print and I can watch it 277 00:22:07,280 --> 00:22:11,440 on the camera. I'm like, yep, it's printing. There's nothing wrong. For the longest time, 278 00:22:11,440 --> 00:22:14,800 3D printing was never just an appliance. It was always this like little, little fiddly 279 00:22:14,800 --> 00:22:19,720 thing you were messing around with the entire time, doing, doing more to it than getting 280 00:22:19,720 --> 00:22:24,480 out of it. And so, yeah, it's it's kind of amazing. 281 00:22:24,480 --> 00:22:26,880 It is. It really is. 282 00:22:26,880 --> 00:22:31,240 All right, Paul, what's our final thing for this week, this month? 283 00:22:31,240 --> 00:22:37,080 This one I'm very excited about. Steph Piper, aka Maker Queen, is a library makerspace manager 284 00:22:37,080 --> 00:22:41,520 in Queensland, Australia. She's come up with a great way to visualize your skill level 285 00:22:41,520 --> 00:22:46,600 and skill gaps in 45 different disciplines with her Maker Skill Trees. Similar to how 286 00:22:46,600 --> 00:22:50,440 in a video game, when you get enough experience points to level up, you can put skill points 287 00:22:50,440 --> 00:22:54,960 into your character to get better. This is the same concept, but for you. It can help 288 00:22:54,960 --> 00:22:59,120 gauge where you are in your learning journey for each of the different skills. 289 00:22:59,120 --> 00:23:03,360 Each skill is made up of 73 hex boxes. You start at the bottom with some of the basic 290 00:23:03,360 --> 00:23:08,240 tasks and work your way up at both skill and complexity. You don't need to do them in order, 291 00:23:08,240 --> 00:23:12,480 though they get a little harder the higher up you climb the skill tree. What's also neat 292 00:23:12,480 --> 00:23:16,800 about it is that it might show some gaps that you might have in a given skill. And now you're 293 00:23:16,800 --> 00:23:21,680 incentivized to try something new and fill in those gaps. Or for the person that says, 294 00:23:21,680 --> 00:23:26,440 what do I do or what do I learn next? This has plenty of inspiration to draw from. 295 00:23:26,440 --> 00:23:32,800 Just a few of the almost 60 disciplines are 3D modeling, 3D printing, amateur radio, coding, 296 00:23:32,800 --> 00:23:37,600 cooking, crochet, Dungeons and Dragons, music, PCB design, robotics, sewing, woodworking, 297 00:23:37,600 --> 00:23:42,040 and so many more. That's just a small selection. So you can see that there's things from the 298 00:23:42,040 --> 00:23:46,460 maker world, the crafter world, the coding world, and all of those have Venn diagrams 299 00:23:46,460 --> 00:23:50,160 that probably cross over each other in so many different ways. 300 00:23:50,160 --> 00:23:54,320 Steph Piper gave a great talk on the maker skill trees at the Open Source Hardware Association 301 00:23:54,320 --> 00:23:57,960 earlier this year. It's under 10 minutes long and it's worth watching to learn more about 302 00:23:57,960 --> 00:24:01,880 it and I'll link to it in the show notes. She's also working on a book that uses the 303 00:24:01,880 --> 00:24:07,680 skill trees, including 13 makerspace skill trees, self-reflection activities, and a life 304 00:24:07,680 --> 00:24:12,480 progress dashboard for how you are doing across many of the skill trees. I think it's definitely 305 00:24:12,480 --> 00:24:17,680 worth checking out and everyone, every makerspace should have a copy of this printed, I think. 306 00:24:17,680 --> 00:24:22,000 That's really cool. And I am a really visual person, so I appreciate having it all laid 307 00:24:22,000 --> 00:24:27,320 out. Some things that you may forget or not think about or not know about, like you said. 308 00:24:27,320 --> 00:24:30,480 So it's really cool. I really like Steph Piper's work in general too. 309 00:24:30,480 --> 00:24:33,960 Yeah. She made a very conscious decision that it should be printed on paper and should be 310 00:24:33,960 --> 00:24:38,840 a tactile experience filling in those boxes, circling them or crossing them out or however 311 00:24:38,840 --> 00:24:42,360 you want to do it and get you off the screens to do it. 312 00:24:42,360 --> 00:24:49,360 I like how it's laid out as a sort of a honeycomb hexagon thing rather than a tree where there's 313 00:24:49,360 --> 00:24:53,040 like a linear progression because when you're learning stuff, it's never a linear progression. 314 00:24:53,040 --> 00:24:58,960 You might bust ahead really far on certain topics, but still be like down farther on 315 00:24:58,960 --> 00:25:02,840 other topics. And so having it be sort of like, oh, this amorphous, you just kind of 316 00:25:02,840 --> 00:25:07,040 grow from the bottom towards the top is really nice. 317 00:25:07,040 --> 00:25:08,040 I agree. 318 00:25:08,040 --> 00:25:09,040 Yeah. 319 00:25:09,040 --> 00:25:14,360 Well, that's our show. For detailed show notes or to join our newsletter, please visit thebootloader.net 320 00:25:14,360 --> 00:25:19,480 and a special thank you to Liz Clark for joining the show this episode. Until next time, stay 321 00:25:19,480 --> 00:25:20,000 positive.