Embedded - 184: Not Likely, Possible, or Safe

Episode Date: January 18, 2017

Ben Krasnow (@BenKrasnow) spoke with us about prototyping, Patreon, and staying current. And a whole bunch of stuff. January 28th Hats and Hacks Party RSVP Ben’s YouTube channel is Applied Science. His recent videos have been shot with the high speed Chronos camera (whose creator David Kronstein was on The Amp Hour #325). Ben has a Patreon page which funds randomness. (Embedded also has a Patreon page, for randomness and mics.) Ben was previously on the show: 119: Do Your Neighbors Have Any Idea? For BLE prototyping, Ben mentioned the OSH Chip by Philip Freidin (146: The Loyal Opposition) and using Processing for Android to make quick-n-dirty test applications. We mentioned the Wazer desktop waterjet. Chris brought up this video describing impedance with a mechanical model. One of Ben’s favorite videos that he did was the first one with an electron microscope, way back in 2011: DIY Scanning Electron Microscope - Overview. Ben gets a lot of his news from Hacker News: https://news.ycombinator.com/ Ben’s Twitter criteria was that they didn’t post updates often too often for his one-a-day check and that they focus on tech: @bunniestudios @vk2zay @elonmusk  (for updates on my car's firmware) @LongHairNasaGuy @szczys @samykamkar @PaulStoffregen @mikelectricstuf @johndmcmaster @michaelossmann @macegr @Chris_Gammell @EMSL @mightyohm And some of his favorite YouTube channels (Ben said it was very difficult to distill as there are many great choices): mikeselectricstuff tesla500 Matthias Wandel NightHawkInLight The Signal Path Techmoan Cody's Lab This Old Tony Clickspring Nick Moore Gross Science Haas Automation Hackaday Reactions I Build It Alex Dainis bigclivedotcom We also mentioned architect Frank Howarth of the urbanTrash channel.

Hello and welcome to Embedded. I am Alicia White alongside Christopher White. Ben Krasnow is back to talk with us about cameras, cars, and maybe computer stuff. Before we talk to Ben, don't forget to sign up for our party. It is in Aptos on January 28th, which is a Saturday from 2 to 5 p.m. It will be at Little Owl, and it is a hats or hacks event.

You should either bring a hat or wear a hack or something along those lines. RSVP in the show notes. What city is Little Owl in? I said Aptos. You did? Oh, okay, it's in Aptos. Damn it.

I was wondering how to pronounce it properly. Which is in California. And it is not Aptos, which is the Spanish pronunciation, but Aptos, which is the American Indian pronunciation. Hi, Ben. It's good to see you. How's it going? It's been fine.

Good. Could you tell us about yourself? Sure. My name is Ben Krasnow. I'm the host of the Applied Science channel on YouTube. And I think we just turned six years. And I also work at Verily, which is Google Life Sciences, which is Alabets life science uh sort of portion cool before we ask you about all the things we asked you to come over and uh tell us about we want to do lightning round where we ask you short questions we want short answers and if we're all behaving ourselves this this doesn't become an hour-long discussion about robots. Okay. Favorite processor? ARM Core 32-bit.

Favorite fictional robot? What are you the hell, 9000? Most important tool to your daily work? And I'll just leave that open. Soldering iron. All right, that was on the list anyway most important tool to your applied science work slash hobby hmm i guess i'd have to go with camera but that's kind of a cop-out answer well since i want to talk to you about cameras okay that works favorite type of wave um An ocean wave.

Excellent. Excellent. I think we can help you with that later on. What science fiction technology or concept do you think will be real in our lifetimes? Self-driving cars. Least favorite electrical component? Oh, man. That's a good one.

Least favorite. Um... least favorite electrical component oh man that's the good that's a good one least favorite um hmm that's a really interesting question least favorite maybe a processor a microprocessor yes ouch okay what's your most favorite most favorite about a resistor? It's kind of hard to like mess that one up. Well, yeah, but it's not like it's got any personality. When I was thinking of favorite, unfavorite, I was thinking of like what's caused me trouble in the past, you know, career. Well, in that case, microprocessors are my least favorite as well. Yeah, so that's kind of, okay, I interpreted it a little differently. How much time or what year do you think we'll get to the point where

10% of the US has truly self-driving cars? Level 5, as they call it, like fully autonomous. Yeah. 10%, I think it's going to happen pretty quick, maybe in the next 5 years, 10 years. Cool.

I'm looking forward to it. all right do you have any others all right okay so now let's get to those camera questions you got an early chronos camera the kickstarter with the super high speed and and it slows down fast things. What drew you to that camera? Well, I've always been really interested in high-speed video. I mean, it's so cool. And then being on YouTube where the whole medium is about interesting video, playing with a high-speed camera makes a lot of sense because it is about the medium, right? Like it's perfectly suited for YouTube videos.

And then my friend David Kronstein, who invented and built that camera, offered me an early prototype, as with a lot of other YouTube creators. He knew that getting some video out would be a good way of showcasing the camera, so that's how I got my hands on it. What videos have you done with it? I've made two YouTube videos that show a bunch of clips that I've collected. And the first one, I have a mechanical rig that spins the camera around like a center of rotation so that the camera is always facing what's at the center. And it spins in an arc that's maybe, you know, 60 centimeters in diameter or something like that. So it's kind of like a matrix bullet time effect where you can film something in the center of the arc. And then as the spins around it you know if the thing is happening quickly it you know shows it in slow mo it looks pretty cool you know in one of your videos you mentioned bullet time effect and i had

no idea what you're talking about until you you just said matrix and i'm like oh right yeah i know i actually realized that not many people know like what the effect names are and so then i was trying to think of how could i summarize it in a title and like spinning camera rig with slow-mo action or something and so i settled on bullet time but i actually did one get one guy that said i disliked this video because i was expecting bullets in it yes well i mean high speed bullet videos can be really interesting especially when you see the shock waves and they're supersonic and all that but but yeah i can see so i don't have a gun otherwise i could have done bullet time of bullets which actually maybe i should do because that actually sounds pretty good okay that leads me to another question have you ever been injured in the filming of your videos

um wait like on camera that I edited out later. Sure. Yeah. Actually, potentially, but it's always like, like not exciting stuff. Like not like, um, like incidental injuries. Like I remember I got hurt pretty bad. I was, um, showing how to temper steel. And so I had these thin bars that I was pulling.

Like I had a very heavy weight that I would add to the bar and then it would buckle. It break at some point. I'd measure how much weight was in the bucket to figure out how strong the bar was. And so, I hardened the steel and then the bar became so strong that I had a bucket with like 80 pounds of steel in it trying to rip this eighth inch bar apart. And finally, it let go and one of the pieces of steel like came down on my thumb and it left a big gash and everything. It's not a very like, you know, sort of exciting injury, right? No, although it does sort of want you to say, you know, I'm glad I wore safety glasses. Yeah, I think I'm kind of, I think safety glasses are pretty good,

but they're kind of like, I don't know, over like they're not going to save you from a lot of things. And then when you do have them on, it's kind of, I don't know, it's more like dust and particles flying and stuff. But yeah, I do wear safety glasses. I do recommend it. So back to the camera. The lighting ends up being a huge part of being able to make it work. Yeah.

Daylight is good. So you'll notice a lot of channels like the slow-mo guys on YouTube shoot in daylight a lot because it's free and it's everywhere. But indoors, yeah, you need a lot of photons. So David Kronstein also built like the world's largest LED array. I don't know if you've seen this on his channel. It's like 10 panels of 100 watts each all aiming at his high-speed studio, which is pretty incredible. Water cooled everything. That's because at the frame rate we're dealing with, there's just not enough light coming in.

Yeah. It's a physics problem. There's nothing you can do. I mean, you could make your image a little, your sensor a little bit more sensitive, but with existing CCD technology, yeah, it's your exposures are down to a couple hundred microseconds at the highest speeds that that camera can go. Is it a really small, uh, uh, F-stop? Um, that's the other thing you you can try to open your lens up to like f2 but then you can't focus on it yeah your depth of field is short and then also the lenses don't perform very well at that setting so you kind of want to stop down as well what other settings do you have you played with um white balance is good so another thing that

makes videos look much better than you know if you look at a video you can't really put your finger on what makes it look pro white balance is definitely one of those things and i was happy to see that david has a custom white balance in the camera so you can aim it at a piece of paper in your scene and hit custom and then the colors come out really nice i don't think we mentioned how fast it is uh it's at 720p i think it'll do 1500 frames a second okay and then at at its tiny resolution of you know like 100 by 100 it does maybe 10 000 frames a second something like that have you tried that um yeah you know it's funny for a lot of things it's actually too much yeah so if you're doing like bullets then you need speed that high but if

you're doing things like water pouring or um even like things breaking like glass breaking and stuff you actually don't need that much speed because it's not going quite that fast and then you get way more resolution and so it's a much more you know compelling looking image have you thought about uh using the extremely high speed or the high speed mode uh coupled with a microscope to look at like bacteria processes or something like that? Yeah, it's funny you mentioned that. So at work, we have a high speed microscope, and it's definitely meant for research. It's black and white only, and it has, you know, metrology tools. But what's cool is that you can take any

old lens on a camera and just move it away from the image sensor and get it to focus much closer. So you can turn almost any old camera into a you know a microscope just by extending the lens which i've done with good effect so i've got a video of like an ant and it's almost filling the frame i mean it's pretty cool you ended up building a lot of support equipment uh to move the camera and i know there are some people out there i think who who got to the Kickstarter and said, I'm going to get this fancy camera. I'm going to be able to do all these neat things. They don't realize it isn't quite point and shoot.

Actually, no, I'd say it is. One of the nice things if you're shooting high speed, you don't really need a tripod because your clip is less than a second long. In a lot of cases, you just can't even shake fast enough for it to matter that much. It's actually funny little side effect but if you go out to the park and you want to get like some high-speed video of a river or a waterfall you should just hold it and you'll never see the camera you'll never see the hand movement in the final footage i noticed you uh you acquired a hummingbird for a little while how did that happen he flew into the garage actually and i was trying to like escort him back out but he hung around long enough where I could get the camera fired up, and I got maybe 30 different clips of him and picked the best one and put that one out.

That was pretty cool, because hummingbirds are just one of those things that are pretty endlessly fascinating. It is, yeah. It's a very weird creature. He was in there long enough where I put out a little cup of sugar water for him because he was hanging out for quite a while. Did you name him? No. So what advice do you have for people who are going to get a camera like this? I think that the fact that there's more high speed cameras out in the world gives people more flexibility to be artistic with them. So up until now, most of the high speed footage that people have seen is like sort of setups like, you know, balloons popping, or, you know, things in a lab setup. But if you have one, and it's portable, it's battery powered, you can

just take it to the park and find an interesting looking bug and capture it. I mean, you have some pretty amazing footage. And so just getting more cameras out there, I think, will open it up and people will come up with more artistic looking shots. See, I like the really low speed cameras that you leave open in the sky for an hour. So it's funny, there's this huge dynamic range between things you can look at with a high speed camera and things where it's almost the same technology, too. You want low noise and that kind of thing. Yeah, it's funny playing with time compression. In fact, I've been thinking about weird stuff to do where you could even have like log time in cameras. Maybe you could go all the way from time lapse into real time into high speed.

And I was thinking of a process that could make use of that. Plant germination. Ah, that's an interesting one. Like flowers opening and then. Yeah. Huh. That could be rather interesting i like this yeah what else so you can also do tilt shift photography like if you've seen this where you

take like an image from the top of a building and if you tilt the lens you can get both distant distant things out of focus as well as close but the middle is in focus so it looks like a miniature it's a really you gotta look this up it's a very strange thing in our brain it's super super like unusual and then when people combine the technique with time lapse it looks like a tiny little village of like little machines moving around it's really crazy you should look it up and i was thinking wow i wonder if anyone's ever done tilt shift high speed this could be like pretty crazy too and for that you'd need to probably does it work with a standard lens mount yeah it's interchangeable that's the other cool thing it's battery powered which is unusual for a high speed camera and it

also has interchangeable lenses so you can put any old c-mount lens on okay yeah what plans you have going forward do you have a list of things yeah so the first one was this um bullet time effect rig which was pretty cool And then I linearized it. I basically stretched the camera out onto a linear rail so that I could move laterally really quick. And then the next step after that was to put a pivot on the lateral transition so that it was always looking at the same point but moving really quick past it. So it was kind of like this drive-by effect. That was pretty good. And then what I wanted to do was tilt the whole thing upright so that I could move the camera like faster than gravity and then pour water or something off the table and have it sort of chase the water or even go faster than the water. Figured that might look pretty good.

I haven't done that one yet. That sounds pretty cool. Sounds like you're slowly reinventing the ILM motion capture stuff that they did in the 70s. Yeah. And so it's fun playing with a camera that, you know, I think everyone should have one right because there's so much unexplored territory right i mean it's just all kinds of things you can do with it i'm thinking your rig that does linear now you need a little little tiny horse to run along so you can recreate the original camera system exactly or insects flying and stuff like yeah it would be really good to have these sort

of controlled setups and stuff another trick i did was to change the lens focal length as it's moving on this linear rail and so this technique is called dolly zoom i don't know if you've seen like vertigo you know or something like that where the camera is moving physically toward the subject but then you're also changing the zoom of the lens. So you end up with this really strange, like perspective change. And I pulled it off in high speed and the effect was, I think it was pretty good. It kind of showed the potential, but the camera was already doing like five meters a second, top speed, or I don't know, maybe about quite that, but three meters a second, top speed. And like the action wasn't quite enough to keep up with the effect. So I needed to go even faster, but it was still pretty cool.

It's the first time anyone's done dolly zoom in high speed, I'm pretty sure. You had a video where you described all of these different techniques. And I don't play with cameras very much and definitely not videos. So it was neat to really understand the difference between pan and zoom and tilt. I just, yeah, listeners, you should look at that. It was pretty cool. The lingo goes on and on too. I think I only mentioned a few of them, but the camera operators for pro video have like, you know, a dictionary of terms that just mean things like side to side, but that, you know, it's like in boats, right? It's like a whole different terminology for everything. How did you learn all this well you know i spend a lot forever on the

internet just you know yeah no i understand that one so you have a patreon patreon yeah we're something depending how i'm feeling about yeah the fans supporting me you can switch between patreon and patreon maybe right why why well you know it's um it's kind of like the death of tv right like we're basically witnessing the end of broadcast video and so it's pretty cool to actually put your money where your mouth is and say no i'm actually going to pay for what i want to see and then it gets created you don't have to wait for a network to decide to fund a television show anymore. So I think the whole concept is, it's super revolutionary. It's going to put, I mean, cable's already on the way out, but it's just like the last nail in the coffin, basically.

So that makes sense from a, I like to support people side and the people that I want to give money to when they do neat stuff. You have the other side you receive support what went into making the decision to do that um it certainly makes it easier to buy like kind of i wouldn't say silly stuff but kind of like you know if you're on ebay and you see a a weird looking device for you know you're like, it's too much to play with. But if you have Patreon funding, you could do an interesting teardown and everybody wins. It's basically crowdfunding this teardown that might be pretty interesting. So I thought it was a way of sort of literally just distributing the cost of literally what's in the video. So this linear rail that I bought from eBay was a few hundred bucks or more and the motor was another few hundred. So basically the cost of

that video was paid for by the people that saw it. Pretty cool. We recently signed up and one of the reasons we signed up was because you had one. It feels odd. I mean, we have jobs and these are our hobbies and you, so hobbies normally are a giant money sink. So it's in the tradition of sinking money, yeah. And so, it felt a little odd. I agree. And so, I usually get about, I set it up such that I pay back about 10 or 20% of everything I take in to other creators. And that makes me feel like I'm part of the ecosystem and it's helping other creators get along too.

I had a lot of people sort of request that I start taking money. So back in the days before Patreon, I had YouTube fan funding, which funny enough was actually formally announced to be shut down yesterday by, by YouTube. And it wasn't very, it wasn't used by very many people, especially the tip jar basically. But people were like, yeah, you know, I want to get in here and, you know, you should make more videos and I'm going to pay you and this, you know,

and everything. And, you know, almost at the fan's request. And then it became, you know, a pretty decent way to pay for all this stuff. So it worked out pretty well, I'd say. It's working out pretty well for us too. I did it because I failed to send a guest a mic and there was no reason why I didn't send it other than our two loner mics that tend to get shipped out around the country hadn't come back yet and and the guest said they had a mic and I was like okay fine turned out the mic they had was terrible and I whined on our slack channel

and Chris Gamble said well why why why do you not just send the mics like because this hobby sinks enough of my time and energy he was like this is why you have this is why you do the funding because that shouldn't be it and now i'm i'm set up to send out a whole bunch of mics on monday no it works out really well there's only one downside that i've encountered is that um i kind of feel like my own videos have to have a higher standard of you know quality because everyone's paying all this money in and so there's a slight like guilt factor it's like oh geez i can't make a two-minute video that's like not substantive enough because i you know see the amount of money pledged per video and it's like oh geez i can't really do that so it's it's funny the people that

were paying and hoping that i would make more videos you know it sort of backfired since i'm actually going to make fewer videos now that might be higher quality, but definitely the number is not going to be as high. That's not so bad though. I mean, I think it's nice to have some motivating factor for things that's a little bit outside and, you know, money is a classic motivating factor, but, you know, but it's also tied to guilt a little bit, right? It's like, well, you're paying me, so I better keep making stuff and I better, you know, use it as an opportunity to improve. So I don't think that's all bad. Since you asked about cameras,

camera gear funded through Patreon especially, for a while I was on this kick of increasing my production value quite a bit. You know, better lighting, more cameras, much better framing of the shots, better audio, everything. And funny enough, I don't think the fans particularly liked it. The range was like slightly positive to like somewhat negative you've sold out yeah well there's a little bit of that and then people are used to youtube having like this look yeah the homemade look yes and so

even like huge stars you know pewdiepie and all these guys it's a very like different look than television it's not it's low production value and that's actually what people kind of prefer the sound too yeah which yeah which a crappy sound and everything right and so you know it's hey that's great because that's cheaper and easier so why not this is good info if i ever start a youtube channel see i would go completely nuts into the high production value i've definitely talked to other creators that started a channel with high production value and people like what is this it looks too much like television are you funded like is this a studio like this doesn't look like youtube and so there all these questions cause people to not be subscribers and then it's

like i know why are you punishing professionalism so you're telling me we should lower our values no no no so audio is actually one of the things that um can always be done better and it doesn't sort of distract people if it's perfect right like audio is one of those special things where it's like i'm not talking about like animations or title sequences if you have really good audio on your channel everyone appreciates that and it doesn't get in the way of like you know perceived youtubeness or whatever you know oh. Oh, I think that's very true, yes. And knowing the listeners that we have, if our audio quality went way down, they would be very unhappy because that's been one of our strengths. Well, when I say our, I mean his, because Chris is all about

having really good audio. It is a nice setup. And I had one guy that would complain on every one of my videos saying, you know, your audio sucks. When I was using my, my camera's microphone and he's like, why is it so loud in the left channel and not the right channel? And I'm like, well, that's because I'm standing on the left side of the camera. It's true.

And he's like, well then fix it in post or something. I was like, come on. Yeah. Hours in the day. But you on back to the Patreon, Patreon, whatever we're going to say that um

you you actually have the goals and stuff and do you have special content for um sort of you mean the rewards like yeah the supporters yeah um i sent out stickers so that's a good that's really that's good yeah so that's a common support tier. And then I didn't really, I was already pretty strapped for time. And so I didn't really want to have a reward that would take a lot of time. So I was trying to figure out how to do that. Originally, I thought I would have no tiers. You know, it's totally, you know, egalitarian. Everyone pays what they want.

You know, there's no like tiering or anything. And then I thought, well, maybe it'd be cool if people sent me in stuff to show on the channel. Because that would be less time on my part. And people would get their sort of, you know, piece on it and everything. And so, it hasn't really worked out quite that well, but some people did sign up at those tier levels anyway and never sent me anything anyway, so. Well, that's because you didn't send me your address to send you a sticker, but that's okay, since you're in studio, there's no way you're living here without a sticker. You didn't get it? I already sent it to you. No, you sent me a sticker. I was going to send you something to put on your wall.

Yeah, I got it. Oh, I have embedded FM stickers too. I think I gave you my address. Well, anyway. Well, anyway. What's happening? Ben will be leaving here with stickers. I think if you check the latest video, it might even already be on the wall. Maybe.

Someone should watch the guest video before they arrive maybe i'm not sure but i think so so we mentioned we're having this party have have you met many of your subscribers um over the years i mean because you've been doing this for years six years this this season actually very close to six years exactly, probably, yeah. Have I met very many of them? Probably only incidentally at like the Hackaday meetup at the end of last year. And yeah, and Maker Faire. That would be the other big one, yeah.

Is it strange? To not meet them? No, to meet them. To meet them. To not meet them is just like normal, right? I'm constantly not meeting people. Yeah, me too. Well, you know, I'm not really a social creature. And so, you know, is it weird to meet them is just like normal, right? I'm constantly not meeting people. Yeah, me too. Well, you know, I'm not really a social creature. And so, you know, it's just weird to meet them.

Most people are pretty cool about it. Occasionally it gets a little too fanboy-y and that's slightly weird, but it's pretty cool. I like meeting them. Sometimes I find they know more about me than I remember ever having said. And that's a little strange, but. I have encountered that. Sometimes someone will say, you know, a video you made, you said such and such like four years ago or something like, really? I don't even remember that video.

That happened last week for a podcast that we made two weeks ago. So, I can't remember, you know, it's gone. Yeah. Who exactly were you making the fondant cake for? That was my housewarming party in Redwood City. So that wasn't a mystery. No, and that was one of my most popular videos for a long time, actually. There's more cake bakers in the world than there are science hardware tinkerers. So most of your channel, for people who haven't seen it, is you make something odd and then you talk about the science and the creation of it.

Where something odd could be miniature $1 bills because you use ammonia and pressure to just squish them. But then you have a series of videos where it's literally you baking a cake i haven't done that in a while that was that was early days right but i mean you know it's time to bring it back yeah well you know so okay so format changing is dangerous right like you've seen youtubers that try to like shift their focus and oh man people hate that more than more than bad audio even right well you you had a comment in there about and I don't usually read the YouTube comments because it's just horrible. Although you get some pretty good ones. Yeah, I like it.

But there was one in there about how you had an itty-bitty kitchen in a giant garage. Yes. Well, that's true, actually. That is true. And really, they were talking about your priorities. And I thought that was very fair. Oh, yeah. No, when I was looking for that house, I told my real estate agent, I don't care that much about the house, but the garage has to be

substantial. And she's like, well, I can't type that into MLS. And so it was kind of a long search. In fact, it was about a year and a half of searching. As Ben came over today, we mentioned that we sort of found our house by accident on the first try. That's cool. A year and a half of searching. Yes, I do understand because we did that before, but it was easier this time. That's nice. Sort of.

Okay, so moving along, a coworker of yours said that you build mechanical solutions to do things that we would do in firmware. What did he mean by that? You know, I'm not exactly sure. It sounds intriguing though, doesn't it? It really does. Was this, so potentially, so the lighthouse tracking system that's used in the Vive is kind of a mechanical solution that other people used cameras for. Probably not what this person was thinking of, but it is in fact a very mechanical solution. A lot of people are like, ah, like spinning motors and mirrors that's going to be terrible but it actually turned out to be a very successful tracking system for vr we had alan yates on the show and we talked about how that

worked and it did sound implausible because you have all of these timing requirements and this reproducibility and moving parts that shouldn't be that precise. We were actually really surprised when we fired it up the first time and got, without even trying that hard, a hard drive motor to be precise within like 200 nanoseconds, spin to spin or whatever it was. He measures it in parts per million, like how far it's off for one rev. And it's pretty incredible what you can do with just a pretty bog standard hard drive motor. It is really incredible. Of course, I should point out a pretty bog standard a hard drive motor it is

really of course i should point out they don't use standard hard drive motors there was more effort that had to go into it but um it was the starting point yeah but it's still it's interesting to take advantage of a technology from a completely different field and i wonder how many other things are out there like that that haven't been done yet like okay hard drive motors hard drives have to be very precise they can't have a lot of walk axial walk because you know they have to read very precise positions so there's got to be other stuff like that that's been totally commoditized but now okay this is really precise and we can take this and we can make something that would have been a million dollar instrument

a decade ago for a couple hundred dollars exactly and that's sort of been my job kind of all along it at most of the places i've worked is sort of finding those really quick and easy solutions that at least to get us past the prototyping stage so that did that brings up a question from a listener who asked who has a goal for 2017 to do more prototyping oh that's good that's a really good goal a listener an enemy podcaster uh do you have any advice for how he can do that wow so like specific technical advice or like mindset maybe let's go with mindset yeah that's a good one um you know i mean failure i think i think the talk of like failure has gone a little too far like a lot of people's talk about oh you know i mean failure i think i think the talk of like failure has gone a little too far like a lot of people's talk about oh you know failing is great fail fast yeah fail fast it's

great you know not failing you're not trying exactly if you're not failing you got to do this and that and it's like my take is like failure is kind of part of everything it's part of engineering it's part of life but you don't really try for for it. And so, don't try to fail, basically, is what it comes down to. And some people, I think, hearing all this talk of failure, especially in earlier education where people haven't really understood that there's nuances to the meaning of failure, getting hung up on it. And so, my point is basically, just keep going. Don't think about how you're going to fail or how it's not going to work just kind of keep putting more effort into it and concentrate on the on the day-to-day i agree

with that i i hate that fail fast thing because i think i understand its original intent but i don't think it's i don't think that's stuck for very long and the companies i've worked out that have pushed that uh tends to mean let's not plan let's just hack around you know randomly trying things with like a shotgun approach and we'll just throw out everything that doesn't work and you know kind of a darwinian approach to design and i don't think that's a good way to do things yeah exactly and that isn't really the ultimate goal anyway so it's kind of like well you know i mean succeeding is always better right so you know, I mean, succeeding is always better, right? So, you know, it's, yeah. Yeah, and you do learn a lot from failure.

That's fine, too. That's different. Yeah, it's definitely different. Yeah, exactly. I wouldn't even call it a failure. I mean, if you tried something and then you ruled it out, that's progress. I mean, that's part of how it works. There's a lot of truth to that, I think, also in research, where a negative result is not something that is desirable, and yet it could be very interesting and publishable but people want the positive results exactly so they

actually almost the same problem that silicon valley does where they're like so i guess maybe it's the opposite of the inverse problem whatever but they they won't publish negative results and so then people figure oh yeah well you know i can't ever even though that is valuable stuff yes because if you don't publish then other people go down that path too and then they don't publish and then it's all a giant waste of time so let's also talk about the paths that don't work yeah exactly and if the cost of publishing were way lower and you could just easily access everything it would be great and so we could rag on like science journals for a while. Let's go back to actually,

no, let's stick with failure. How do you deal with it? Say in terms of applied science, your video channel. I'm sorry to interrupt, but have you had videos that you've tossed? Yeah, sometimes the making of, I usually shoot B-roll footage first.

For viewers that don't know, B-roll is sort of like action shots of me using a lathe or something. So there's no audio or there could be audio, but it's not that important. And then A-roll is me staring at the camera and talking. So I always shoot all the B-roll first as i'm doing the project then shoot the a-roll at the very end and then splice in parts of the b-roll to make it you know more interesting so i have b-roll that i've never made into a video and um usually i'm so strapped for time that if i go through the trouble of building something i'm going to squeeze a video out of it pretty much but um there's there's definitely a bunch of B-roll footage that's on the cutting room floor for sure.

Yeah, but I mean, that's kind of by nature. Of course, you go ahead and set up a camera because you're working anyway. Why not? Yeah, sometimes I forget, though, and then I don't even have as much B-roll. And you can tell those videos because they'll be like a shot of the camera just looking at a desk or something with audio. I mean, I'm talking over it, I'm narrating it, but I don't have any B-roll for that section. And so it's just nothing basically.

So you have never like mostly finished a video and said, wow, this is a really boring topic. What was I thinking? Have I ever gone that far? Probably not. No, I think once I'm that far along, there's no turning back. At some point, you just ship it. Yeah. Now, I have sometimes made videos that I didn't charge Patreons for. So, I have the option

of posting a video and then saying it's a freebie video. And I've done that a few times. That actually might be the best solution to do the short ones. I was thinking the two-minute videos, you can absolutely do that. Exactly. I think that's probably the best balance. I should do that more. It's fun and it keeps things going. I know with the podcast, over the holidays, we ended up recording a whole bunch in a short time. And then we had big blank spaces on either side. And there was some amount of getting back into the pattern.

And so, yeah, there's some. Yeah, definitely. I used to have a pretty sweet work schedule where I was doing four days a week at my day job. And so the extra day I had for videos, and I didn't have to, but sort of strongly incentivized to give that schedule up. And so now it's really crunched.

I had way less time to make videos. Was that part of your negotiation? Yeah. Well, you know. It's like, I'm going to do, I have this video thing. Is that okay? I mean, did it come up? Oh, absolutely.

And so when I was working at Valve in Seattle, because of the geographical distance I was here, they were in Seattle, I negotiated this day here to sort of finish up things. And it just sort of got grandfathered in. And then somehow I transferred that over when I started working at Alphabet. And I know that when you started working at Alphabet, you already had the videos. They didn't. That was, I mean, that was one of the reasons they knew you. Exactly. And do they still support that now or do they want all of your attention?

No, no. They want the five days a week, but I can, you know, make videos talking about technical stuff. They're totally cool with all that. Yeah. They just want all your time. Well, it's reasonable. I actually don't feel overworked. And I think Verily has a really good work-life balance. It's just having that extra day is like a really significant thing.

It's like, even if it were like four, 10 hour days, that would be better. Cause it's really like having that whole separate space to like play with. And so, you know, it's, it doesn't really work to be out of the office an entire day a week. And so I can kind of see their point. Yeah.

Almost. Well, like I say, they made it a pretty good, it was, it was a very tempting offer. So yeah, it was,

it worked out. But it's a tough balance. It was. It is. In fact, they must've gotten the number just right since it was it was a very tempting offer so yeah it was it worked out but it's a tough balance it was it is in fact they must have gotten the number just right since it was like man this is a tough choice so so you do have a full-time job yeah and you have the videos how much time do you put in to the videos um not if you know it's not so much anymore i mean at the peak it was almost a video a week which was absolutely nuts even at one even at having a spare day i mean even that was pretty crazy i think my average probably during the the heyday was maybe a video every other week and now it's like a video a month you know or even less so um but you know weekends are good if i get like a full saturday that's a really good time to kind of like a video a month, you know, or even less. But, you know, weekends are good.

If I get like a full Saturday, that's a really good time to kind of get a video nailed down. Yeah, if nobody convinces you, you need to go to the beach. Right, right. How do you choose what level to explain things to? Sometimes you take very complex scientific things. Do you have a mental model of your audience member do you have somebody you actually are explaining it some person that's actually yeah it's an interesting idea i don't but i like that idea a lot i think that actually i'm going to use that in the future actually

i'm not thinking about explaining it to a person that i know but i am using my own frustration at having things explained to me in very convoluted ways so like electrical impedance was like a perfect example oh geez i know i still can't get that stuck in my head properly it's like why do they make it so complicated so in in school um i had to take a few ee classes even though i was an me in school and it's like okay we get to ac circuits and they have to explain impedance and they do it in such a way that no one could possibly understand right the first time they come out i mean they're using imaginary numbers they're using really complex math i mean it's like why you could just talk about it in sort of

terms like just what is it it's resistance that depends on frequency it's it's really that yeah they never said that they never said that they never will either and so it's very frustrating and so i actually get a kick out of explaining things in really simple terms because I wanted to explain it out to me when I was learning it. Everybody does. Yeah, I know. So why does no one do it? antenna so the interesting stuff about impedance is about reflections and efficiency so they just skip to that and then ignore the easing you into an analogy that makes sense you know from basic electronics yeah because the plumbing analogies with the little pipe and the big pipe not not really you don't like that one it's not the worst but then you lose then you get this mental idea

that it's a block instead of having it depend on frequency i see um but then i don't have a mechanical right implementation of impedance like that it does start to get the analogies start to break down when you push on them too hard so yeah it's there was a great video on youtube i found and it's from either the 60s or the 50s or 70s something very old where a guy is explaining impedance. I think it's with Bell Labs. I'm going to find it. But he has this mechanical model

where he can make the waves go with this, it's kind of like a cascade of tinker toys. Nice. And you can see the reflections and he talks through the whole thing and it was a really good explanation of that aspect of impedance because you could actually visualize it and you could lock things down and change change the basically

the impedance of this mechanical system um i'll find that it was a good video you're saying instead of asking what is your least favorite electrical component we should be asking what is your least favorite electrical property yeah i mean yeah and people would say impedance because they don't understand it but yeah that's another thing you know a lot of these like 50s and 60s videos were actually way way more uh clearer than modern educational equipment when it's like well that's i mean yeah there's something about you know they had to work really hard to make this mechanical model it was very large probably it was very expensive and it explained it better than you know somebody putting up a video screen with you know lots of colors and waves going around everywhere i mean it looks

fancy when you're doing that but this was like oh this is very simple it takes up a hotel whole table but i like it yeah i really like that kind of stuff my second career was always going to be in education too and maybe it still will be. It is partially already. Yeah, exactly right. You're headed in that direction for sure. Yeah. It's funny. I've always thought that education and how we explain things does fall to a Darwinism sort of thing.

The better you can explain something, the more likely somebody else will use that explanation to teach the next person that's good yeah i like that and yet somehow impedance has well there's cracks in the facade yeah there's like an island over here with good impedance explanations i would say a whole lot of basic introductory physics is broken too it's like an emphasis on mathematics which i don't quite understand and it's it's useful to use math to solve problems but i don't think it's very useful to use it to describe sort of how to understand something at least not for me yeah i i would agree with that i like math but i like it second to concept exactly uh and you said you were a mechanical engineer for some reason i thought you were electrical i know software so that that this question of how did you learn all this mechanical stuff

is sort of pointless isn't it how'd you learn all this electrical stuff yeah exactly well i actually wanted to be an ee but the math killed me now we see the truth no it's and it is true um I never actually formally signed up as EE, but I mean, I knew what was going to happen when I started looking into it. And so I was like, yeah, I think we'll go with mechanical. What's funny is the opposite happened to me. I started as an engineer and said,

I'll go back to math. Really? I'm kidding. Huh. Oh, so you must, and you don't even like the mathematical descriptions of these properties. Um, I think they're useful depending on what your path is in education. I think they're very useful if you plan to go, you know, very deeply, but I think for people who are, you know, stopping at a

certain, at another point, like, oh, I'm going to be an engineer. So let's get these physics concepts down. I think it depends on the audience. and I think there is no tailoring of it to the audience, especially early on in undergraduate. Cool. Okay, so for people who are not mechanical engineers, how can they gain more facility with doing these things like you do? Yeah, that's good.

I think having a good space to work is pretty important. So I've always had the benefit of having like a really big shop. So even when I was growing up, my dad had a really well stocked garage, lots of tools and space to just play around. And if people don't have that space of their own, you know, finding a good maker space or something like that could work out pretty well. But sort of having like a dedicated spot that you can always go to and tinker out there is a pretty important part of understanding this stuff, I think. That's an interesting advice. I always try to like play with origami on the couch,

and it really doesn't teach me the mechanical things I want it to teach me. So, yeah, going to a space and having the physical tools would help. What tools do you think are most important? I like to have a milling machine. That one's probably the most versatile. And it's true that, you know, it doesn't work as well on sheet goods as a laser cutter. You know, maybe it doesn't do 3D printing as well, because you can't have undercuts with a milling machine. But generally, you can get almost everything you want out of a decent milling machine. What's a decent milling machine? Well, again, it's sort of a sliding scale. So ideally, it would be a pretty hefty three axis, you know, 4000 pound machine. But if you can't have that,

some of the desktop ones are actually not bad these days do you have another mill uh no actually i don't okay that seemed like a tool that i would expect you to have because it is desktop and it's very oriented towards the maker community yeah i think i i really like that tool for pcb milling actually it's a pretty cool application for it i'm spoiled i already have a bridgeport c in my garage, so kind of. Who needs a tiny desktop when I have this monster? Yeah, and I mean, well, actually, so it would be better for PCB milling, another mill for sure. That actually would be pretty nice, yeah.

Do you have a laser cutter? Nope. I've got a nice one at work. And so, again, if I need to cut something one or two times, it's not, yeah. There is some advantage to working with that company. They do have some nice tools. Yeah.

Yeah, and Valve too. I mean, they, you know, employees can use the shop for, you know, little projects here and there. So what spiffy tool are you excited about for the future?

Have you guys seen this Wazer? This, like, desktop water jet cutter? What? I know. Oh, no, that jet cutter? What? I know. Oh, no, that doesn't seem likely possible or safe. No, no, no, it's real. In fact, it's a Kickstarter, so it has to be real.

It is, in fact, real, and they have videos of it working. I'm not sure precisely how well it works, but they successfully Kickstarted it. I almost signed up. I was really close to getting on the Kickstarterstarter it's like maybe four or five thousand dollars or something like that so it's pretty hefty machine um it's desktop in the sense that it's at the height of a desk but it's it's very heavy of course it's a water jet um i would like to have that if it works as well as i think it does i mean that's pretty cool so water jet works by shooting very very very high pressure water at whatever it is you want to cut yeah or kill

sort of like a laser jet but much much more so yeah the cool thing is that it doesn't care what material you have and so it'll cut stone metal paper glass glass foam i mean any like hardness any like material you've got as long as it's too thick and even that is pretty i mean it can even do the big ones can do like five inch thick slabs even so there's almost nothing it can't cut and it also doesn't push on the work like a milling machine does like the cutting force is almost zero so that you can do really intricate stuff and you don't even have to clamp your work down you can just put it on there it works just like a laser cutter it's kind of magic like it's one of the most like magic-y tools out there and to have like a desktop one for five grand is actually pretty awesome the the big ones cost two three hundred thousand dollars it's just

not even close for you know that's because generating that much pressure yeah and having enough uh stuff on the side so that you don't cut through and hit the people on the other side it's pretty cool and so i've looked into i actually bought a nozzle for a water jet cutter and was thinking of connecting it to like the biggest baddest um pressure washer that i could get just to see what would happen and if you could only cut like eighth inch aluminum that would actually still be pretty interesting right by hand um initially but i was going to get a cheap cnc machine and kind of test it out but you should do it by hand well i was of course this is youtube i mean of course we're gonna like you know and i i got close i haven't done it yet but i've got like most of the parts except the

actual water jet pump okay so could you please add like a blue laser to it a blue led to the near the nozzle so that it looks like a lightsaber yeah that's you could it might be able to do that actually um the amount of power in those things is staggering like it's like 40 or 50 horsepower for most of the largest you know mid-size water jet cutters and when you think of this in electrical terms i mean the motor itself is like you know 500 pounds of motor just to generate that much power so it's i don't know what you're going to get with just a plain old pressure washer but you know i think these i think the waser guys basically did this and they got a nice you know pressure washer motor and it doesn't cut quite

as well as the big machines but you know so what for eighth inch aluminum it might be just fine and sure would be fun to try i can i can totally see that yeah another thing people may not know is it's actually abrasives that do the cutting so the water is just sort of like the motive for us it's actually a little sand they actually add sand to the jet oh do they yeah so it's it's actually an abrasive jet cutter um and without the sand it won't cut hard materials that makes sense then we say of course but they do use a water only to cut things like um food like beef and cake and all kinds of stuff in industry because then you don't want the abrasive in there don't you have to submerge it how do you cut cake with a water jet and not

have soggy cake you'd be really surprised it's it's on like a it's on a these are the deep penetrating questions well you know i wondered about it too so we had we had like foam inserts water jet cut uh at valve for a while ago and they came back you know bone dry and like how did they even like what did they dry them out afterwards or something but if you put the piece of work on a spoil board so that the water kind of goes through and doesn't come back up, like splash back up very much, the jet is going so fast and it's so concentrated that it's literally like a knife. It doesn't even get the work very wet. All right. Well, I know what I want to use to cut my next cake.

No. So, how do you stay current? how do you learn about these things um i think twitter is probably oh i read twitter a lot and i'm sorry well you've got to know how to use it right like you gotta it gets worse though unfortunately i also read also read Hacker News, which is- I'm really sorry. I know. So that's actually worse in a lot of ways, but I can't stop doing it.

In fact, it's bad because I'll have like Hacker News typed into my computer before I'm even fully down in my chair yet, you know? And, you know, I mean, I don't really like listening to people argue, but for some reason it's okay when it's on Hacker News. But I mean, there is a lot of good news on there too. And that is how I get a lot of information.

Okay, so for Twitter, because I'm on Twitter, I don't find Hourly Kitten to be exceptionally good in making me current in tech. Who do you follow that is good? I think people can check my stuff. Yes, but you're not going to... Who do I know off the top of my head? Who posts good stuff, basically?

Oh, geez. I don't know. We'll ask you after the show. We'll come up with some answers. And Hacker News. Hacker News is Y Combinator? Yeah. News.ycombinator.com.

And so it's mostly, I like that because it's a little bit more neutral in a lot of ways. Even though they somehow managed to still have lots of flame wars, there's no politics. There's absolutely no political news on there. Really? Because I don't follow very many people on Twitter anymore. No, it's just angry tech people. Yeah, so they're very opinionated, but it's not about, at least they're not, I mean, you know, occasionally a topic will drift into it, but there's no headlines

where it's like, you know, what did Trump do now or whatever, you know, it's like, I mean, that's totally off, which is very refreshing. And so, you know,

so that's nice. In fact, it's the only news source that I know of that's filtered in such a way that it's literally only about tech and business news. So, that's nice.

We occasionally post our show there, but we don't have any particular like automated process so it is only occasionally it's there's a lot so back in the day it was um much less heavily you know there was less competition to basically get on the front page and now it's just insane i mean it's yeah we've never gotten on the front page it's it's it's tricky i mean's, and then even like big news stories will get bumped up and down.

And if it's submitted multiple times, all the votes get spread out to like all the different ones. And so, yes. Any place else you'd suggest people look for? Let's see. It's funny that social media is on your list. Only Twitter.

I have a Facebook account that I haven't logged in in probably a year or something like that. I have Tripoli Magazine. I actually like iTripoli Spectrum. It's still a good magazine.

They occasionally infuriate me. They were very against electric cars. Yes. Or maybe still are. But I just knew that was probably the one because I remember

looking at that too. Like, what do you guys mean, man? Really? And articles on and on about how it was never going to happen and they were horrible. And then there'd be like a tiny blurb about Tesla has sold a whole bunch of them. Yeah. It was just sort of boggled. I thought that was kind of weird too.

So yeah, occasional stuff like that. But they have good stuff. Sometimes they have really good stuff and youtube too actually um a lot of research uh institutions have a youtube channel and so they'll post their stuff at the same time to youtube that they do to like you know pro news outlets and so even just watching youtube is actually pretty good that's a good point do you find have you connected with other youtubers in any way? Is there like a little cabal of... Yeah, so if you're into that, VidCon is where you go. So VidCon is the conference for YouTube creators.

It's in LA, I think, once a year. And I've talked to people that have gone. It doesn't really sound like something that I might like that much. I mean, I probably would, but... You might have to meet people. Well, I know that's one downside. It's a little like high school where it's like you know if you meet someone they'll say oh uh you know how many subscribers do you have and you say oh you know

about 800 they're like oh i'll talk to you later i'm sorry i'm gonna go over here now yeah 800 million yeah right is that you mean 800 oh okay well so there's a little bit of that but um i have spoken to other youtube creators a fair bit over you know back channels and email and stuff and usually we're discussing things like hey did you get ripped off on facebook again yeah how are you dealing with it oh do you have a name of a lawyer that we could talk to to prevent getting ripped off again it's mostly yeah that kind of help. Oh. All right. I can see why you wouldn't run out to do that. Yeah.

I mean, occasionally there'll be a nice call out. Like occasionally someone will say, you should check out this person's channel and we'll have a few back and forth and stuff. It is pretty neat. Cool. Yeah. And I'm going to ask you for a couple of channels you'd recommend. Oh, yeah.

But I can do that offline. Okay. I'll put it in the show notes. Good. Do you have any advice for mid-career engineers about getting burnt out? I mean, you do a lot of stuff and you always seem enthusiastic and passionate about what you're doing. Yeah. So I have to admit, I got pretty lucky.

So the job I have is all about prototyping, which is what I really like to do. And so that's, you know, not everyone gets kind of the, you know, the gravy jobs like that, right? But I mean, I, let's see, advice to not get burned out. To avoid burnout or to stay connected to technology. Right. I think, I mean, you know, it sounds redundant, but maybe watching YouTube is a good way to see what's exciting, right? So, the nice thing about YouTube is that it's self-filtering

where people don't make videos about stuff they don't like. And so, it's kind of inherently exciting because people only talk about what they're interested in as opposed to like the real world where you have a job and you might be forced to do something and so it's not as it's unfiltered basically um similarly going to like a meetup the only people who would go are the ones that like talking about this kind of stuff so they kind of sell filters for people that are already enthusiastic maybe i don't know i have resolutions new year's resolutions about going to places and meeting people it's not that i like it it's just i have to be there because i made a resolution yeah i mean like you say i'm totally not a social person so i i completely understand and i think you can still communicate with people asynchronously too i think that's actually one of the best things

about you know the internet of course i do think you're where do you go to learn things and and your how do you stay interested or connected because you you mentioned the same place for both and i think some of that is one way to stay connected is to learn stuff yeah yeah like the hackaday uh meetup was pretty good at the end of last year that was a big one a fair number of youtube fans there um and also old friends too actually that i had worked with in the past. It was a pretty nice meetup. Yeah, I heard some Valve folk were there. Yeah. Sounded fun. Was, yeah.

So you were talking about prototyping. Can you talk about your BLE prototyping techniques? Yeah, so this is pretty cool. This is actually something I recently did, too, so it's all fresh in my mind. For a lot of projects that I do at work now, our job is to get sensor data from a little battery powered device into the cloud, basically. And one pathway is from a sensor through Bluetooth into a phone, which then uploads it to the cloud. So the link that we have to work on is sort of

the sensor and Bluetooth into an Android phone, basically. And so I came up with this framework to do that. And it's really easy. So I don't actually write that much code. Like I definitely stay more in the engineering, like hardware mechanical side. So anything that saves me writing code is like a really, that is exactly what I want. So I found this perfect solution that I don't think many people know about. The sensor itself is Philip Frieden's OSH chip with a Nordic NRF52 on there. And that's great because you can program it through embed. You don't even need a programmer. You can just connect.

Well, actually, I use Philip's programmer. But you don't need to have a compiler setup. Embed is an online compiling tool. And then it's cool. You plug the thing in, and it appears like a USB drive. And you just drag the file that the compiler downloads into your browser onto the thumb drive and it programs the device, which is pretty sweet. So that, I mean, it's easy and they already have, for example, like an Eddystone broadcaster. So you can just drag and drop that into your project and now you're broadcasting Eddystone packets.

I mean, it's a little bit more work than that, but it's pretty good. The advantage to Embed is there is a lot of code out there that's pretty simple to use and the nordic chip has a lot of uh example code that is relatively straightforward to if you don't want to modify it exactly you want to modify it gets more complicated it's true and so i'm not yeah so i i like c i i don't like c plus plus and unfortunately most of the embed stuff is like full-on heavily abstracted like super out there and so that wasn't so great but at least it's working and it's easy to program so that's that part done so now you have a sensor sending ble packets and on the phone side i don't know if you've ever tried to program an android app it's yes it's heavy duty yes really heavy duty and so

i was like gross no way i'm gonna do do this. So I found a perfect solution. Processing.org has an Android mode and you can use processing to make an Android app in five minutes once you get it installed. Processing is the C++-ish language

that is used to program Arduinos. Yeah, and it's Java in Android mode even. Okay. Yeah. So you download the JDK and the Android SDK

and this and that, and then you basically just say, send it to my phone, and it literally just shows up on your phone when it's plugged in through USB debugging, which is slick. Now it gets even better. Now there's no Bluetooth, of course,

because it's just an Android app and processing doesn't know about BLE. But I found a library on GitHub of this guy that wrote a ble library for processing.org in android mode why he didn't compile it i don't know so on github it says oh all you have to do is just fire up eclipse and compile my library and away you go i i almost died because that was i mean that that almost killed me downloading eclipse and getting his library compiled but i did it. And so now I have

BLE working in processing.org and can make an Android BLE app in three minutes. That sounds pretty cool. We're going to need more links. Yeah. And even better, once processing is, it's on the internet, so you can take that data and do HTTP get requests and pump it into the cloud with basically no effort. It does. I mean, it's, yeah, it's, it's good stuff. Neat. I'm glad I asked. And we have had Philip on the show. So if anybody out there is wondering why they've heard of this, it may be that, or maybe you've heard of it somewhere else. And I'll have that show link in the the show notes clearly i am losing the ability to talk so let me ask a few more questions do you have a favorite video maybe something that is

usually overlooked um a favorite video that you've done just to be clear we're not we're not talking star wars or anything that'd be okay well um i think my favorite would have to be my first electron microscope video because that's really the one that launched this whole endeavor so that was like early 2011 um yeah well so i had videos before that but that was the first like really big one and um that was kind of what started the whole thing probably the reason i'm talking to you now is because of that video so it's uh yeah it's pretty cool it made it to hacker news but funny enough it doesn't have as many views even now as some of my more recent more popular videos because youtube has grown so much in the past few years and they don't they don't prefer like old videos like they they recommend mostly recent stuff on the idea that yeah i mean

there's it's the algorithm there's reasons for that but yes um it is funny when somebody writes in and says have you ever done a show about x and we're like yeah that was show 11 well maybe we could do another one it's it's been a while uh let's see do you have a vr system i know you worked on valves vr system but do you actually have one yeah so i have an hdC Vive Pre and my computer at home wasn't powerful enough to run it so I brought it into work and we actually have a dedicated room setup. So originally I set it up at my desk and there was literally like a line of people

out the door waiting to use it which was fun but super distracting. So we actually set up a dedicated space and that's worked out really well. What do you play? I like the creation tools so Tilt Brush is one of my top favorites. Yeah.

That keeps getting better and better. Yeah. I need to go back and retry that because I keep adding stuff to it. So that's the one where you have a paintbrush or sort of clay and you can make your own whole art. 3D drawings. 3D drawing. Yeah. And I did this weird poster place with a tree and some water and it was really

fun and at the end my arm was very sore because you actually have to paint you have to move a lot it's pretty cool to tell someone you can paint fire in midair and it will just hang there i mean people instantly get it and it's super enjoyable right from the start very intuitive yeah and you can put stars in the sky and they twinkle. It's really cool. Now you can, I think they recently changed so you can change how big you are. Oh. So you can work on fine detail and then, you know, work.

So you can build a whole city. Actually, I haven't seen that. That sounds pretty awesome. Very small and shrink down and do fine details. Nice. That is really cool. I feel like we're going gonna need a camera to go inside the VR system

pretty soon I know that probably exists just to see you can see what someone's painting well like like you do with the

Kronos camera but now we need it inside oh oh I see a camera like on a rail or something yeah oh yeah huh

I don't know what that means I don't know what that means I don't know what that means because it's all head mounted and that's where the camera lives Soulbrush has some interesting

they have like a like a it sort of tries to give you perspective like it'll kind of move your viewpoint back and forth and it'll make

an animated gif of it to show the 3D-ness of it on like a on a flat monitor oh if you export it yeah exactly that's kind of neat

it'd be like a 20 minute gif sometimes oh no no it's just back and forth oh oh if you want the yeah exactly that's kind of neat it'd be like a 20 minute gif sometimes oh no no it's just back and forth oh if you want the full fly through yeah i don't know if they can do that just yet maybe maybe cool have you seen the new uh google earth app in vr no actually i haven't i i was a little sad that they don't have um street view yet oh yeah because i've heard they've done street view of the Louvre, and I wanted to go walk through that. I was just so jazzed by that idea. But you can go to cities, and you can step on the Eiffel Tower,

but of course it doesn't have any destruction, which is another feature I would ask for. But you can't, I mean, you start out kind of in orbit, and then you can zoom down to arbitrary size. Then you can grab the sky and move the sun around and stuff. It was kind of cool. In different times of day. It was pretty neat.

Sweet. It shows you just how dense some cities are and how huge other spaces are. It was pretty cool. Yeah. In five years, you'll be doing 3D VR videos of your stuff so people can look over your shoulder while you're doing stuff. Some YouTubers have done full 360 videos to kind of play in with the form. Some people, Frank Haworth is a great sort of pioneer of interesting YouTube formats. And he did a 3D video where the camera was sort of like in the middle of his workshop and he was working

like around it, like the table saw was on one side, you know, the shaper was on the other side. And so he would walk back and forth and you could actually like turn your head back and forth following him around the shop it's pretty creative there's so much on youtube you need to get another chronos camera put them side by side and then and do 3d high speed that would be that's a really good idea you should tell david before because he's gonna have so david's gonna have like hundreds of cameras before he sends them all out to the backers. And someone's like, you got to do interleaved framing where you could have a billion frames per second. All right.

With that thought, I think I'm out of questions. Chris, do you have any? Questions? For Ben? For Ben, no. No, I don't think I have any more questions. What about you, Ben? Do you have any more questions what about you Ben do you have any last thoughts

you'd like to leave us with um we didn't talk about self-driving that much but I think that my

final thought is going to be um how the bar for AI always moves like people say do we have

artificial intelligence no but if you had said you know 15 years ago oh yeah self-driving cars

will be around in 20 years. No problem. They're like, that's crazy. We don't have AI yet. So the bar for AI is always higher than where we are now, even though we've probably crossed it many, many times over. Yeah, we'll have fully, you know,

humanoid androids walking around talking to us and people, that's not real. They're not really intelligent. Right, because they're not really creative or whatever. They'll always have an excuse. They're like chess playing or go playing. Like, oh no, no, that's just an algorithm. So they'll be saying that while you're not really intelligent. Right, because they're not really creative or whatever. They'll always have an excuse. They're like chess playing or go playing. Like,

oh no, no, that's just an algorithm. So, they'll be saying that while they're exterminating us. Yeah. Those are really intelligent.

It's true. I actually think it's, well, I mean, I don't have to give you exterminated, but it's true. The bar will keep going up until it's at ridiculous levels,

I think. And at one point, it'll be above our intelligence. Yeah. Therefore, we won't be classified as intelligent. I did fail to ask you about cars because your car doesn't have any driving, self-driving capability yet.

Not yet. There was an update last night and I was like, oh, right. You know, and I installed it and nothing happened. So, yeah, that's what, that's what you get for being on the bleeding edge. I know. I know. Well, maybe we'll take you out and see if you want to try our self-driving opportunities. But if we do that, we have to stop recording.

So our guest has been Ben Krasnow of the Applied Science YouTube channel. You can check out his Patreon page by searching for Applied Science on that site. Thank you for being with us, Ben. Thank you. It was great. Thank you also to Christopher for producing and co-hosting. And of course, thank you for listening. Do sign up for the party. The RSVP link is in the show notes. If you'd like to say hello, hit the contact link on Embedded FM or email show at embedded.fm. You can also subscribe to the YouTube channel or the newsletter. Those are all on the site, embedded.fm.

And the blog. And the blog, yes. I'm sorry, I shouldn't forget the blog because the blog is very cool. And now a final thought. This one's sort of convoluted. It's from Henry Marsh, a doctor, a neurosurgeon, in his book, Do No Harm. On one of my regular trips to the neurosurgical department in America, where I have an honorary teaching post, I delivered a lecture entitled All My Worst Mistakes. It had been inspired by Daniel Kahneman's book, Thinking Fast and Slow, a brilliant account published in 2011 of the limits of human reason and of the way in which we all suffer from what psychologists call cognitive biases. I found it consoling when thinking about some of the mistakes I have made in my career to learn that the errors of judgment and the propensity to make mistakes are, so to speak, built into the human brain.

I felt that perhaps I could be forgiven for some of the mistakes I have made over the years. Embedded is an independently produced radio show that focuses on the many aspects of engineering. It is a production of Logical Elegance, an embedded software consulting company in California. If there are advertisements in the show, we did not put them there and do not receive money from them. At this time, our sponsors are Logical Elegance and listeners like you.