 And welcome to DeskLadyAda. Hey everybody, and welcome to my desk. It's Lady Aida. It's a Sunday evening. It's a hacker o'clock, Lady Aida o'clock. And I thought we would spend 15, 20 minutes at my desk, and then we have a 15 minute great search as well. And we'll just see what's going on on my desk. Yeah, we got a lot of cool samples. Let's dive right in. Okay. So first up, so let's go to my computer. And I will go to Aida for where I'm dealing with Puppa. I don't know if you knew, but there's a part shortage. So the LSM60S33 is a part that we've used. It's a very low cost IMU. Seen here, it's partnered with a magnetometer. The magnetometer we got in, but the IMU we still can't get. So it's the LSM60S33. It's discontinued. It was a very inexpensive, low cost, but fairly good quality IMU. So it was good for basic sensing because IMUs can easily get into like $6, $7 range. And this one was like a couple bucks. So I really liked it. And it is featured in a couple of things. First off, of course, we've got two breakouts. But it also made it into the Feather Sense. Had this chip here, the LSM60S33. I don't know if it's listed here. I think we just say an IMU. But yeah, it's a XL gyro for the IMU. And then also the clue, which we do have some leftover in stock. This one also has the XL gyro, but then, you know, we can't get it anymore. So looking at other IMUs, I found that I could get an ST still, ST doesn't make the LSM60S33, but they do make the LSM60S3. And they make a couple of different versions, including one called the 3TRC. I don't know what the C is, but it's basically an entry level, low cost IMU, which is perfect because, you know, again, I can get a really high quality IMU, like the ISM3330 is in stock, but it's like $6, $7. And I know it would basically double the bill of materials cost for something like the Feather Sense, which is supposed to be inexpensive or the clue, which likewise, you don't want to keep it inexpensive. So I picked up some of the LSM, so this is the desox, and let's just look at the front. So the, what is nice is ST decided like, basically every IMU after the DS33, they're like, okay, we're gonna do this kind of standard QFN, DFN with, I think it's three pins on the top, four pins on the side, and it's kind of rectangular, but they've kept the same pin out. So if we go to the overhead, which is a little bit clunky here, this is a, this is actually an LSM6 desox breakout, but I removed the 6 desox and I soldered on the DTR, which I've got here, sorry, this one, the LSM6 DS3TR, see? And the register map did change a little bit. I will notice all these chips, they're pin compatible, but like registers, things will move around a little bit. That said, I did get it working. So going back to the computer, I've got my Arduino working here, and you can see it's got temperature, accelerometer and gyro data coming out, and it's just fine. So, and I've got Potombo working and I got the shake motion detection working, because pretty much all I used the LSM6 DS334. So I do have to do a board re-spin for the feather sense and the clue, but the good news is that the sensor is available. So, you know what I'm gonna do is I'm just gonna purchase like basically two years worth because I know I'm gonna use them and I know that this is just part George's an ending. So I might as well get a year or two's worth of this chip so I can keep the feather sense in stock and the clue in stock and also, you know, I'll revise those two basic breakouts so it can have a low cost IMU breakout. One thing that's interesting is as I was looking, because I went to st.com and I think I looked for, I don't know if it's under sensors, MIMS sensors, and then I think I looked at iNEMO, which is what they call their gyro accelerometers. So one thing that was interesting about this is everybody's favorite, hold on, 9DOF sensor, the LSM9 DS1 is, I don't know if you can see, it's very small text, but it says, and it's like, hi, I wanna resize your screen. It says NRND on there. So, you know, if you're using the LSM9 DS1, I know that Arduino using a couple of boards, I know other people have used it. It's adorable because it's like a 9DOF IMU and it's all in one and it was pretty inexpensive, it was a couple dollars. It's basically being discontinued. So, you know, thankfully, I didn't use the 9DS1 in these designs, so it's like I'm only changing at one part and the part's changing to something similar, but it is kind of a downer. Anyway, so I ended up looking through here for what was available. The only thing that's a little annoying about ST's part selection site is they don't let you search for pricing, whereas, I will say TI is really nice, it'll give you budgetary pricing. Budgetary pricing isn't the actual pricing, it's always gonna be a little bit last, but at least gives you an idea like is this about a dollar, is it about two dollars? So I had to look up each one of these chips to see how much it really cost, and that's how we ended up with the LSM6DS TR. So it's a revision party always here, but thankfully got this firmware going, so I feel confident I'm gonna spin the clue, gonna spin the feather since, luckily this new center's smaller, so I'll just like, space will just change a little bit, but the firmware should be very minimal, and thankfully you can detect which one you have by reading the, there's a who am I register and it gives you different values, so you'll be able to know which sensor you've got. So I've seen this in a bunch of different boards, boards that have sensors, they're kind of constantly swapping them around. So that's the LSM6DS33 swap out, and then the reason I was actually kind of forced to do this is I wanted to, you know, they're revising the feather sense, but then Scott also was like, oh, you know, we really useful to have a NRF 52 feather with a TFT on top, like the ESP feather that, you know, we haven't made, it's like got the TFT on top, and on the bottom there's the ESP32 S2 or S3, and I was like, oh yeah, like, you know, the NRF 52 module is much smaller, there's even a little bit of space, so I thought I would also put some sensor pads down. So this kind of like forced me to like, okay, I gotta swap out the IMU for that LSM6DS chip because, you know, I wanted to put IMU here, and I was like, well, you know, now I'm like, I have to like get it together and do the swap out. So this will end up revising like six different boards, much like the CP 2104 to 2102N, it's just kind of life, you know, sensors and chips. I'm noticing that, you know, it's not that, it's not just that you can't get certain parts, but there's acceleration of, if it was going to be end of line, then the end of line just comes much faster, and if it does get end of line, you're not gonna be able to necessarily do a last time by, so everything is just much more sped up. Okay, what else did I say I was gonna show people? Well, we have samples that, I did a little bit of a preview on our Twitter, is it's the filament stuff, we got, and, Oh, did you wanna do chip shortage? Would you do that later? No. Right, sorry, correct. Matt, can I do it? No, yes. Let's show some filament samples, it's been a day. Let me get my samples. Yeah. And, we're gonna do chip shortage this week on Ask an Engineer, so we'll just be, we'll just do the one that you sent me this week. Yeah, that's cool. Okay. So it's done in advance. Yeah, yeah, yeah. Yeah, let's do that. Don't worry, there's always no parts. Yeah. Okay. Yeah, no parts is still a true statement. Okay. Okay, so let's go to the overhead, and we'll check out these cool filaments. So I got, he has funnies, I've been trying to get filaments for, like literally years. We actually got some LED filaments, literally like five, six years ago, but they were 70 volts. And I was, I actually was about to put them in the store in the last minute. I just said, like, you know, nobody's gonna wanna deal with these 70 volt filaments. So I paused, but then I found a supplier that would sell me a small quantity of three volt filaments. And I was like, that's what I'm talking about. So these are kind of neat. These are, you know, they look like spaghetti noodles, but they're actually LED, there's a ladder of, hold on, let me get some, wires. Wires are good. Let's get a power wire here. Let's get a ground wire. And we'll just, they can run off of three volts, which I really like. So what's interesting is that if you look, you know, you're not gonna be able to see it very well here at all. But on the back, there's like this little ladder of a very thin metal piece and LED diodes that are bonded on and then I think the side with the hole is negative. No, the side with the hole is positive. That's what I meant to say. Okay. So this, for example, is a pink color and you see you get a really nice, smooth color effect. You know, I'll hold it up here so people can see what it looks like. You know, not blow it out as much. Yeah. So there is, if you look really closely, there is a slight, you know, dotting to it, like you have to be really close to see it. It's definitely like one LED per millimeter, maybe one every one and a half millimeters. And what's really nice is that it's like, it's really flexible. I mean, like it's, you know, you can pretty much move it just like EL wire, it's even more flexible than EL wire. And it's got that full color to it. Like it's a much more solid and it doesn't have the buzzing sound and you don't need an inverter. So for small illumination or, you know, a decoration, I think LED filaments are kind of neat. Okay, so this one is green and now I know the side with the hole is positive. Okay, so this is like a cool lime green color. Again, very flexible. I haven't really been able to break them although I haven't really tried. I think if you really pinch them a lot, maybe you could break them. But they seem very, very durable. So great for wearables or cosplay. You know, if you wanted to have like a cool, like neon strip that's very thin and you know, doesn't use the LED neon. So that's good. And I got a couple other colors. So I showed pink and green. I have, I think this is blue. Yes, this is the blue also. And these are all about 30, so it's about a foot long, about 30 centimeters. And then lastly, I'll show off the warm white because I feel like that's kind of what, that's what people want to see. Or maybe this is yellow. This side's got the hole. And these are used for, yes, this is like a warm white light. So these are used for, you know, when they make LED light bulbs that have funky, you know, like Edison bulb shapes, but they want to fake it. They just use this and they wind it through to get that cool, like Edison lamp effect. Okay. Let's do, do we got a question about the filament and then we'll do a general question after you show these things off. No, you can't use it outside. The LED filament is not outdoor suitable. It's not outdoor suitable. You could put it in something that is weatherproof. But it's not designed. I mean, like- It itself is not. Well, it'll work for outside, but only I wouldn't recommend it for more than half an hour. The answer is no, because if we say, well, you could, then someone will put it under water. Yeah, they would do that. Can the filaments be cut to shorter lengths? Not really. I mean, I think if you were skilled and you were comfortable with it, you could cut it and then scrape off the rubber and you might be able to solder to or bond to the metal, but pretty much now. I think we'll just cut, we'll just carry lots of different. Yeah. It pretty much- But anyway, that was the naming ideas. We want to have a catchier name than just like the part number and stuff. So. Yeah. And this is a non flexible type, but the trade-off is it's much, much brighter. So this is actually what you could light up a room with it. You know, it's not super bright, but you could, you know, it's reading light, bright. And it's got two, I think it's got two layers of LEDs, like one on each side, because there's no dimmer side. Suggestion is magic wire. Magic wire. Okay. So that's the filaments. All right. And there's other colors, just orange and like, you know. Do you want to show some more stuff or do you want to answer some of these more general questions? Can you see if there, what else I said I would show off? I don't remember. Okay. Well, let's just, let's answer the questions. Okay. So I can help out with this. Yeah. Both of them. Chip shortage question. How long do you expect a Raspberry Pi shortage to last in response to what you thought was just the way it's going to be now. So. Just the way it's going to be now. Yeah. So my prediction is we're not going to see, I can order a Raspberry Pi anytime I want to just get here overnight for at least a year. That's, that's our current prediction, just based on how everything is so delayed for so many things that you just can't predict what's going to happen. And so it looks like just, you know, someone has to make a guess. So that's our guess. And then the other question is, can you talk about how the part shortage has affected your future planning beyond current shortages? I anticipate things have changed. Enough that procurement philosophy has changed and it isn't going back. Yes. So one of the things that we used to do was we used to design for manufacturing. So LaMoure sits 30 feet from the pick and place and the oven and all the things that we use to build and test things. Now we are designing for purchasing. So the difference is we look at what is possibly available, how long the lead times are, how much stuff we have. And while you did planning like that in the past, there was never like, well, if I need 10 things to make a product and one of them's gone, oh, I'll be able to get that other one part soon. That's not how it is now. So we have to redesign, we have to do lots of revisions and we have to design backwards from availability. If something says- It's not what I want to work on or what is interesting to me. It's like what I can get. Cause there's chips that I really was interested in looking at and I just can't get them. So it's like my, instead of just saying like, oh, this sucks. I can't work on it and just complaining. I'm like, well, what can I work on? And so for example, I can get expressive modules. Express has been really good about getting me chips. And so it's like, you're seeing a lot of ESP boards because I can get ESP chips. I can get ESP modules. And so I'm gonna keep doing boards with what I can get. I'm not going to come out with like, I took a break from doing trinkets because it's just very hard to secure SAMD 21s. Like even when I order them, they don't ship. So, or like the dates, you know, one thing that's definitely happened more, it's like I'll order something and the date gets bumped and bumped and bumped and bumped. And so, you know, I can sometimes get samples of parts, but then, you know, I don't, I'd have a policy for like a new product. I haven't really purchased a new component early, if that makes sense. It's like, if I only have a sample, because it's so unclear when I'm gonna get it, I wait, I actually tell the company, tell me when you can ship it to me and I'll book it. I'm not gonna future order parts that I don't already have sales predictions for because I'm just gonna get screwed. So, instead I just wait until that part's available. And like, like I do see like parts come in, like there's like this, like, there's this tide that comes in and like, oh hey, like that sensor is like suddenly available now and then like the tide will go out and like, and then the tide will come in and be like, ooh, this like chip is available. It's not everything is available at the same time, but I do see like, you know, the NAU-7802, which was a breakout we put in a couple of weeks ago. Like I designed that years ago and it was just not available, but like now you can get it everywhere. Like they really did make, you know, a couple hundred thousand and they kind of seeded the market with them, but you have two years, you couldn't get it. So it's just, it's just when the number comes up. Yeah. And you know, one of the things that we've learned is you just have to be endlessly adaptable when you're in the technology world and more so than ever, there was a lot of times where things were easy. You could do a Kickstarter and oh yeah, when the time it's funded, I'll just get all the parts and I'll do it now. You just have to have quite a bit of insight and be prepared to take some really creative risks on which direction you're going and also have lots of contingency plans. Yeah, in the last week, Phil and I were chatting four different companies dropped manufacturing lines. Yeah, they're like we're done. They're like, this thing that you actually sold quite a few of, we are not making anymore and we're done. And they're from one person maker companies to multi-billion dollar companies that just said we can't, we just can't deal with this anymore. And I'm like, oh, this product was pretty popular. And they're like, we just can't, they, it's just not, you know, I don't want to put words in their mouth because I don't know what they're thinking, but like read between the lines, it sounds like it's just not worth it to stress and you know, bust your ass and manage all the inventory and then you still can't get the last component. So they're just focusing on, you know, if it wasn't their most popular product, their most profitable product, they pretty much have. Well, the other thing is, you know, we've diversified for years because we didn't want to be stuck with one particular chipset. We didn't want to get stuck with one particular programming language for microcontrollers. And so I think sometimes folks are like, oh, why don't you just do one thing, commit to one chip line, commit to one, you know, only Arduino or commit to only SAMD. And I'm glad we didn't because we've been able to zig and zag as things have changed. And the one thing that's for sure, so we don't know when Raspberry Pis will be instantly available, but I can tell you one thing for sure, change is coming. Everything's gonna constantly be changing. If you need a project that has multiple Raspberry Pis, it's probably, you might want to think of other projects that you've been putting off because I'm sure you have a big bin of projects that you haven't finished. But everything's gonna keep changing. And that's just like how progress in life goes. Like there's scarcity and there's surplus for some things and scarcity for others. And you just have to be financially and mentally prepared and have a culture at your company that's like, okay, we're just gonna roll with it. We're gonna surf these waves, build better boards and hang out with cool surfers. Okay, well, we're running out of time. So let me, let's just show this one switch sample I got. So long answer. JP wanted us to stock these. These are like using some, you know, I think like 808 synths or something. You know, there's an LED that's separate than the button, but I just connected them together with some wires. And I got these gray with red and then, you know, red with red and then black with red. So I might carry, you know, these three colors. He wanted to make like a little, like a synth keyboard. So this is, I thought it'd be kind of cute. Okay, so let's go on to Richard. Richard. Hey, every single LED is your user powers of engineering and more to search on digikey.com. Take a digikey for the parts that you're looking for. This is a great search, Lena. What are you looking for this week? A common request. We have people who have emailed us multiple times and they said, I want to pick up a low cost oscilloscope, my first oscilloscope, and they want some suggestions. And I was like, oh, I don't really have a good suggestion. I haven't looked at what's available in the market. I thought I would use that for the great search, check out digikey, see if there's a couple of good options available. And also just talk about some of the trade-offs. Basically, if you have infinite money, you can get the best scope in the world, but if you don't have infinite money, let's say your budget's like $500, which I think is pretty fair for a good desktop scope, you do have to trade off what you're willing to spend money on and what you're willing to let go of in order to get other specs that are better, because it's all gonna be, all the specs are going to be, they're gonna affect the price. The more bandwidth, the more expensive, the more channels, the more expensive, the bigger screen, the more expensive. So what's important to you? So it's a trade-off. Again, if you have $5,000, you don't probably don't have to trade off any of these things. You get exactly what you want. But for 500 bucks, let's see what's available. So let's, let me go to, yeah, I was gonna do a great search about IMUs, but there are no IMUs, so it was like, it was a total downer. Maybe when there's more available. Okay, so we're at digikey, this is gonna be the computer. And we're gonna type in a oscilloscope, which is tough to spell, but they do have them. I think I got my first oscilloscope from Digikey as well. So there's a couple scopes. You know, there's the standards type scope and there's also, you know, these smartphone tablet scopes. Now, I'm biased. You will get a higher spec scope if you don't have a screen, but I really, really, really like the benchtop ones with the screen. I don't, there's some people who can do the USB ones. I just don't like them. It's personal. I just don't like them. Other people can do a video where they say how much they love them. I really like to have the knob. I like you able to see the waveform and I'm twisting the knob at the same time. I don't necessarily like computer style. So that said, let's just do basic. First off, I want benchtype. I want to be active. And I'm only gonna look at ones that are in stock right now because I'm not gonna suggest a scope that you can't get. So again, there's gonna be a lot of trade-offs. So the things that I find most important in a oscilloscope is number one, the number of channels because that's, you know, there's no upgrading. You know, it's like once you get the channels, that's fixed. You pay more for more channels. There's basically either two channel or four channel. Those are your options. Now I have four channel scope and I've always kind of purchased four channel scopes, but I'll say, you know, 99% of the time I don't use four channels. I use two channels, which are trigger and there's a response, especially if you're doing analog. But even with digital, because instead you'd get a logic analyzer that you would use for your digital signals. And part of the reason for that of course is digital signals are usually much, much faster than analog signals, like ironically. It's not unusual to have a microcontroller that is running the SPI line at 60 megahertz, 80 megahertz. ESP32 can do that. The Raspberry Pi can do that. Sometimes you're looking at a crystal, but usually crystals aren't that fast. Usually crystals are like, if you're looking at oscillation on your analog scope, it's gonna be 12 or 24 megahertz. So, you know, if you have a logic analyzer, you can get away with a two channel scope. If you don't, there have been times where I'm like, I want to look at the chip select line, the data clock line, and, you know, whatever some other line, it's possible. Or I want to look at the power supply voltage while I'm also looking at, you know, I squared C. So it is possible that you'll need four. I'm going to assume however that you're gonna want two, but you definitely want not one. One channel is kind of useless in my opinion. I think you really need two. So let's look at the two channel scopes. Next step, the next thing you're gonna pay for is bandwidth. Bandwidth is kind of sort of the fastest signal you can look at. It's a little bit more complicated than that because if you're looking at a square wave, you know, the square wave, you're like, oh, it's a 50 megahertz square wave, but the square wave has higher frequency components that will not come through. And so you'll kind of get it assigned you so wide instead. You know, again, if you have a logic analyzer, you could probably get away with a slower scope. However, I really think that a hundred megahertz is like kind of a minimum. I did look at this and there were scopes that had a 60 megahertz bandwidth. I think if you want to last a couple of years, get a hundred, you know, it's not just how fast is the frequency of the thing, it's how good is it going to be at reproducing rise times? And if you're doing, you know, if you're looking at some digital signals, you might need to look at, you know, skew or slew rate. It's not a bad idea. So, you know, I definitely, 20 is too small. 20 is like a handheld thing. I'd say, you know, 70 plus is good. For beginner scope, you're probably gonna get about like a hundred-ish, you know, my desktop scope here is it's five gigahertz, five giga samples per second, one gigahertz bandwidth. But it's also a very fancy scope. So, you know, what you're gonna do. But it wasn't my beginner scope. My beginner scope was a 100 megahertz bandwidth. It was a TDS, Tektronix TDS 2014, I think, which was a hundred megahertz four channel color scope. The good news is that pretty much every scope is LCD. So that's nice. They're gonna be lightweight. You're gonna be able to carry it, put it on your desk. We can't search by price, but you can sort by price. So since we have a budget of about $500, I did that. And there are a couple good options. So there's quite a few on the, you know, right under $500 range. You know, here's another thing, you know, and in addition to bandwidth, memory size. So if you're zooming into or you want to scan through a waveform, memory size is going to be something that comes up. It seemed to me while looking at these options that there's kind of like either 20 kilopoints or like a mega point and higher. I actually kind of feel like you could save a lot of money on lower kilopoint memory, but I actually haven't been really annoyed. I remember once I used a scope with very low kilopoint memory and it was a really annoying because I couldn't trigger something and then look later at the signal or there was like, it was like, you couldn't zoom in. There's something nice about you trigger on a signal and you zoom in and you're like, ooh, I can like keep zooming in and I can get the detail even if, you know, you looked at the longer signal first, you trigger on the longer signal and you zoomed in. So even though there's these cheapy, you know, like $300 scopes, I actually, I'm going to say like, get at least a mega point. I don't care about the interface as much. I've never used ethernet or USB or Wi-Fi. At the most I've used like grab a screenshot, but everything has a screenshot capability these days. That's not a big deal. And of course, phones are really good. And with an LCD, you can just take a photo with your phone. You know, it's, I know it's a little tacky, but there's no real reason. So there were a couple options here and I didn't look at, you know, like why the, you know, the why in which they're all kind of the same family, these O-ON, SDS, seven series. It looked like what you're paying for basically is bandwidth at this point because, you know, we selected a range. So 170, 70 and 100, 200. You know, if you're, if you're, if you're really up to about 500 bucks, you know, this 200 megahertz, 10 mega point one is pretty good. So like the 7202, you know, has a big screen. It's only two channel, but it is under, you know, you'll get it for 500 bucks and it's got a gig of sample per second and 200 megahertz. That'll last you quite a bit. You know, you get this in college or something and you'll be using it, you know, easily for five, six years. And I've seen people use these scopes. You're the only other thing that you don't really get to know from specs, just purchasing and why it's good if you can find someone with the same family of scope or the same model is how good it is at triggering. I've always liked the tech scope triggering. I found it's really solid. It's very reliable. I remember a long time ago, Wiggle scopes did not have very good triggering, but I think that's probably improved. That was like 10 years ago. I think probably it's a lot better. But if you do have somebody with these scopes, you'll put on a signal when you want to trigger on something and look at, is it stable or is it kind of like freaky? Can you do a single shot? And like it'll actually get the single shot. But as for actual capabilities, pretty much every scope has the same capabilities these days. You could always, you know, have triggers and you can have, you know, measurements on screen and you can do math and you can do FFT. That's super, super standard. So this would be, I think, a good two channel scope for beginner under 500 bucks. And then they also have a four channel one. It's like 650. I think if you can spare the extra 200 go for the four channel last, you know, it'll definitely make you very happy. But I think this two channel scope is a good beginner scope that will also take you pretty far. So, you know, assuming you are, you continue to be interested in electronics, you'll be able to use this for almost everything. All right, that's a great church. Okay, so that's our show for tonight. One question that came up is a TDS 2014, a good one to buy used. I wouldn't get one used. Honestly, they're really old at this point. I mean, any tech scope used is probably good. I will say that I had to stop using my TDS 2014 because the controls literally started falling apart because I used it so much that the main channel, the ordering code was really flaky. I know I could have fixed it, but I ended up was like, I need to update this anyways because the screen was quite small. It was kind of a passive matrix LCD, like it was not like IPS, it was a pretty small screen and there's a much better scope these days. Used is fine, if you're gonna get used, I would get like a tech scope or LaCroix. You're better off getting a used high quality scope. I would not get a used Weagler or O-On. Like you don't, I don't really trust the used market for those. I'd get a used high quality scope, especially since there are like authorized resellers that you can get, so you can get a good scope. Next up, and how does it work when that it's 200 megahertz, but one gigasample a second, I expected 20 millisecond samples or 40 Shannon. No, it's not the same. Does the analog limit it to 200 or something? There's the internal analog front end which has a bandwidth limitation and there's how many samples per second. It's not, you know, yet in theory, you know, you have like a Nyquist rate of two, but you can't actually reproduce the frequency. You need, you'd have a much higher sample rate and also usually the sample rate is shared between both channels. Okay, and then internally, we're gonna start to think of names for this filament. So here's some of the ones that came in. Sure. Magic wire. Magic wire. Spaghetti, lily-dee. Spaghetti. Spaghetti. Spaghetti. Light emitting spaghetti, do not eat. Glow worms. Glow worms. Filament, ha-ha. Fill element. Your name, Phil? Yeah, and then the current one that looks like there's some folks that like it is Aida's love laces. Aida's love laces, that's funny. You could make them into laces. I mean, I wouldn't use them to hold shoes together. Love laces is pretty cool. We'll probably weave them into a pair of shoes to take a photo because that'll just look cool. Yeah, we'll give them to a friend who does, she has hair design. She's like, oh, this would be a cool hair thing. So keep them coming if you want to put it into the name machine. And thank you so much, everybody. We very much appreciate it. We'll see everybody during the week. We're very thankful that y'all continue to watch our shows and you support us as a company. There's a bunch of topsy turvy times ahead, but we're in it together. And we thank you for being part of this journey with us, so we'll see you next week. Bye, everybody.