 So we're here with Microchip and there's a couple new chips here to Microchip, so how much do we do? There is. I'm Jacob Lassen. I'm working with Microchip, part of our product marketing team. So we have two new products coming out this week. So the first one is the PIC 16F18446 and this new device is having a very nice ADC. This is the first small PIC with the 12-bit ADC with computation as well. 12-bit ADC, what does that mean? 12-bit ADC. So the resolution here means that you can actually do more accurate measurements. So the intention here is that we want to be giving customers a better solution for censoring applications and measuring analogue stuff. So in addition to the 12-bit we have, we also have oversampling in hardware. And that means that you can either suppress noise better or you can actually gain better resolution by signal processing, digital signal processing. So you have 12-bit analogue performance plus additional 3-bits of digital performance. So you're actually up to 14, 15-bits in real-world applications. So the demo we have here, that is with the new device that we have, the 18446. Here we have set up the board. We have a small standard development board that we have soldered on to the base board. And we're then showing different types of sensors, sensor applications when we're using these new analogue capabilities. So the demo that is running right now is doing proximity sensing. So here you see the circle as it goes larger, it means that I'm closer. It has a capacitive touch sensor along the edge here. So that is used to actually detect whenever I'm approaching the board. We then also have other stuff here, other cool demos. We have a pressure sensor, so the small sensor here is detecting pressure. So if I'm just tapping it, you can see that it starts to produce a little bit of ripple here. We have one more. We have the temperature sensor. Let's see if I have warm fingers today. So if I hold this one, it should go up, I'll have cold fingers today. It's cold. It's my fingers are colder than the other one. It's crazy cold in here right now. Right. So the other thing there, we have battery that's not to demonstrate. So the last thing here is the metal detector. So here you have a coil, which is capable of actually detecting if I'm approaching. So I'm using my watch here. And you can see first the gold is starting to react. And then the beeping frequency is increasing. So here we're doing all the metal detections. So all of these things are things you can do with this new analogue capabilities. One chip. One chip. This is the one running it. So what's it going to be used for? Here it says something about the washing machines or what's it going to be used for? I mean anything where you need sensors. So if you have a washing application like this one, you can use it for many of the things where you need to detect water or temperature. You need to detect maybe some of the particles in the water if the washing cycle is automated and so on. So a lot of the things you need are quite good ADCs. So is this a very important new chip? I mean it's nothing like that before? No, because here we have the 12 bit ADC and a very small package. So that is a new part that we think are relatively far on customers. Is the microcontroller with nothing else or something like that? Oh no, no, no. There's a lot of peripherals to it. So you have all the standard communications peripherals. You have a lot of very cool timers that you can use either for detecting like I showed you with the fidget spinner where we have a rotational detection timer. Yeah, I can do that. So here we have a cool speed timer. This is the microchip fidget spinner. Is this the coolest fidget spinner in the world? That's the coolest fidget spinner in the world. So what it does is it's actually using one of our special timers to do a rotational speed detection and synchronizing with a small sensor we have on the board here. So if I turn it on, what I can do is that I can spin it and it will show a microchip text here. Nice. And with eyes, I can definitely see microchip right here. My camera has not the right shutter speed, but it's okay. But it says microchip, like super cool there. Yeah. It can be programmed with to say something else? Yes, it can. Because here we have added our Bluetooth module as well. So it means that you can actually put it into programming mode by holding these two buttons here and you can then load something cool for your wife, for your girlfriend or something that you want to show. You can load it with an app. So basically it senses the speed. It senses the speed. Certainly just you can barely see it here, but you have a small map in here on the backside and you have a whole sensor on the front here. So that is detecting the rotation of the board so that it can actually synchronize to that. Nice. How long is it going to work on the battery? Oh, a very long time. Yeah. The faster you spin it, the longer it lasts. The three batteries because it needs to be balanced. Exactly. So actually there's too much power. There's too much power here. So the main reason for having that, first of all you want to have some weight through it so that it continues spinning. The second thing is of course you want it to be balanced. So for that reason you have large batteries in three areas here. Like 12 year battery life or something. This will last a while. But that's not exactly the same chips that you were just talking about? No, this is another one. So this was just a smaller one that we used just for this demo because that was what we had used when we made it a couple of months ago. We used a different chip. Right. But you know the pick chips are very similar so the peripherals that you would see in this one is things that you would see in the other one. So you're talking about the pick here. What is this Atmega? So as you may know, Atmega was acquired by Microchip two years ago. And with that acquisition all the Atmega products and AVR products that came in. So the AVRs are 8 bit microcontrollers as well. So Atmega is the AVR? Yes, that's an AVR. So here we have the new device that came in the Mega AVR series. This is the Mega 4809 and it actually comes with more flavors to it. We have if you just switch to this one. So we have a number of different variants of this one. We have the 4809, the 4808. So these are respectively a 32 pin and 32 pin and you also have the same in 32k versions. So 48 pin and a 32 pin version of that. So these new cool devices they have a new improved ADC. So the demo we have here are just turned off. Let's just get this oven running. I don't know why it's turning off now. Give me a moment. Something happens here. So some applications are better to use the pick, some are better with the AVR, some are better with the arm. How do people choose? That's a good question. So what I want to show you here is one of the reasons why we recommend 8 bit for some solutions and 32 bit for other solutions and just get this oven running again. There we are, now we're running. So as you know if you go to arm devices you typically have a high operating frequency. So that is something which is really useful if you want to do massive processing. On the other hand an 8 bit is typically with lower power consumption and what we are doing here with the 8 bit is we're focusing on real time performance. So while for the machines they often use the CPU to do a lot of real time performance that can be a limitation as well. So what we're instead doing for the pick and the AVRs is that we put that real time functionality into the peripherals. So the arm suggests that it's the Cortex-R for real time? Yeah, so but still, so we have the arm products next door here. So what we say that you can do a very good call but if you do not have excellent peripherals you can actually not meet your real time requirements. So the demo here is focusing on the analog performance. So here we have the Mega4809 on an Explained Pro board. This is a standard development board that you can get right now. In addition we have made a signal generator and the signal generator is currently generating this signal here so this is a raw signal on top and on the bottom you have the measured signal. So that is transmitted from the board up to this tablet PC. So what we can show is that all right you have this is an ideal situation. This is what you would see in the lab but in reality you typically have noise. So let's turn on random noise. So what you see here is just get this back to where I wanted it to be. Here you see the standard operation of an ADC. The noise that is on the signal is actually measured as well. But the ADC here has some fantastic hardware capabilities. You can do hardware averaging meaning that you can actually do filtering in the ADC itself. You don't have to spend CPU cycles to do that. So if we start turning on this, if we're starting to add the oversampling you can see that the noise, the random noise is attenuated and if we go to the maximum oversampling you can see from that source we can measure that signal by doing that in hardware. Real-time averaging. Real-time averaging. So the CPU is not doing any of this is only in the ADC. The other thing we could look at is, all right, here we have some averaging which can remove random noise but if we have periodic noise that's a different situation. So we still have the averaging turned on but here we have a higher frequency component overlaying our signal which is not possible to remove with the averaging. What we can do instead is that we can add jitter to the ADC we can uncorrelate the noise source and here we turn it on, we do add the sampling delay and you can see we actually are able to uncorrelate the noise. So just to make it worse we're adding the random noise again you can see this is representing a sine wave and a lot of noise and this is what we can actually measure with the ADC without spending a single CPU cycle. And the ADC is, I mean what are the use cases for this? Where does it go? So again what we're focusing on is censoring applications and typically what we see here is that we need it for industrial applications and where you have harsh environments, a lot of noise. There these fantastic features would be fitting well in because you can say you can have noise from a 50 hertz you can have noise from a lot of different things but we are able to remove that. It could be, well we wouldn't use this specifically for radio but it would be for industrial applications where you want to have good analog measurements even though that you are in an electrical noise environment. So this processing that you would normally have to do in software this would require a quite powerful CPU to do the same processing this is something we can do without spending a CPU cycle. So that is coming back to your question about well when do we recommend ARM, when do we recommend 8 bit products? So if you have an application where you need to do good noise suppression and still want to have low power consumption this is typically an 8 bit application where you want to use these kind of features. So these two new chips that you just presented there they are going to enable new products in society right? They are, they are. Like this is going to mean what? New machines, new... So what we see, many of the things that we see today is that a lot of systems are being connected we are talking IoT, connected devices and in all of these situations typically what you do is you are measuring environmental parameters it can be temperature, air pollution, a lot of things and you are collecting data about that and transferring that to a centralized system. So this is part of a larger connectivity story where you actually need good unlike performance to do connectivity and share that information in IoT applications.