 So we're here at the Sharp booth and this is Memory LCD. So hello, so who are you? I'm Jeremy Locke. I've been working on Memory LCD for quite a number of years within Sharp. And so this is some of our technology. The idea very much is that it's a reflective display. And so it's for very good outdoor readability, extremely low power. So we're talking about microwatts, very few displays that are out there at the moment with the capability of performing a microwatts type of values. So this is our memory in Pixel. We have other examples here. If you want to see, here's a demo, a simple demo of a moving picture. So you can get an idea of the capability of the technology. Each Pixel stores data. And so we're able to hold that data in a still image and just have that extremely low power functionality. I don't know whether you can see that. So are you basically as responsive as any LCD? Yeah, this is typically in the same way. It's just the same materials that we use with Memory LCD as you would with a standard LCD. But it's just one bit, so it's a very simple display, very easy to integrate. And here you can see this is for energy harvesting. So we've got a very small solar cell, which also we've been doing some research on. So this solar cell is quite a powerful device and is for indoor energy harvesting. So everything is being driven just by that solar cell. So right here, this could be a smartwatch and this could be on the wristband or somewhere. Yeah, absolutely. And then everything can be driven purely by that solar cell. Maybe a super cap or something like that, or some form of storage might be useful for nighttime use as well. So it would run the display and could be an arm processor and some Bluetooth and some other stuff. Yeah, well at the moment this is already running, if you look on the back, this is already running quite a little bit of electronics anyway. We've got an MPU on the back there, there's also Wi-Fi capability and other aspects within this device. And so we're able to run that type of thing with our solar cell. So you're already in this very, very popular brand right here. This is in all the pebbles. Yep, that's right. So this is one of the products. We're producing this product typically for the sports type application. And you can see this pebble watch here. And then we've got several other industrial applications that we're using the device in. And a lot of interest at the moment really in terms of the device. Right, and so you're showing some other displays right here? Yeah, so we've got quite a range here. We're going from about 0.96 inch through to, this is a 2.7 inch. You can see there that's 400 by 240, so it's quite high resolution. What kind of, could this be shelf labels or what kind of... It could be shelf labels. It could be information display. It could be in sort of the home sort of situation where it's just information displays that we're using within home energy management systems and that type of thing. All right, and this color? Yep, that's right. So this is a simple color device. So again we're trying to keep the power and the simplicity there. So this is just basically an eight color device, but you can see that even without a backlight. So this is, we're still talking about maybe 25 to 30 microwatts of power that this is requiring. So it's very, very low, but you can see the clarity of the image even in the light performance in this room. So memory LCD could be perfect for smartwatch, right? For smartwatch, and for other applications actually. There's no reason why with, it's just smartwatch. So we're talking about sports applications. There can be cycle information displays, automotive. There's lots of areas really that this can be used. And one of the key things as the number of displays increases within the industry, then this is extremely low power. So you're not using much battery life. And typically we're looking at trying to have a product that maybe will last for one year on a typical button cell type battery. So I was at the China Sourcing Fair. I saw some awesome memory LCD smartwatch, 20 day battery life. So this is a whole different class than Apple Watch and some other stuff. That's right. There's no comparison. It's not competing in that type of market. This is where somebody wants significant lifetime in terms of their battery performance. You know, typically if you have no access to power, then it's no good to have a smartwatch. You know, if for instance you go away for a week or a weekend, then you're in a position where you can have a device like this and you've not got to worry. It can be there, it can be as an emergency backup or anything in terms of the information that it gives you. So it could be for IoT and displays everywhere. That's right. Everywhere. In terms of, at the moment, this is a fixed glass substrate. But ultimately there's no reason why you can't start to move into more wearable or flexible types of product. And it supports a backlight? It does support a backlight. It's not a frontlight. Typically we try and avoid that, but on this you can have a backlight. It's got some transmissive components, but it could also have a frontlight as well. The backlight would just be optional, right? It's just optional. And also you can use it for information again. So you could use different colour backlights so that you're able to maybe indicate a warning or to say everything is okay by green or red LEDs that you use. And so by doing that then you can keep the power right down as well. So it's just a display and could it maybe run software like Android Wear? Well, there's no reason. It's very, very simple interface. So this is just one serial pin data. So this is just a Raspberry Pi. Just knocked up really fairly, fairly quickly. So you can see here we're just using a standard Raspberry Pi here and using the GPIO pins. So basically it can be connected to anything fairly, fairly simply. So if anybody wants to make Android Wear with like 10, 15, 20 days battery life, they should use a memory LCD? Absolutely. And you'll find on the internet already. There's lots of guys who have been looking at memory LCD and they love it because it's so easy to interface with some of these little systems. It gives them the information they want. If they're doing some censoring then you want information from your sensor. You want to be able to know what actually you're measuring and what's going on. And so the information display that we have, the memory and pixel can give you that. So when you do solar cells, you do a PV, are you optimizing for wearable or are those extremely special solar cells? Well, these are a new type of solar cells. So this is giving us a factor of improvement of something around about six compared to your typical PV. Six times better? Six times better. And the purpose really here is for indoor use. So it's energy harvesting indoors. So even in low light conditions, you can see even in the low light conditions it's able to drive this memory and pixel display along with the electronics associated with it. So it's quite significant really in terms of the power output of such a device. So you say that it's been designed for over the last bunch of years. When did it start? Well, memory and pixel probably started in something like 2008, I think it was. This type of technology is just an extension of your standard LCD. So Sharp have been doing sort of transfective displays for many, many years. And so this is just pushing the transfective to a simple reflective display and then having that memory store in each pixel. So it's not transfective? It's not transfective in the sense that we normally talk about transfective. It's a reflective display that has a small transmissive part so you can have a backlight. But it's predominantly reflective so that you keep the power right down. You're not actually introducing any significant hit from a backlight. What is the power that always runs? What is it used for? When it was a little bit power, right? There's a tiny little bit of power. Typically liquid crystal requires something just to keep it operational. And so you're actually running this so that the pixel gives you a clear image. You could actually drop that down but then you start to lose the image quality. So it's much better to keep it running. And how low power are you talking about? Let's say it's normal smartwatch battery runs how long with this standby mode? Well, I think typically you're talking about in excess of a year. A year is possible? Of course, you've got to remember it depends on what other electronics is using that battery. So from a display point of view, when you're talking about maybe 15 to 20 microwatts, it's extremely low power and you don't have to consider that. And if you're only updating once a second, the power will go up maybe to 35 to 40 microwatts. So it starts to increase when you refresh, but it's still extremely low power. So it's two or three times a little bit more power just to change the pixels? Yeah, because you've got to drive the system to actually give you a new image. One significant advantage as well is that you don't ask when you refresh the whole display. You don't have to refresh the whole display. You can refresh parts of the display. So if you just refresh a part, then the power comes down? Only there that it uses power. So you were saying yesterday at the keynote that every pixel has a memory. How does it work? That's right. So it's just a simple SRAM block within each pixel. So there's a few transistors. So we're using a low temperature polysilicon type technology here. And we call it CGS, which is our continuous brain silicon. So the performance there enables us to have transistors that actually are very good switches. And we can build logic gates and other systems. So within this memory LCD, you'll find that all the driver circuits are actually on the glass. So there's nothing actually that's required outside of the glass. And so in each pixel we have several transistors that are actually storing the data. And so then you only need to refresh or change that when you want to update your image. So each pixel has their own kind of electronics driving them. And it's different from how LCDs and normal LCDs are. A normal LCD just has one transistor which acts as a switch. And so you drive that and that switching then and driving the LC material. Here you're actually storing that data and then you're able to drive the LC material. And how's the price? Is it going to be potentially hundreds of millions of devices because it's very affordable? Well, one thing is there's a lot of technology in this. So a low temperature polysilicon is a little bit more expensive than your typical amorphous silicon. So there is a price hit. But as you see with a standard watch like this, then obviously we're having customers that the price is not such an issue. So you get an idea. Like for display that size is just $2 more or something compared to a LCD that would have only one day battery life. You're just talking about a few dollars. Because all this is already mass production. Everything is ready for mass production. It's just any hardware maker can just order as many as they want right now. All the displays that you have. Yeah, so typically the displays on show here are ones that are already in mass production. And so they're available either from our distribution channels or depending on the type of quantities direct from shop. So when I go to China Sourcing Fair, I see several memory LCD devices. So is it easy for them to just implement, take the LCD and put it on their PCBs and it just works? Yeah, pretty much. It's very simple structure because it's standard structure. It's very simple in terms of its interface because of the data requirement to drive it. And it's just all in all, it's very simple for anybody to get a hold of and to actually get it working. Could you describe a little bit how the viewing angles and maybe the contrast, how does it work in memory LCD? Well at the moment this is, again, it's just typical LCD type structure. So depending you can have potentially different technologies that are driving the actual device. Something like IPS and TN? Is it something like that? So typically this is TN at the moment, but there is no reason why you couldn't have other types in LCD. TN is satisfactory for what we actually need, so there's not an issue there really. But at some point in the future I'm sure this will go to maybe a wider view type VA application. And Sharp is doing an IGZO and stuff like that. Is that compatible with this or is that something else? Well IGZO is the backplane, so at the moment we're using CGS. IGZO is relatively new and our focus for IGZO is in other areas at the moment, particularly in the sort of tablet type industry. Whereas with this potentially IGZO is a backplane that we could use in the future, yes. How about more colors or more saturation in colors or something? Is that possibility or how's the consideration for this demo right now? Yeah well at the moment what we're doing, we're just trying to keep it very very simple. So this is an eight color device. So eight colors? Yeah it's eight colors. Why is this eight colors? So basically you've got your RGB and so from that simple structure you can get your red, green, blue and also your cyan, yellow and magenta and then black and white. So there's your eight. And so it's because we're just working with a single bit switch at the moment. It will add complexity into the pixel structure once you go for more colors. Alright. Is there a difference in the visibility when it's color and when it's just black and white? Well typically color is, for instance the white is being built up from the RGB color filter. So you're going to get a performance difference in terms of what white looks like with the color version. But one thing the color version gives you is actually better contrast. So there maybe is a trade-off in different areas in terms of performance.