 We are the Mobile World Congress with ThinkSilicon. So how is it? Who are you? I'm George Ziroblos, co-founder and CEO of ThinkSilicon. And who are you? I'm Uli Müller. I'm vice president of marketing. And ThinkSilicon, you say here it's the world's most power efficient GPU. Is that true? Of course. So you're making GPU? Yes, we are doing graphics processors for extremely low power applications, like smart watches or smart glasses. And we are beating the competition in terms of power, cost and performance. How do you do this? We have secret source in different domains. One of the very important part is the compression of the data. So we compress everything, we shrink everything. And the world source shrink the hardware and the software. So here you have some demos. Can you show what you have here? So yes, sure. What we have here from right to left. Right on the right side you see here our 2.5D GPU on the FPGA board. And you can see here is a classical smartwatch face, what you can see here. What we can try, what else we can do with certain other applications we can run here. And here's a rotating clock and here you see the 3D effect. So what's really interesting here is that we don't need a lot of power for that. So we basically built this architecture towards, like George already said, to small display devices with 1.5 to 6 inch. Here you have an instrument cluster even for automotive. You can drive an automotive display with that. So low power, how low power? And how much do you measure? Do you show the difference? Yes. So how low can it go? For this kind of applications we can go down to 1mB, the big power. 1mB what? In 1mB. That sounds quite low, no? It is impressive, yes. This is one of the most important things to have long battery life. That's quite hard. You want to have something that looks good but lasts for months. Yes, actually what we're trying to achieve is to have a, usually in the face like your smartphone, but have it on your smartwatch, while the battery life is one or two weeks. While now either you have a poor experience on your smartwatch, either the battery life is one day or even less. And you have some other demo here, which one is that? So here, this is a very interesting project here. This is actually an FPGA from Lattice, and we programmed this FPGA with our display graphic processor here. And this is even smaller than what we've seen before. It's an ultra-small FPGA with just 5,000 LUTs. And what we demonstrate here is that we can drive a display with so little resources and it's battery powered. You see, it's not attached to any, or not connected to any device or to any outlet. And this is usually used on, always on displays, what you have on your mobile phone, would you have your data or your date or your time. And here we're demoing certain kind of apps and applications, what you can do with that. Here the battery, the power consumption is even lower. It's even below one milliwatt. Below one milliwatt. So is the idea that potentially you can have an always on OLED display running some nice animation even all the time? Even the rest system is totally powered off. So you can extend the battery life of the standby. Either you can have very nice and fluid graphics, while the system is on, and keeping the consumption extremely low. And do you have some customers doing this already or is it potential in the future? We had last week a present list with Seconds Communications, a very big company in the LTE domain space. So they integrated an LTE modem, a system on chip based on a non-processor and our GPU. We have the latest project now. One of our big customers is Microchips and Dialog. All right. Since when did you do this? We did it for the last nine years. Nine? Yeah. We built all the hardware, the software, the compilers, the processor, everything. Did you say how many it was shipped in total in nine years? How many? Of your end chips that might have yours? Are you not allowed to say? Yes, we have some problems to talk about. Okay, no need to say. All right, but maybe millions? Yeah, yeah, yeah. Millions of devices out there are easily available. And the last part is that beyond the GPU and the hardware, we also give our customers the best case to make our GPUs very easy to use and allow them to go very fast to the market, decreasing the time of the application development. So here we have an example. This is a graphical interface. We don't have a... We don't have a mouse pad. But... The usual case, the demo is not... Okay, yeah, no problem. But then, how does that work? I think go for the laptop. Okay, no problem. It's disconnected. So what is special about it? There's a UI. It's a UI which is very close to your mobile phone. There are some 3D effects. There are some projections, so it's really high quality. And the differentiator is that you can build this UI in a couple of hours or days instead of weeks or months without using our tools. But what is the kind of software that runs this? It's our software. So we build this... We call it WeBuilder Toolkit and translates all the software to calls to our GPU. So it gets the benefit of the acceleration from our GPU in order to run this user interface. So it's about compression? It's about optimal code optimizations? It's how to transform and translate a graphical user interface to calls to our GPU in order to get the most benefit out of the GPU. Including everything. Hardware, software compression, whatever we offer. And here at the Mobile Oil Congress, you're looking for all kinds of potential customers, right? We have potential customers, potential investors, press coverage, so as a start-up we need some exposure. These are the places to be. What's the target? How big are you going to be? Well, seeing we want to be the next imagination would probably be a very ambitious target, but we want to be somewhere in between, maybe between Vivanti and Imagination, which is one of our competitors. So you're going to have more products, more different GPUs? Not only we have a roadmap about the GPUs and the next release is about the AP. A very important part is that using our technology which is validated, we have a number of patents in several fields and our major advantage is the low power. The question is how we can apply this technology to new kinds of applications like virtual reality or augmented reality. And this will be a challenge for us. Can you run Android? Yes, we do. You do. But then that's one of your bigger GPUs. You have some small and the bigger ones? It's the middle, the high end. All right. Cool. All right, so thanks a lot. And there it is. People can read more here. That's the stuff here.