 Hi, my name is Grant Jennings from Go Win Semi Conductor and we're showing a variety of FPGA solutions. We're an FPGA manufacturing company, mainly China based and we've been around since 2014. So we sell and manufacture FPGAs with a variety of additional functionality. This your chip? Yes, that's our chip. This is our GW-1S-E2 FPGA which has a hardened Cortex-M3, 2-4K look-up tables as well as embedded puff-based security. So, physically inculnable function basically regenerates keys at power-ups so they're not saved in flash, providing an ultra-secure solution. What's the main use of your FPGA? What do people do with it? So, it's a really broad market. FPGA is a really broad market but we have a lot of opportunities in video interfacing, bridging, AI, mainly at the edge. So, consumer applications that need a little bit more performance than say a microcontroller but the cost power and size of a microcontroller kind of on a equivalent level. So, it's affordable FPGA but sometimes FPGA is very expensive, right? That's true. Our FPGAs are very low cost, very low power, very small form factor. In a millimeter by half by millimeter and a half you can have an FPGA that can do video interfacing. So, very flexible, high bandwidth, low power. Most affordable FPGAs in the world? Yes, that's our claim. And also very feature driven. What's happening here? So, this is doing a picture frame viewer. It's just a solution example showing Ethernet loopback as well as driving a display. Kind of a all-in-one solution for driving a display, driving Ethernet and driving additional interfaces for video. And this one? Oh yeah, and one more thing about this. This is running a RISC-5 core. So, a popular RISC-5 IP core and we have a variety of hard and soft processor cores that are available for free with our device. So, it's hardware RISC-5, it's not FPGA emulated RISC-5. Our RISC-5 cores currently are soft cores but on our ARM cores we have available for free an ARM Cortex-1, ARM Cortex-3 soft cores as well as ARM Cortex-M3 hard core. Cortex-M0 you mean? Yes, yes. It's an M1 which is basically an M0 but optimized a little bit more for FPGA. So, that's available on soft core but the M3 is soft or hard core. So, there's also a lot of new trends around the FPGAs, right? People can do a bunch of new stuff or people are thinking of new things to do? Yeah, that's true. We're doing a lot of new things such as AI with our FPGAs and because we have the ability, particularly in the camera area and video interfacing area, we have the interfaces to do it as well as extended block RAM and extended memory. So, we have devices that have, in addition to your 4 to 18k lookup tables, you have an extended PS RAM of 4 or 8 megabytes of additional memory. So, that's really good for doing our AI accelerator that allows you to use that memory as scratchpad memory or even doing frame buffering with your video stream. What's the main difference between Xilinx, Altura and the Go-Win? So, really, even though we're both all three of us are FPGA companies, Xilinx and Altura really focus on either high performance or high density. We really focus on power, cost and size and really our focus on applications in consumer electronic devices, as well as industrial and automotive applications. So, price sensitive applications, power sensitive applications, that's where we play. And something over there, in that corner, what's happening there? Yeah, so we have our Edge AI demo. So, this is using a Go-Win FPGA all-in-one chip solution using a CIFAR 10 data set and AI network. It's pre-trained network, so this doesn't go to the cloud or anything, it just is directly on the Edge and directly all in the device. So, if I hold up an image, say I have a horse, it should detect, there it goes, so it detects it as a horse. It can detect 10 inference outputs in hundreds of milliseconds, which is about 80x faster than what it would do in just a normal, regular processor. So, is that one of the things with FPGA, as far as I understand, it gives freedom to do different things than ARM cores or something like that, right? Yeah. What is this with the says ARM Design Start right here? So, what's your relationship with ARM? Well, we are key partners with ARM in their Design Start program, which allows us to provide Cortex-M processors with our device for free. So, actually, if you download our software tools, you can use ARM Cortex-M1 or ARM Cortex-M3 free of charge with Go-Win devices. So, that's soft. And then also our hard Cortex-M3 you can also use for free in the devices that have that available. And is there an advantage also for developers to use soft cores before they go hardware? Yeah, it just depends on what your needs are. We have a lot of customers like, for example, this HDMI to LVDS solution. So, this is driving 1080p video from HDMI port and driving the display directly, a single link or a dual link LVDS display. And this is basically a cost reduction for HMI and HMI interfaces, both in automotive, industrial, kiosk type applications. We have a lot of customers where we sell our very low cost FPGA, but then once they have that FPGA, they start integrating more features into the FPGA to consume other components from their bomb. So, they actually end up going to higher density to take other components, actually just add those into the FPGA. What are you showing here? So, this is a new device. This is our GW-1NR. So, it's the first FPGA with a built-in Bluetooth transceiver. So, it has a Bluetooth transceiver, 32-bit processor, hard processor, and 4.6K lookup tables of FPGA fabric. And so, for unique applications where you need the flexible I.O. or maybe performance of FPGA, but also need some wireless communication link, this is a great device for that. We see a lot of IOT edge applications using that. So, what's the big plans for the future? What's happening at Go-In, maybe, strategy? Well, there's many plans, but yeah, we just focus on really providing a differentiation with power cost size and features for consumer, industrial, and automotive applications that can benefit from FPGA fabric to do applications that you can't do with any other product except for spinning a custom chip.