Speaker 1: Welcome to everyone joining us in Taipei and around the world today as we host Compex 2024 at AMD. We love games. Today, I’m excited to show you what’s next for Ryzen PC gaming. Our new Ryzen 9,000 CPU is the world’s fastest consumer PC processor, bringing our new Zen 5 cores to the AM 5 platform. Supports the latest IO and memory technologies, including PCIE 5 [00:00:30] and DDR 5, I’m now excited to show you that our new Z 5 Core Zen 5 is actually the next big step forward in high-performance CPUs. When you look at the technology behind this, we have a lot of new technology. We have a new parallel dual-pipeline front end that is designed to improve branch prediction accuracy and reduce latency. It also allows us to deliver higher performance per clock cycle. We also designed Zen [00:01:00] Five have wider CPU engines and instruction windows that can run more instructions in parallel, resulting in lead computing, throughput, and efficiency. So compared to Zen 4, we get double the instruction bandwidth, double the data bandwidth between floating point units, and double the AI performance with full AV X 5 12 throughput. All of this is concentrated in the RISE 9,000 series of products, and our IPC increases by an average of 16% [00:01:30] Compared to Zen 4, it covers a wide range of application benchmarks and games. Speaker 1: Now let me show you the top of the line goes up 9 99, 50 times for the first time. So be it. Speaker 1: We have 16 Zen Penta-cores, 32 threads, up to 5.67 GHz boost, [00:02:00] 80 MB large cache with a TDP of 170 watts. This is the world’s fastest consumer-grade CPU. We know all of our fans love games. 99 50 x delivers best-in-class gaming performance across a variety of popular games. Now for desktops, we know enthusiasts want to have an infrastructure that allows you to upgrade across multiple product generations. We partner with Verizon. we just did [00:02:30] That. Our original Ryzen platform, Socket AM 4, launched in 2016 and is now in its ninth year, and we have 145 CPUs and APUs across 11 different product families in Socket AM 4. We’re actually still launching new products. We actually have some Ryzen 5,000 CPUs coming out next month, and we’ve taken the exact same strategy with Socket AM 5, which we now plan to support in 2027 and beyond. [00:03:00] So you’ll be seeing our AM 5 processors for many years to come. Now, in addition to the 99 50x improvement over the top version, we’re also announcing 12-core, 8-core, and 6-core versions that bring Zen 5’s leading performance to mainstream price points, all of which will be available in July. Speaker 1: We are very happy to announce the launch of the third generation artificial intelligence processor today. Our news on artificial intelligence keeps growing [00:03:30] series actually significantly improves computing and artificial intelligence performance and sets the standard for what a co-pilot PC can do. Thank you, Drew. here we go. This is STR. strix is our next generation processor for ultra-thin and high-end notebooks, combining our new N5 CPU with faster RDNA 3.5 [00:04:00] graphics card and the new XD NNA two NPUs. Thanks. Speaker 1: When you look at the product that we have, it’s really all the best technology on one chip. We have a new NPU that delivers an industry-leading 50 top results. Today we’re going to talk a lot about tops. 50 top-level computing technologies can support new AI experiences with extremely low power consumption. We have the new N Penta Core [00:04:30] It delivers all the computing performance of an ultra-thin laptop, and we have a faster RDNA 3.5 graphics card that really brings best-in-class application acceleration and console-level gaming to the laptop. Now we have some deviations. The flagship upgrade to 9, HX 370 has 12 cores, 5 cores, 24 lines, and 36 trillion cash. The industry’s most integrated NPU and our latest RDNA graphics card. strix is simply the best [00:05:00] Mobile CPU. So let me talk about what’s special about this new NPUN PS, they’re really new, they’re really suitable for all these AI applications and workloads. Speaker 1: Compared with our previous generation XD A 2, which now has a large number of 32 AI blocks, the multi-tasking performance has been doubled. It’s also an extremely efficient architecture, running at twice the energy efficiency of the previous generation [00:05:30] A generation of artificial intelligence workloads. In addition to Microsoft, we work with all the leading software developers, including Adobe, epic Games, SolidWorks, Sony, zoom and more, to accelerate the adoption of AI-enabled PC applications. By the end of 2024, we expect there will be more than 150 ISVs developing for the MD AI platform across content creation, consumer gaming, and productivity applications. give now [00:06:00] Let’s take a look at some of the upcoming CoPilot+ PCs and let me welcome our next guest, a very close partner and good friend Enrique Lores, HP President and Chief Executive Officer. Hello. Hello. Speaker 2: We’re really excited about the new product line that’s coming out in a few weeks because Speaker 1: I think you have something to show us. Is it right? OK I hope so. Speaker 2: This is actually new [00:06:30] Now that we’ve done a generation together, we can show it together. That’s great. When we integrate the latest AI 300 series, which is the next generation omnidirectional book, as Lisa said before, it will be the first to integrate 50 tops into the device. As Parveen said, performance is crucial because [00:07:00] Allowing us to continue to provide incredible experiences to our customers. Speaker 1: Now I would like to invite Luca Rossi, President of Lenovo Smart Device Group. Now, Luca, you’ve been holding something, and I thought you would show us what it is. Is it right? Speaker 3: Well, I shouldn’t even show this because it’s something we won’t announce until later this year, but Lisa, you’re right. I think it’s too exciting [00:07:30] The way you look Speaker 1: I’m actually very happy. Speaker 3: Yes, very happy. Straight from our R&D, straight from our R&D labs. This is our first early look at the new Yoga laptop powered by third generation emerging AI technology. That’s Speaker 1: Beautiful. Speaker 3: Maybe we want to do left and right. yes so i [00:08:00] Can share more today. I can can. That’s very beautiful. Thanks. I can tell you that this device represents a major leap forward in the next generation of artificial intelligence computing and will include some of the Lenovo-proprietary IP AI capabilities I mentioned, and that’s just the beginning. Speaker 1: Next, I would like to welcome one of the most important visionaries and innovators in the Taiwan ecosystem and a very, very close partner, Jesus Chairman Charmaine Sheh. Speaker 4: [00:08:30] I think today at 4:00 PM, we’re going to launch a series of cutting-edge IPCs in our portfolio with the new Zen Book pro art. We pre-ordered a Thing and IG laptop powered by a 3rd generation AMD Ascend AI processor. These new land-based laboratories are equipped with the world’s most powerful MPUs, with 50 top-notch [00:09:00] and the superior MB xFi architecture that leads the industry in computing and artificial intelligence performance. The rise of the third generation of AI processors is the catalyst that will bring personalized computing to everyone, from content creators to gamers and business professionals, empowering them like never before. This advancement gives the new notebook higher AI performance than the MacBook [00:09:30] while making it thinner and lighter. In fact, we are very proud and honored to be the first OEM partner to offer consumers the third generation of emerging artificial intelligence systems. Will be ready to buy in July. Isn’t this incredible? Thanks. Speaker 1: So I have something to show you today. It’s actually a preview of our upcoming fifth-generation Epic processors, codenamed Turn. Speaker 1: [00:10:00] So take a look at the turn on your first turn. It has 192 cores and 384 threads, and has 13 different chips built using three-nanometer and six-nanometer process technologies. Turin has a lot of technology. It supports all the latest memory and IO standards and is a drop-in replacement for our existing [00:10:30] Turn, the fourth generation of the Epic platform, will expand Epic’s leadership in general-purpose and high-performance computing workloads. So let’s take a look at some of those performances. namd is a computationally intensive scientific software that can simulate complex molecular systems and structures. When simulating the model 20 million times, Turin’s 128-core version was more than three times faster than the competition’s best version, allowing researchers to complete models faster [00:11:00] This could lead to breakthroughs in drug research, materials science and other fields. Turin now excels at AI inference performance when running smaller, larger language models. Speaker 1: So I want to show you a demo here. Now. This demonstration compares Turin’s performance when executing a typical enterprise deployment of two LAMA virtual assistants, ensuring a high-quality user experience with minimal guaranteed latency. Both servers first load two instances of LAMA [00:11:30] Each assistant was asked to sum up the uploaded document immediately, and you can see that the turn server on the right added double the number of sessions in the same amount of time while responding to user requests significantly faster than the competition. You’ll see other servers stop quickly when they reach their maximum number of sessions. It basically can no longer support latency requirements. Turn continues to scale and deliver sustained throughput [00:12:00] The number of tokens per second has increased nearly four times. Our customers are very excited about the turn, and I know many of our partners are actually in the audience today. I would say we expect to launch in the second half of this year. Thank you for such enthusiastic visitors and wish Compex 2024 a complete success. Speaker 5: Thank you.
