Advantech, a leading provider of AIoT platforms and services, proudly unveils its latest AI acceleration modules: the EAI-1200 and EAI-3300, powered by Hailo-8 AI processors. These modules deliver AI performance of up to 52 TOPS while achieving more than 12 times the power efficiency of comparable AI modules and GPU cards. Designed in standard M.2 and PCIe form factors, the EAI-1200 and EAI-3300 can be seamlessly integrated with diverse x86 and Arm-based platforms, enabling quick upgrades of existing systems and boards to incorporate AI capabilities. With these AI acceleration modules, developers can run inference efficiently on the Hailo-8 NPU while handling application processing primarily on the CPU, optimizing resource allocation. The modules are paired with user-friendly software toolkits, including the Edge AI SDK for seamless integration with HailoRT, the Dataflow Compiler for converting existing models, and TAPPAS, which offers pre-trained application examples. These features accelerate the development of edge-based vision AI applications.

EAI-1200 M.2 AI Module: Accelerating Development for Vision AI Security
The EAI-1200 is an M.2 AI module powered by a single Hailo-8 VPU, delivering up to 26 TOPS of computing performance while consuming approximately 5 watts of power. An optional heatsink supports operation in temperatures ranging from -40 to 65°C, ensuring easy integration. This cost-effective module is especially designed to bundle with Advantech's systems and boards, such as the ARK-1221L, AIR-150, and AFE-R770, enhancing AI applications including baggage screening, workforce safety, and autonomous mobile robots (AMR).

Award-winning embedded solutions provider AAEON (stock code: 6579) has announced the launch of the RICO-MX8P, an NXP i.MX 8M Plus-powered fanless single-board built on the Pico-ITX Plus form factor. Utilizing the i.MX 8M Plus, the RICO-MX8P leverages a platform comprised of a quad-core Arm Cortex -A53 processor, a secondary core Arm Cortex M7, and an optional Neural Processing Unit (NPU) offering up to 2.3 TOPS of inference performance.

Equipped with an integrated Vivante GC7000 UltraLite 3D GPU, a dedicated VPU, MIPI DSI interface, and an HDMI 2.0 port, it is clear AAEON is positioning the RICO-MX8P as a candidate for multimedia applications, with digital signage, smart kiosk, and interactive Digital Out-of-Home (DOOH) advertising earmarked as potential uses. Further benefits to such use are evident from the board's support for APIs like OpenGL ES 3.1, Vulkan, and OpenCL 1.2, alongside its multiformat encoding and decoding capabilities.

Team Blue's marketing team has kicked into hype mode—Meteor Lake is launching at the tail end of 2023—with a shift in naming conventions comes another rebranding of internal technologies. The previously identified versatile processing unit (VPU) is now going under a different moniker: "Intel will usher in the age of the AI PC with the upcoming Intel Core Ultra processors, code-named Meteor Lake, featuring Intel's first integrated neural processing unit, or NPU, for power-efficient AI acceleration and local inference on the PC. Intel confirmed Core Ultra will launch Dec. 14." Michael Larabel, principal author at Phoronix, has noticed that newly updated versions of the Linux kernel have the "IVPU" accelerator driver bundled in. This will: "support Meteor Lake's VPU/NPU, while a patch is pending to extend that support for next-generation Arrow Lake processors."

In his opinion, there are some shared aspects across the two CPU generations: "Intel's Arrow Lake NPU appears very similar to that of Meteor Lake with the driver enablement coming down to adding new PCI IDs...(with) this patch adding the new Arrow Lake PCI IDs to the existing Meteor Lake driver code paths for this DRM accelerator driver is all that's needed to enable the VPU." Team Blue jumped two generations ahead at this week's Innovation event—Lunar Lake showed up in a live demo. Larabel believes that this late 2024 CPU family will revolve the dial to a greater degree: "The VPU/NPU with Lunar Lake brings more significant changes with the open-source driver changes already underway there. With Lunar Lake is the new VPU4 IP and the Linux 6.6 kernel with its updated IVPU driver already has initial support there."

Intel Innovation is Intel's yearly tech conference and the company has revealed some of what it'll share at the event that kicks off on the 19th of September. One of the sessions at the event is called Intel Client Hardware Roadmap and the Rise of AI and during that event, Intel will be sharing its latest "client hardware platforms" which according to the session blurb will include the upcoming Intel Core Ultra processors which currently goes under the codename of Meteor Lake.

It's unclear how much detail Intell will go into and based on the subject of the session, this should most likely be focused on the desktop platform, but could also cover the mobile parts. According to VideoCardz we should expect Intel to detail the integrated VPU which is said to be based on hardware from Movidius, a company Intel acquired a few years ago and that focused on making machine learning hardware. The VPU should be a low-power accelerator that handles AI inference tasks that will be part of at least some future Intel processors, but for now, we don't really know what Intel's plans are for these types of features in its CPUs, apart from offering something competitive with AMD's Xilinx derived AI Engine.

AMD today announced the AMD Alveo MA35D media accelerator featuring two 5 nm, ASIC-based video processing units (VPUs) supporting the AV1 compression standard and purpose-built to power a new era of live interactive streaming services at scale. With over 70% of the global video market being dominated by live content, a new class of low-latency, high-volume interactive streaming applications are emerging such as watch parties, live shopping, online auctions, and social streaming.

The Alveo MA35D media accelerator delivers the high channel density, with up to 32x 1080p60 streams per card, power efficiency and ultra-low-latency performance critical to reducing the skyrocketing infrastructure costs now required for scaling such compute intensive content delivery. Compared to the previous generation Alveo U30 media accelerator, the Alveo MA35D delivers up to 4x higher channel density, 4x max lower latency in 4K and 1.8x greater compression efficiency to achieve the same VMAF score—a common video quality metric.

Intel has quietly canceled its Thunder Bay hybrid system-on-chip (SoC) that combines standard general-purpose CPU Cores and Movidius Vision Processing Unit (VPU) cores. Such chips were aimed at commercial and Internet-of-Things (IoT) applications that relied on computer vision acceleration and edge-computing applications. According to the latest report, it appears that Thunder Bay is canceled as Intel has released a set of patches that removed the Thunder Bay code from the Linux kernel.

Intel kept Thunder bay details well hidden and while earlier rumors pointed to a combination of Intel Xeon x86 CPU cores and Movidius VPU cores, the only Thunder Bay support in Linux patches where showing a combination of ARM Cortex-A53 low-power cores with the Movidius VPU. Intel acquired Movidius back in 2016 and while it is not something that Intel talks about often, there are several products based on Movidius VPUs, including Neural Compute Stick, Intel drones, the Intel RealSense Tracking Camera, and most recently, the Intel Movidius 3700VC VPU, formerly Keem Bay. In any case, it appears that Intel is abandoning an idea of combining general purpose x86 CPU cores and Movidius VPU cores.

Hailo, the pioneering chipmaker of edge artificial intelligence (AI) processors, today announced its groundbreaking new Hailo-15 family of high-performance vision processors, designed for integration directly into intelligent cameras to deliver unprecedented video processing and analytics at the edge. With the launch of Hailo-15, the company is redefining the smart camera category by setting a new standard in computer vision and deep learning video processing, capable of delivering unprecedented AI performance in a wide range of applications for different industries.

With Hailo-15, smart city operators can more quickly detect and respond to incidents; manufacturers can increase productivity and machine uptime; retailers can protect supply chains and improve customer satisfaction; and transportation authorities can recognize everything from lost children, to accidents, to misplaced luggage. "Hailo-15 represents a significant step forward in making AI at the edge more scalable and affordable," stated Orr Danon, CEO of Hailo. "With this launch, we are leveraging our leadership in edge solutions, which are already deployed by hundreds of customers worldwide; the maturity of our AI technology; and our comprehensive software suite, to enable high performance AI in a camera form-factor."

Intel yesterday confirmed its plans to extend its Meteor Lake architecture towards shores other than general processing. According to Phoronix, Intel posted a new driver that lays the foundations for VPU (Versatile Processing Unit) support under Linux. The idea here is that Intel will integrate this VPU within its 14th Gen Meteor Lake architecture, adding AI inferencing acceleration capabilities to its silicon. A sure-fire way to achieve enormous gains in AI processing, especially in performance/watt. Interestingly, Intel is somewhat following Apple's footsteps here, as the company already includes AI-dedicated processing cores in its desktop/laptop Apple Silicon processors since the M1 days.

Intel's VPU architecture will surely be derived from Movidius' designs, which Intel acquired back in 2016 for a cool $400 million. It's unclear which parts of Movidius/Intel IP will be included in the VPU units to be paired with Meteor Lake: whether a full-blown, SoC (System on Chip)-like VPU design such as the Myriad X VPU, or if Intel will take select bits of the architecture (plus the equivalent of five additional years of research and development), sprinkling them on top of their upcoming architecture. We do know the VPU itself will include a memory management unit, a RISC-based microcontroller, a Neural Compute System (what exactly entails this compute system and its slices is the mysterious part) and network-on-chip capabilities.