: Intel and Altera Debut Chips and FPGAs for AI Enhancement

Highlights:

• The new Intel Core processors for edge tasks combine the GPU power of Intel’s 13th Gen Intel Core mobile chips with “LGA socket flexibility,” enhancing system scalability and deployment speeds.
• Intel introduced two new Intel Atom processors: the x70000C Series and the x7000RE Series.

Intel Corp., in collaboration with its subsidiary Altera, recently launched a range of fresh processors and field-programmable gate arrays (FPGAs). Intel and Altera debuted chips and FPGAs to expand artificial intelligence capabilities to the network edge.

The latest releases, reportedly widely accessible at present, encompass a variety of edge-optimized central processing units and graphics processing units produced by Intel, alongside the Agilex 5 Series FPGAs developed by Altera. Intel has expressed that its latest silicon is geared towards enabling AI capabilities such as visual computing, media processing, and business intelligence at the edge. In contrast, Altera’s FPGAs are tailored for a wide array of applications, spanning AI-enhanced video, robotics, industrial and medical applications.

Intel claims that each of its new chips is equipped with integrated AI acceleration capabilities, poised to drive a next-generation wave of smarter devices at the network edge.

At the pinnacle of the lineup are the new Intel Core Ultra processors, delivering up to five times superior performance in image classification inference tasks compared to the 14th Gen Intel Core desktop processors. These processors amalgamate Intel’s Arc GPU with a neural processing unit (NPU) within a streamlined system-on-chip architecture. They are proficient in managing “demanding graphics workloads” at the network edge, applicable across various domains such as retail, education, smart cities, and industry. For instance, they could be deployed for generative AI-powered sales kiosks or intelligent point-of-sale systems in retail environments, interactive whiteboards in educational settings, and AI-driven vision devices within industrial manufacturing, as highlighted by Intel.

Slightly lower in the hierarchy are the fresh Intel Core processors tailored for edge tasks. These processors blend the GPU capabilities of Intel’s 13th Gen Intel Core mobile chips with the added advantage of “LGA socket flexibility,” thereby improving system scalability and deployment efficiency. According to Intel, they boast a remarkable 2.57 times enhancement in graphics performance compared to the standard 13th Gen Intel Core desktop chips from which they derive.

Furthermore, Intel introduced two new series of Intel Atom processors: the x7000RE and the x70000C. According to Intel, these are low-power, AI-enabled chips intended for industrial and manufacturing settings as well as networking and telecommunications. In order to carry out operations like identifying zero-day threats, enhancing packet and control plane processing, and AI-powered quality control, they have integrated deep learning inference capabilities.

Finally, Intel unveiled a new Intel Arc GPU designed for Edge applications, boasting a substantial increase in edge AI performance compared to its predecessors. This advancement facilitates accelerated AI workloads in areas such as graphics processing and media, as highlighted by the company.

Regarding Altera, its newest Agilex 5 SoC FPGAs are programmable chips engineered to bolster AI functionalities within edge devices, eliminating the necessity for a dedicated AI accelerator. These chips offer up to twice the performance compared to the company’s previous Agilex 7 FPGAs.

FPGAs are low-power computer chips that can be configured to handle very specific use cases before being deployed, thanks to their modifiable circuitry. Altera is releasing the chips with tools like the Intel OpenVINO toolkit, Quartus Prime software, and FPGA AI Suite to support this. Using these tools, developers have the capability to program the chips for a targeted workload, generate AI intellectual property blocks, and seamlessly incorporate them into the FPGA’s design prior to their deployment in devices.

Holger Mueller, an Analyst at Constellation Research Inc., noted that amidst the ongoing excitement surrounding generative AI in the cloud, there’s been a tendency to overlook the substantial volume of enterprise workloads operating at the edge. These workloads often cannot leverage the technology due to constraints such as insufficient computing power and limited low-latency communication. Mueller explained that Intel and Altera are striving to resolve this fundamental challenge.

“There are thousands of enterprises running critical apps at the edge, and it’s these that Intel and Altera want to target. So it’s good to see Altera innovating and helping Intel fuel its innovation and show that AI can and will have a significant revenue impact this year,” the Mueller said.

Altera emphasized that its new FPGAs will also expedite enterprise innovation, providing companies with a simpler avenue to harness the vast volumes of data generated at the edge by deploying “embedded AI devices.”

“We announced the launch of the new Altera brand with the goal of bringing leading technologies and innovations more quickly to the FPGA market. Altera is leading the new FPGAi era, tightly coupling programmability with tensor capabilities and infusing FPGA and AI tools to create a best-in-class developer experience,” said Sandra Rivera, Chief Executive of Altera.