» Archived Articles | Login | Subscribe | Free Sample Articles

Current MPR Articles

October 24, 2022

• Zen 3 Goes Embedded
  AMD’s new Ryzen Embedded V3000 family addresses networking and storage with the Zen 3 microarchitecture. It features up to eight CPUs but omits GPUs.
  

• PentaG-RAN IP Targets Base Stations
  Ceva’s PentaG-RAN is a baseband IP block for cellular infrastructure. Packing multiple fixed-function units and two DSP types, it targets SoCs for RUs, DUs, small cells, and beam-forming chips.
  

• Neoverse V2 Pressures x86 Servers
  Arm’s next Neoverse CPU, code-named Demeter, will appear in Nvidia’s Grace processor in 2023, matching the single-thread performance of mainstream Xeon products for the first time.
  

October 17, 2022

• Nvidia Ada Doubles GPU Performance
  Nvidia’s newest gaming GPU, code-named Ada Lovelace, builds on the Ampere design with faster clock speeds, far more cores, a massive cache memory, and faster ray tracing.
  

• STMicroelectronics Enters the Zone
  STMicroelectronics’ new SR6P7x and SR6P6x automotive SoCs aggregate low-level ECUs into domains or zones. With a novel choice of nonvolatile memory for code, the company has a strong focus on deterministic real-time performance.
  

• Marvell Ups LiquidSecurity ECC Speed
  Seven years after releasing LiquidSecurity, Marvell has unveiled a second-generation PCIe card that increases RSA throughput by 20% and ECC throughput tenfold.
  

October 10, 2022

• Intel Debuts Discrete Desktop GPUs
  Intel has expanded its discrete graphics card offering into the low end of the desktop market. It also has two new models for the data center.
  

• AMD FPGA Targets 400Gbps Smart NICs
  Having closed its Xilinx acquisition, AMD is launching a new FPGA with much more CPU power than previous members of its Versal family, targeting it at 400Gbps smart NICs.
  

• Sapeon Adds AI to Edge Servers
  After spinning off from SK Telecom, Sapeon has become the first Korean company to deliver an AI chip. The X220 provides an efficient accelerator for both video analysis and language processing.
  

October 3, 2022

• Raptor Lake Adds Little Cores
  Despite a short design cycle, Intel boosted the performance of its 13th Generation Core processor by up to 15% over the previous generation through higher clock speeds and more “efficiency” cores.
  

• AMD Zen 4 Adds AVX-512
  Zen 4 increases IPC and runs much faster than Zen 3 owing to optimized circuits, a process shrink, bigger buffers, deeper queues, and a larger micro-op cache.
  

• New SiFive CPUs Hit the Road
  SiFive has entered the automotive market, offering small embedded and application cores for ASIL B and ASIL D systems.
  

September 26, 2022

• Ryzen 7000 Upgrades to Zen 4 and 5nm
  AMD’s Ryzen 7000-series processors for high-end desktops employ the new Zen 4 CPU to achieve greater performance and power efficiency.
  

• Snapdragon 6 Boosts Graphics and AI
  Qualcomm updated its Snapdragon 6 and 4 lines, moving to a new process, improving performance, and selectively adding features. The new nomenclature aligns with the 7- and 8-series.
  

• Hopper, Startups Debut MLPerf Scores
  The MLPerf 2.1 inference release includes preliminary results that put Nvidia’s Hopper H100 in the performance lead. Asian startups Biren and Sapeon also made impressive debuts.
  

September 19, 2022

• Updated Trident 4 Monitors Connections
  Broadcom has added a network-scanning engine to its 12.8Tbps Trident 4 Ethernet switch. Capable of fingerprinting every packet, the engine improves network security.
  

• Cerebras Dives Into WSE Architecture
  The startup disclosed new details about how its tiny cores deliver tremendous performance and how its sparsity support boosts performance when training large AI models.
  

• CXL 3.0 Opens New Use Cases
  The newly unveiled CXL 3.0 introduces memory sharing, direct device peer-to-peer memory access without involving a host, and multilevel switching. A new global fabric-attached memory can be shared by 4,096 hosts.
  

September 12, 2022

• Nvidia Grace Supports AI Acceleration
  To tackle the largest AI models, Nvidia has designed a processor to feed its powerful new Hopper GPU. Grace has twice the memory bandwidth of any x86 processor and can hold GPT-3 in DRAM.
  

• Lightmatter Connects Chiplets Optically
  Lightmatter’s Passage substrate is an active photonic interposer for interconnecting chiplets. All the photonic components and supporting electrical circuits reside in a single multi-reticle piece of silicon.
  

• MediaTek T830 Accelerates 5G Broadband
  To capitalize on growing adoption of 5G cellular for broadband access, MediaTek has boosted the throughput of its fixed-wireless-access chipset.
  

September 5, 2022

• Intel PC Processors Move to Chiplets
  Meteor Lake will introduce a multidie arrangement using Intel’s silicon-substrate and advanced packaging technologies. This chiplet approach will appear across all PC price tiers beginning late next year.
  

• Biren GPGPU Aims for the Clouds
  In the data-center accelerator race, the three-year-old startup has burst from the gate with a chiplet-based design that aims to compete with Nvidia for general-purpose-GPU (GPGPU) cloud computing.
  

• Grai Matter Pivots to Floating Point
  The startup has made progress on its Grai VIP deep-learning accelerator, changing native processing from INT8 to FP16 and adding audio workloads to its target applications.
  

August 29, 2022

• Untether Boqueria Targets AI Lead
  The startup’s unique at-memory architecture targets an eye-popping 30 teraflop/s per watt and 12,000 teraflop/s in a single card. The second-generation chip is due to sample in 1H23.
  

• Editorial: MCUs Moving to RISC-V
  Renesas has introduced the first embedded processor with a user-programmable RISC-V CPU. Chinese processors are already adopting RISC-V; more will follow in a direct challenge to Arm.
  

• ST BlueNRG Tracks People and Assets
  ST’s newest BlueNRG microcontroller targets location services and reduces cost.
  

August 22, 2022

• Tomahawk 5 Switches at 51.2Tbps
  Broadcom is the first vendor to sample a 51Tbps data-center switch chip. The 5nm device features 100Gbps serdes and port speeds of up to 800G Ethernet.
  

• Cognifiber Targets Purely Photonic AI
  The Israeli startup has built a proof-of-concept system for a simple optical neural network, but it will require additional effort to handle commercial AI models.
  

• Functional Safety Moves to Chips
  Functional safety affects an increasing number of chip and IP designs. Vendors are precertifying chips, IP, and tools to make it easier for their customers to achieve certification.
  

August 15, 2022

• Google’s Tensor DLA Lags in Area, Speed
  Among flagship-smartphone processors, Google’s first custom design has the largest AI engine but the worst AI performance. In contrast, the Dimensity 9000’s AI engine is the most efficient.
  

• TI AM625 Updates Industrial SoCs
  TI has updated its low-cost SoC family with 64-bit CPUs, lower power, and improved security and functional-safety support, enabling it to target automotive as well as industrial applications.
  

• Intel Finally Gives Up on Optane
  Intel and Micron have developed a revolutionary new memory technology called 3D XPoint, but under the Optane brand it never met expectations as either an SSD or DIMM replacement.
  

August 8, 2022

• Qualcomm Leaps Forward With W5+
  Qualcomm has updated its smartwatch platform with new chip designs. Reduced power enables a visible watch face even when idle; reduced chipset area enables smartwatches that operate globally.
  

• Apple Loses GPU-Performance Lead
  Our die-photo analysis reveals that Samsung devoted too little area to its AMD-powered GPU, contributing to its performance shortfall. MediaTek and Intel both outperform the latest Apple chip.
  

• Intel Shines Light on Copackaged Optics
  Intel’s eight-laser silicon-photonics research chip could be a breakthrough for copackaged optics. It places the laser array on a silicon chip that can eventually be integrated in the package.
  

August 1, 2022

• MediaTek Delivers Efficient Cortex-X2
  Our die-photo analysis reveals that the Dimensity 9000 features the smallest Cortex-X2 design, but Apple’s Avalanche CPU still leads in performance and power efficiency.
  

• Enlightra Team Demos Photonic AI Engine
  A research team led by Enlightra’s founders demonstrated a new method of photonic AI inference that is smaller than earlier photonics implementations. Using photonics also provides opportunities for resource sharing that aren’t available with electronics.
  

• Editorial: OEMs Lack Good Value Options
  Overly cautious about rising development costs and risk, vendors aren’t refreshing their lower-cost processors, hurting themselves and customers. They must consider business cases other than merely chasing the biggest customers.
  

More Articles