» Archived Articles | Login | Activate My Subscription | Subscribe | Free Sample Articles | Purchase an Article

Current MPR Articles

January 27, 2020

• Ryzen Mobile Doubles CPU Count
  AMD’s 7nm Renoir processor brings eight cores to 15W ultrathin notebooks, an industry first. It fills out the Zen 2 generation across desktop and mobile platforms.
  

• Editorial: Nvidia AI Software Hard to Beat
  Several new deep-learning accelerators offer sizable advantages over the best GPUs on a particular benchmark or metric, but many customers still prefer Nvidia for its software stack.
  

• NXP S32G Is Master of Auto Domains
  The S32G vehicle-network processor offers a unique lockstep CPU architecture that supports ASIL D operations. It also has accelerators for both Ethernet and legacy automotive networks. 
  

January 20, 2020

• Analysts’ Choice Winners for 2019
  The Linley Group recognizes 2019’s top products in the categories of AI, embedded, mobile, networking, server, processor IP, and related technology. And the winners are…
  

• Year in Review: Arm Flexes IP to Battle RISC-V
  Arm was by far the most active processor-IP vendor in 2019, but the RISC-V community also surged. Licensable cores for AI acceleration joined a flood of new CPU, DSP, and GPU cores.
  

• Intel Extends 3D Packaging With ODI
  Intel’s new Omni-Directional Interconnect builds on EMIB and Foveros to improve performance, capabilities, and flexibility for products using 3D packaging.
  

January 13, 2020

• Year in Review: 5G Smartphones Ready to Boom
  After an initial launch in 2019, 5G smartphone shipments are set to soar in 2020. The new year will also see greater use of AI, the first 5nm processors, and more gaming phones.
  

• Marvell Accelerates In-Car Networks
  Marvell is sampling the second generation of its purpose-built automotive switch chips. It also bought Aquantia, adding PHYs to its multigigabit automotive-Ethernet portfolio.
  

• GreenWaves GAP9 Goes Faster
  GreenWaves’ new low-power RISC-V microcontroller uses FD-SOI and a second-generation deep-learning accelerator to increase power efficiency on AI workloads by 5x.
  

January 6, 2020

• Groq Rocks Neural Networks
  The startup has created a new AI architecture that implements a single massive programmable engine rated at an industry-leading 1,000 TOPS. It runs ResNet-50 with extremely low latency.
  

• Year in Review: AI Is Livin’ on the Edge
  Embedded-processor trends include adding AI acceleration, stronger chip-level security, multidie packaging, advanced driver-assistance systems (ADASs), and microcontrollers that either omit flash memory or replace it.
  

December 30, 2019

• Snapdragon 765 Enables 5G Roaming
  Qualcomm’s new mid-premium platform, the Snapdragon 765, is the industry’s first to integrate a 5G mmWave and sub-6GHz capability. The first phones using it should arrive in 1Q20.
  

• Editorial: Custom Chips Are Back in Vogue
  Hyperscale-data-center operators and wireless-infrastructure OEMs are creating new demand for ASICs. New ASIC vendors such as Intel and Marvell use large IP catalogs to allow customers to focus on their own IP.
  

December 23, 2019

• Snapdragon 865 Dis-Integrates
  Qualcomm’s newest flagship smartphone processor requires separate 5G and Wi-Fi chips but delivers leading-edge performance and works with millimeter-wave networks.
  

• MediaTek Rides 5G to Flagship Phones
  The Dimensity 1000 is MediaTek’s first 5G processor and allows it to target premium smartphones with the latest CPU and GPU cores plus strong AI performance.
  

• Year In Review: Competition Heats Up Data Centers
  Many new server processors, AI accelerators, and 400G Ethernet chips are headed for production in 2020. Hyperscale-data-center operators are also developing their own processors and accelerators.
  

December 16, 2019

• Tomahawk 4 Switch First to 25.6Tbps
  Broadcom is sampling a 400Gbps Ethernet switch that packs an unprecedented 512 serdes to double bandwidth relative to the shipping Tomahawk 3. 
  

• PowerVR Readies Ray Tracing
  Imagination Technologies has overhauled its PowerVR GPU architecture and is introducing new GPU cores that address all levels of graphics performance. The new A-Series also delivers AI throughput that competes well with other licensable GPUs.
  

• Editorial: Intel Looks Beyond x86
  CEO Bob Swan has replaced a long-time strategy—pushing x86 CPUs for all applications—with a broader offering of GPUs, AI accelerators, and FPGAs for workload-specific optimization. 
  

December 9, 2019

• Mobileye Expands Into Robotaxis
  The company announced plans to expand beyond EyeQ processors, including lidar sensors, complete self-driving systems, and robotaxi services, along with greater licensing of map data.
  

• Photonic Computing Finds Fit in AI
  Optical neural networks are moving closer to commercialization as photonic computing moves out of university labs. A few startups are leading the charge and starting to attract funding.
  

• Graphcore Up and Running
  After taking nearly a year to tune its software, Graphcore is finally taking the wraps off its C2 deep-learning accelerator, which will be available through Azure Cloud and in some Dell EMC systems.
  

December 2, 2019

• Centaur Adds AI to Server Processor
  Centaur, Via’s chip-design arm, disclosed a new deep-learning accelerator and an x86 microarchitecture that combine on a single chip to form a highly integrated server processor. 
  

• Exynos 990 Enables 5G Smartphones
  Samsung’s newest smartphone processor and 5G modem provide a sneak preview of its next flagship Galaxy. They both use the company’s 7nm EUV process.
  

• Ryzen Threadripper 3 Thrashes Xeon
  AMD Threadripper 3 processors for high-end desktops and workstations exceed their predecessors by upgrading their CPUs, clock speeds, caches, DRAM controllers, PCI Express interfaces, and internal designs.
  

November 25, 2019

• Alibaba Takes AI Performance Lead
  The Chinese Internet giant has developed its first ASIC to accelerate neural networks, posting ResNet-50 throughput that’s 10x better than Nvidia’s fastest GPUs.
  

• MLPerf Debuts Inference Benchmarks
  The MLPerf group has released a modest set of inference benchmarks and use cases. The first results include submissions from cloud-service providers (CSPs), server vendors, chip vendors, and professional-services companies.
  

November 18, 2019

• Chiplet Technology Proliferates
  Chiplets are getting lots of attention in IC-design circles, as real products are shipping in high volumes from leading vendors. Package and interface technologies are advancing chiplet-based designs.
  

• Renesas RA MCUs Strengthen Security
  Renesas Advanced (RA) microcontrollers will include the company’s first models that implement Arm’s TrustZone technology. The 32-bit RA family comprises chips with Cortex-M23 and Cortex-M4 CPUs; future products will add Cortex-M33.
  

• Editorial: Third-Party Chiplet Challenges
  The ODSA hopes to shift the chiplet market from in-house to third-party designs, but to enable this transition, it must overcome several design and manufacturing challenges.
  

• Arm Offers Bespoke Cortex-M CPUs
  Arm is opening the Arm v8-M architecture to allow custom instructions. Cortex-M33 is the first CPU to support this new feature, but all future Arm v8-M cores will as well.
  

November 11, 2019

• Emerging Memories: Why Now?
  How will an emerging memory technology rise to prominence? Both technology and market dynamics will lead to the replacement of today’s technologies with something altogether different.
  

• ARC VPX5 Sports Four-Way VLIW
  The new VPX5 is the first Synopsys ARC DSP to employ a VLIW architecture, combining a scalar CPU, a 512-bit vector/SIMD unit, and an optional vector floating-point engine. 
  

• SiFive Secures RISC-V
  SiFive recently disclosed its Shield secure-platform architecture, which includes hardware and software as well as provisioning services. Shield’s baseline capabilities are similar to those of Arm’s TrustZone.
  

November 4, 2019

• Intel’s Tremont: A Bigger Little Core
  Tremont is Intel’s newest low-power x86 CPU, boosting IPC by 30%. The new core design will debut in the Lakefield processor—a heterogeneous multicore SoC for thin notebooks and tablets.
  

• NovuMind Launches Tensor Processor
  Startup NovuMind is preparing to tape out a neural-network accelerator that it estimates will deliver 150 trillion operations per second (TOPS) while consuming just 5W, achieving impressive efficiency.
  

• BittWare Puts Achronix On Board
  A new partnership with BittWare gives Achronix another route to market for its Speedster7t FPGAs. The companies recently announced the VectorPath PCI Express accelerator card based on Speedster7t.
  

• Mythic Reveals Data-Flow Architecture
  AI startup Mythic revealed more details of its first chip, which performs matrix multiplication in the analog domain. Conventional digital circuits, however, manage data flow and other tasks.
  

• Cornami Right-Sizes Systolic Arrays
  After testing FPGA prototypes of its dynamically configurable AI processor, startup Cornami projects its first chip will run ResNet-50 at an incredible 105,000fps while burning only 30W.
  

More Articles

Purchase an Article