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

Current Articles

November 29, 2021

• AMD CDNA2 Targets Supercomputers
  AMD’s Instinct MI250 and MI250X GPUs deliver extreme floating-point performance through massive parallelism, targeting high-end scientific and government research with leading FP64 throughput.
  

• Imagination CXT Fires Photons
  At 1.3 billion rays per second, the Photon architecture in Imagination’s new PowerVR CXT RT3 GPU enables hardware-accelerated real-time ray tracing for the first time in mobile devices.
  

• Kneron KL530 Targets Automotive
  Kneron is targeting emerging computer-vision models with its KL530. The new SoC delivers 0.5 trillion INT8 operations per second (TOPS) and supports INT4 format for greater throughput.
  

November 22, 2021

• Cracking Open Snapdragon’s 4-Pack
  In its latest mainstream push, Qualcomm delivers a fresh four-pack filled with CPU, GPU, and camera upgrades. It’s raising performance across multiple segments from low-end to mid-premium.
  

• Editorial: The Next Microprocessor Decade
  Some trends that are big today began more than a decade ago as inconspicuous events. How will recent events look in 10 years? I foresee new architectures and new accelerators but the same transistors.
  

• Epyc Roadmap Gets Specialized
  AMD recently disclosed new details of its Epyc server-processor roadmap, including Milan-X with V-Cache, Genoa based on Zen 4 cores, and Bergamo scaling to 128 cores.
  

November 15, 2021

• History of the Microprocessor, Part 3
  As we celebrate the anniversary of the Intel 4004, this article extends our microprocessor history by covering the top trends from the past decade, including the rise of AI, Nvidia, processor startups, and RISC-V.
  

• 50 Billion Transistors in 50 Years
  Microprocessors have improved by seven orders of magnitude—from 2,300 to 54 billion transistors—in only five decades. To illustrate these changes, we highlight a handful of industry-defining products.
  

• Apple Scales M1 GPU to the Max
  Apple’s M1 Max processor delivers a massive boost in integrated-graphics performance compared with the original M1, offering throughput similar to that of Nvidia’s top-end laptop GPU card.
  

November 8, 2021

• Pixel 6 Packs Semicustom Processor
  For the first time, Google designed the main processor in its flagship smartphone. The Tensor SoC is based on a Samsung design but integrates Google’s AI accelerator and image processor (ISP).
  

• R-Car S4 Boosts Connected-Car Safety
  The Renesas R-Car S4 serves as a vehicle’s central networking hub, but it also integrates ASIL-capable application CPUs, real-time CPUs, and safety microcontrollers along with multiple hardware security features.
  

• Syntiant Grows Into Computer Vision
  Following its success in low-power wake-word detection, Syntiant has expanded into computer vision with its newest chip. The NDP200 can handle simple video tasks such as presence detection.
  

November 1, 2021

• Alder Lake Steals Gaming-CPU Crown
  Intel’s Core i9-12900K chip delivers a double-digit performance advantage over AMD’s fastest Ryzen processor. To do so, it runs at a sweltering 241W while using new superfast DDR5 DRAM.
  

• Marvell 5nm Switch Handles 5G RAN
  Marvell’s first Ethernet switch built in 5nm technology is for carrier access rather than data centers. The Prestera DX7321 joins the DX73xx family, which handles Ethernet speeds up to 400Gbps.
  

• Cadence HiFi Family Adds a Little One
  The new Tensilica HiFi 1 DSP consumes less die area and power than the popular HiFi 3 but adds new instructions that improve performance on Bluetooth 5’s Low Complexity Communication Codec (LC3).
  

October 25, 2021

• Apple A15 Extends Battery Life
  The new A15 processor doubles its system cache to an industry-leading 32MB and drives double-digit performance gains across the CPU, GPU, and AI engine, while boosting battery life.
  

• Editorial: Intel AI Strategy Ready to Reboot
  To avoid customer confusion and strengthen its market position, Intel must rationalize an array of overlapping AI chips that have taxed the resources of even the world’s largest processor vendor.
  

• Coherent Logix Debuts Edge-AI SoC
  After serving niche applications with its reconfigurable designs, Coherent Logix introduced an SoC that will serve as a development vehicle for edge AI, software-defined radio, and other real-time applications.
  

October 18, 2021

• Google Builds Video-Transcoder Chip
  To better support its massive YouTube business, Google has created its own video-transcoding chip. The VCU design offers a 40x performance gain compared with CPUs.
  

• Loihi 2 Advances Spiking R&D
  Intel Labs has developed a second-generation neuromorphic processor using the Intel 4 process. It’s also offering the open-source Lava framework to advance neuromorphic software.
  

October 11, 2021

• Anton 3 Cracks Molecular Dynamics
  Billionaire David E. Shaw has spent two decades quietly building highly specialized supercomputers. Anton 3 is his newest ASIC for accelerating molecular-dynamics simulations.
  

• Samsung Demos In-Memory Processing
  The company disclosed architecture and performance details for the processor-in-memory (PIM) version of its popular Aquabolt HBM2. The HBM-PIM, dubbed Aquabolt-XL, adds compute to a drop-in HBM2 stack.
  

• Cadence Takes Tensilica AI to the Max
  The new Tensilica AI lineup comprises three product tiers that Cadence calls AI Base, AI Boost, and AI Max, along with a new accelerator and a multicore model that can deliver up to 32 TOPS.
  

October 4, 2021

• Enflame Lifts Veil on Training Chip
  The well-funded Chinese AI startup revealed architectural details of its first chip shortly after announcing its T20 second-generation design with 4x better inference performance plus TF32 for training.
  

• Dimensity 810 Trims Phone Power
  The new Dimensity 810 boosts the single-thread CPU throughput by 17% compared with the Dimensity 800 while reducing power consumption, while the Dimensity 920 upgrades gaming performance.
  

• Kompanio Powers 5G Tablets
  MediaTek has added a new brand to its portfolio by introducing two tablet processors, refreshing its Arm compute strategy by reusing its latest Dimensity smartphone chips.
  

September 27, 2021

• Qualcomm Spills Cloud AI 100 Guts
  Shortly after demonstrating outstanding performance per watt in recent MLPerf Inference results, Qualcomm revealed internal-architecture details of its Cloud AI 100 accelerator at last month’s Hot Chips.
  

• SOT MRAM Improves Access Speed
  Startup Antaios is leading the commercialization of spin-orbit-torque (SOT) magnetoresistive RAM (MRAM), which promises greater longevity and faster access times than previous MRAMs.
  

• Editorial: Intel, AMD Differ on Chiplets
  Intel has developed several high-performance chiplet technologies, but their cost restricts them to low volume. AMD’s lower-cost approach raises volume today but won’t meet future requirements.
  

September 20, 2021

• Tesla D1 Tackles AI Training
  In a surprising move, Tesla has developed its own AI training accelerator. The D1 design crams 25 chips into an enormous wafer-scale package that generates an industry-leading 9 petaflop/s.
  

• Cerebras Performs Extreme Training
  The company hopes to meet the demands of huge neural networks by offering its new MemoryX and SwarmX systems, which aim to reduce GPT-3 training time to less than one day.
  

September 13, 2021

• Sapphire Rapids Spans Tiles
  The next-generation Xeon processor is Intel’s biggest leap forward in years, delivering greater core counts, more per-core performance, integrated accelerators, and the latest memory and I/O interfaces.
  

• Alder Lake Extends Battery Life
  Alder Lake is Intel’s first mainstream PC processor with a hybrid architecture, combining big and small cores to increase both performance and battery life.
  

• Mount Evans Puts Intel in DPU Range
  The company recently disclosed architectural details of its first data-processing unit, combining a control plane and a programmable data plane to handle Ethernet speeds up to 200Gbps.
  

September 6, 2021

• Intel Stacks Tiles on Ponte Vecchio
  The new high-performance GPU is an engineering marvel, using Intel’s EMIB and Foveros technologies to create a heterogeneous processor combining 47 discrete die (tiles) built in five different process nodes.
  

• Golden Cove Adds Matrix Units
  Intel’s new flagship CPU microarchitecture delivers a big performance boost for general-purpose code while AI operations increase 8x using the new Advanced Matrix Extensions (AMX).
  

• Gracemont Bulks Up in Performance
  Intel’s family of “little” CPUs hits a bold new performance level by doubling the execution width and making other changes, offering a 20% jump in per-clock performance over its predecessor.
  

More Articles

Purchase an Article