Our privacy policy has been updated to incorporate GDPR considerations. See our updated policy here.
Order a Guide Order a Guide Order a Guide

Most Recent Guides & Forecasts

A Guide to Processors for Deep Learning
Covers processors for accelerating deep learning, neural networks, and vision processing for AI training and inference in data centers, autonomous vehicles, and client devices.
Communications Semiconductor Market Forecast 2018–2023
Provides five-year revenue forecasts for more than 20 categories of communications semiconductors, including Ethernet products, broadband interface chips, embedded processors, server processors and FPGAs.
Communications Semiconductor Market Share 2018
Provides market share data for more than 20 categories of communications semiconductors, including Ethernet products, broadband interface chips, embedded processors, server processors and FPGAs.
A Guide to Multicore Processors
Covers 32- and 64-bit embedded processors with four or more CPU cores that are used for wires and wireless communications, storage, security, and other applications.
A Guide to Ethernet Switch and PHY Chips
Covers data-center switch chips for 10G, 25G, 40G, 50G, and 100G Ethernet. Also includes physical-layer (PHY) chips for 10GBase-T and 100G Ethernet.
A Guide to Processors for IoT and Wearables
Covers processors and connectivity chips for IoT clients and wearable devices, focusing primarily on single-chip solutions with integrated radios.

More Guides »

White Papers

Accelerating AI Performance in an Open Software Ecosystem
Nvidia's software stack, based on its proprietary CUDA, creates a high barrier to entry for challengers. SYCL, the only open alternative to CUDA that has multivendor support, enables customers to move to higher performance hardware while retaining software flexibility. A commercial version of SYCL, branded ComputeCpp, is offered by Codeplay.
Arm Ecosystem Reduces SoC Design Cost and Time to Market
This paper discusses three critical facets of the Arm ecosystem: design verification, physical design, and software development. Each one directly affects the time required to develop a complete SoC and software stack.
AI Requires Many Approaches
Artificial intelligence (AI) is being applied to a wide range of problems, so no single processor can support them all. This paper describes and compares the various approaches needed for data center, autonomous driving, and consumer/IoT applications. 
Universal RDMA: A Unique Approach for Heterogeneous Data-Center Acceleration
Now available in 10G, 25G, 50G, and 100G Ethernet server adapters, RDMA reduces network processing overhead and minimizes latency. Cavium's FastLinQ stands out by handling both RoCE and iWARP, which are alternative protocols for RDMA over Ethernet.
More White Papers »
Linley Reports Linley Reports Subscribe to our Reports

April 6, 2020

Connected Cars Get a Cellular Plan
China has approved the use of Cellular Vehicle-to-Everything (C-V2X) technology and plans to start deployment next year. In the US, however, conflict at the FCC has stalled deployment.
Qualcomm UltraSAW Targets FBAR
Qualcomm’s new UltraSAW filters match up well against Broadcom’s dominant FBAR technology.
Ampere Pushes Neoverse to 80 Cores
Ampere Computing is sampling Altra, an 80-core server processor based on Arm’s Neoverse N1 design. In addition to the high core count, the 7nm chip boosts the peak CPU speed to 3.0GHz.
First 112Gbps PAM4 PHYs Arrive
Inphi and Marvell are first to sample 400Gbps Ethernet PHYs that enable 112Gbps-per-lane chip-to-module interfaces. The former is sampling Spica for optical modules, whereas the latter offers its 88X9121P for the system side.

March 30, 2020

Intel Snow Ridge Gains 5G Wins
The new Atom P5900 combines up to 24 Tremont CPUs, a 440Gbps Ethernet switch, and packet accelerators, making it well suited to 5G base stations and other networking equipment.
Versal Premium Targets Core Networks
Xilinx is expanding its Versal family with new FPGAs for high-speed networking and communications. New features include 600G Ethernet, 400Gbps crypto, and 600G Interlaken.
Unisoc Joins 5G-Smartphone Race
The Tiger T7520 is Unisoc’s first processor to integrate a 5G baseband, and it’s also the first announced product to employ 6nm EUV technology at TSMC.

March 23, 2020

AMD Develops Compute GPUs
AMD won its second major supercomputer design, El Capitan, using the 5nm Genoa processor, a next-generation Infinity Fabric, and a new GPU called CDNA that’s optimized for AI and HPC.
Ceva XC16 Stays Ahead of 5G Rollout
Ceva’s XC16 DSP is a dynamically configurable multicore/multithread design. At run time, users can configure it as two symmetric dual-vector processors or as a single quad-vector processor.
Editorial: Deep Learning Gets Small
Google and the TinyML Foundation are encouraging the development of small neural networks that can run on microcontrollers, but even these tiny AI models benefit from deep-learning hardware.

March 16, 2020

Marvell Upgrades to Octeon TX2
Marvell is shipping its first major upgrades of former Cavium products, including Octeon TX2 high-speed networking processors and Octeon Fusion processors for 5G macrocells.

More Articles »

2020 Linley Spring Processor Conference - Register Now


Linley Spring Processor Conference 2020
Coming April 6-9, 2020
Register Now!
Linley Fall Processor Conference 2020
Coming October 28-29, 2020
Hyatt Regency, Santa Clara, CA
More Events »

Media Appearances


Linley Newsletter
Analysis of new developments in microprocessors and other semiconductor products
Subscribe to our Newsletter »