-
Hard Macros Are Back in Town Thanks to the Arm Client CSS
Alongside the new Cortex-X925 and Cortex-A725 CPUs and Immortalis-G925 GPU, Arm has unveiled client CSS, predesigned chip blocks for peak performance, power and area. However, Arm must clarify the CSS’s target market and the CPUs’ and GPU’s performance gains. continue reading
-
AmpereOne Adds Cores as Ampere Toys with Two-Edged Swords
Ampere has announced a new 256-core, 3 nm AmpereOne processor and an AI collaboration with Qualcomm. Ampere’s efficiency focus may not be enough against competitors and for cloud giants building their own chips. Arm’s ascendancy in computing suggests Ampere consider a niche beyond data centers. continue reading
-
1Q24 Processor Market: Four Billion is Not Enough
On the back of rising Epyc sales and a strengthening MI300 outlook, AMD is gaining data-center share on Intel. The PC-market recovery continues, while the embedded-market recovery should be underway in Q3. continue reading
-
Google Trillium Blooms, Promising AI Efficiency and Speed
Google unveils the Trillium TPU, its next-gen AI chip (NPU) that features larger MXU matrix units to boost throughput and updated SparseCores to handle embeddings. continue reading
-
Plug: The SHD Group is Developing a Complimentary Edge-AI Processor and Ecosystem Report
I’ve joined the SHD Group to produce an edge-AI report. Intended to be a resource for both product developers and technology suppliers, it will detail processors integrating AI, accelerators, software, and services and deliver insights into edge AI application requirements. We’re partnering with the Edge AI and Vision Alliance, and our research dovetails with recent continue reading
-
Using It to Make AI Chips, SoftBank Would Risk Arm
Arm may design and sell AI chips, targeting production by 2025. Directed by majority shareholder SoftBank Group, the move is risky. Selling chips is less likely to succeed than licensing AI IP, which Arm already does, and won’t alienate Arm’s customers. continue reading
Popular
Sponsored Content and Plugs
- BWR 12: The Memory Episode
- Slash Server Costs Without Hardware or Software Changes
- Ceva Boosts NeuPro-M NPU Throughput and Efficiency
- MEXT Helps IT Leaders Find the Sweet Spot
- Chips Act Backs Chiplets
Other Sites You Might Like
Read More
Amazon AMD Arm auto Broadcom business Cerebras Ceva consumer CPU data center DPU DSP edge AI embedded FPGA Google GPU Imagination industrial Intel interconnect Marvell MCU MediaTek Meta MEXT Microsoft MLPerf NPU (AI accelerator) Nvidia NXP OpenAI PC process tech Qualcomm Renesas RISC-V SambaNova Semidynamics SiFive smartphone SoftBank software Tenstorrent