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