AMD Su Zifeng visited TSMC and others: NVIDIA's strongest graphics card killer MI300X is going into mass production
Views : 104
Update time : 2023-07-25 15:52:41
July 19 news, following the end of May after the Taipei Computer Show, AMD CEO Su Zifeng recently visited partners once again, including TSMC, Sun Micron, PEGATRON and other companies.
AMD officials did not disclose the specific content of Su Zifeng's talks with manufacturers, but the industry believes that this visit to the supply chain manufacturers, mainly for the Q3 quarter mass production of MI300 series AI graphics card outposts, to ensure production capacity, after all, NVIDIA's AI graphics card has encountered the problem of oversupply.
Today's AI graphics card market is hot, but NVIDIA is a dominant player, accounting for 90% of the share, A100, H100 two graphics cards priced at more than 100,000 to 250,000 yuan is still in short supply, the domestic special supply of A800, H800 is even more price increases to snatch up the AMD naturally will not miss this market.
At the end of June AMD officially released Instinct MI300A and Instinct MI300X two AI-accelerated graphics cards, of which the former is the first time the CPU + GPU combined products, the latter is a pure GPU, mainly used for AI, HPC high-performance computing and other fields.
MI300X is a product that benchmarks NVIDIA's strongest graphics card, H100, with up to 8 XCD cores, 304 groups of CU units, 8 groups of HBM3 cores, and the memory capacity has been boosted to 192GB, which is equivalent to 2.4 times that of NVIDIA's H100 80GB, and at the same time, the HBM memory bandwidth reaches up to 5.2TB/s, and the Infinity The HBM memory bandwidth is up to 5.2TB/s, and the Infinity Fabric bus bandwidth is 896GB/s, which is also far more than the NVIDIA H100.
The MI300X's 192GB of video memory is unprecedented in both performance and capacity, and is especially suited to AI computing, with a single card supporting 40 billion parameter models.
Price, MI300X did not announce the official suggested price, but the benchmark H100 means that the price is close, only slightly cheaper, guessing that it is also in the 200,000 yuan or more.
The LDK220 series of low-dropout linear regulators (LDOs) is a high-performance device designed specifically for low-power consumption and high-precision voltage regulation, widely used in scenarios such as consumer electronics, industrial control, and po
Xilinx Spartan®-7 FPGA Family stands as a defining solution in the mid-range FPGA landscape, blending high performance, energy efficiency, and cost-effectiveness to redefine versatility for industrial, IoT, and consumer electronics applications. Built on
The Altera FLEX series was more than a lineup of FPGAs—it was a blueprint for how programmable logic devices could evolve to meet diverse market needs. The FLEX 8000 laid the architectural groundwork, the FLEX 10K redefined functionality with embedded mem
Among TI standout offerings, the LM4765 and LM4766 are dual-channel amplifiers designed to cater to diverse audio needs—from compact setups to high-fidelity systems. While sharing the same product lineage, these chips differ significantly in power output,
