- Nvidia's H100 AI GPUs cost up to four times more than AMD's competing MI300X — AMD's chips cost $10 to $15K apiece; Nvidia's H100 has peaked beyond $40,000: Report https://www.tomshardware.com/tech-industry/artificial-intelligence/nvidias-h100-ai-gpus-cost-up-to-four-times-more-than-amds-competing-mi300x-amds-chips-cost-dollar10-to-dollar15k-apiece-nvidias-h100-has-peaked-beyond-dollar40000 26 comments technology
- Nvidia's H100 AI GPUs cost up to four times more than AMD's competing MI300X — AMD's chips cost $10 to $15K apiece; Nvidia's H100 has peaked beyond $40,000: Report https://www.tomshardware.com/tech-industry/artificial-intelligence/nvidias-h100-ai-gpus-cost-up-to-four-times-more-than-amds-competing-mi300x-amds-chips-cost-dollar10-to-dollar15k-apiece-nvidias-h100-has-peaked-beyond-dollar40000 147 comments hardware
Linking pages
- Nvidia's new entry-level gaming GPU is slow but very power-efficient — RTX 3050 6GB benchmarks show 20% lower performance than RTX 3050 8GB but draws 50% less power | Tom's Hardware https://www.tomshardware.com/pc-components/gpus/nvidias-new-entry-level-gaming-gpu-is-slow-but-very-power-efficient 191 comments
- Nvidia provides the first public view of its fastest AI supercomputer — Eos is powered by 4,608 H100 GPUs, tuned for generative AI | Tom's Hardware https://www.tomshardware.com/tech-industry/artificial-intelligence/nvidia-provides-the-first-public-view-of-its-fastest-ai-supercomputer-eos-is-powered-by-4608-h100-gpus-tuned-for-generative-ai 115 comments
- Silicon Valley wants to deploy AI nursebots to handle your care | Popular Science https://www.popsci.com/technology/ai-nurse-chatbots-nvidia/ 63 comments
- AMD's new cheap gaming chip is 25% faster than Intel's more expensive alternative — Ryzen 7 5700X3D is faster than Core i5-13600KF in early reviews, even on aging AM4 platform | Tom's Hardware https://www.tomshardware.com/pc-components/cpus/amds-new-cheap-gaming-chip-is-25-faster-than-intels-more-expensive-alternative-ryzen-7-5700x3d-is-faster-than-core-i5-13600kf-in-early-reviews-even-on-aging-am4-platform 58 comments
- Nvidia is the 'kingmaker,' says analyst firm Omdia — company projected to make $87 billion from data center GPUs alone in 2024 | Tom's Hardware https://www.tomshardware.com/pc-components/gpus/nvidia-is-the-kingmaker-says-analyst-firm-omdia-company-projected-to-make-dollar87-billion-from-data-center-gpu-revenue-in-2024 28 comments
- Intel’s “Gaudi 3” AI accelerator chip may give Nvidia’s H100 a run for its money | Ars Technica https://arstechnica.com/information-technology/2024/04/intels-gaudi-3-ai-accelerator-chip-may-give-nvidias-h100-a-run-for-the-money/ 26 comments
- China's top AI accelerator and CPU makers are bleeding tens of millions -- Longsoon and Cambricon losses continue despite billions in government subsidies | Tom's Hardware https://www.tomshardware.com/pc-components/cpus/chinas-top-ai-accelerator-and-cpu-designers-are-bleeding-tens-of-millions-longsoon-and-cambricon-losses-continue-despite-billions-in-government-subsidies 16 comments
- TinyBox AI accelerator now available starting at $15k, available in AMD 7900XTX and Nvidia RTX 4090 variants | Tom's Hardware https://www.tomshardware.com/pc-components/gpus/tinybox-ai-accelerator-now-available-starting-at-dollar15k-available-in-7900xtx-and-rtx-4090-variants 10 comments
- Faulty 12V2x6 connector found with RTX 4080 Super card — it could be a one-off problem, but caution is advisable | Tom's Hardware https://www.tomshardware.com/pc-components/gpus/faulty-12v2x6-connector-found-with-rtx-4080-super-card-it-could-be-a-one-off-problem-but-caution-is-advisable 9 comments
- Three Shifts Re-defining the Semiconductor Landscape https://semiconductor.substack.com/p/three-shifts-re-defining-the-semiconductor 1 comment
- Chip race: Microsoft, Meta, Google, and Nvidia battle it out for AI chip supremacy - The Verge https://www.theverge.com/2024/2/1/24058186/ai-chips-meta-microsoft-google-nvidia 0 comments
- Nvidia's first CPU+GPU chips come to AI workstation desktop PCs — Grace Hopper Superchip systems start at $41,500 | Tom's Hardware https://www.tomshardware.com/pc-components/cpus/nvidias-first-cpugpu-chips-come-to-ai-workstation-desktop-pcs 0 comments
- Chips all the way down https://press.airstreet.com/p/chips-all-the-way-down 0 comments
- FuriosaAI’s RNGD at Hot Chips 2024: Accelerating AI with a More Flexible Primitive – Chips and Cheese https://chipsandcheese.com/2024/09/11/furiosaais-rngd-at-hot-chips-2024-accelerating-ai-with-a-more-flexible-primitive/ 0 comments
Linked pages
- Amazon’s new AWS charge for using IPv4 is expected to rake in up to $1B per year — change should speed IPv6 adoption | Tom's Hardware https://www.tomshardware.com/networking/amazons-aws-new-charge-for-using-ipv4-is-expected-to-rake-in-up-to-dollar1b-per-year-change-should-speed-ipv6-adoption 331 comments
- AMD announces it has preorders for $3.5 billion of its AI GPUs; stock tumbles in after-hours trading anyway | Tom's Hardware https://www.tomshardware.com/pc-components/cpus/amds-announces-it-has-orders-for-dollar35-billion-of-its-ai-gpus-but-the-stock-tumbled-in-after-hours-trading 269 comments
- Chips aren't getting cheaper — the cost per transistor stopped dropping a decade ago at 28nm | Tom's Hardware https://www.tomshardware.com/tech-industry/manufacturing/chips-arent-getting-cheaper-the-cost-per-transistor-stopped-dropping-a-decade-ago-at-28nm 175 comments
- Nvidia Sold 900 Tons of H100 GPUs Last Quarter, Says Analyst Firm | Tom's Hardware https://www.tomshardware.com/news/nvidia-sold-900-tons-of-h100-gpus-last-quarter 118 comments
- NVIDIA Announces H100 NVL - Max Memory Server Card for Large Language Models https://www.anandtech.com/show/18780/nvidia-announces-h100-nvl-max-memory-server-card-for-large-language-models 117 comments
- AMD's integrated graphics performance shines in AI tests with extreme memory — Ryzen 7 8700G with DDR5-9000 up to 15% better | Tom's Hardware https://www.tomshardware.com/pc-components/cpus/amds-integrated-graphics-performance-shines-in-ai-tests-with-extreme-memory-ryzen-7-8700g-with-ddr5-9000-up-to-15-better 48 comments
- U.S takes the China chip war to the next level - will soon stop Chinese companies from using American clouds for AI training | Tom's Hardware https://www.tomshardware.com/tech-industry/artificial-intelligence/us-takes-the-china-chip-war-to-the-next-level 37 comments
- OpenAI CEO Sam Altman will be at Intel's next foundry event — and he's currently looking for chip partners | Tom's Hardware https://www.tomshardware.com/tech-industry/artificial-intelligence/openai-ceo-sam-altman-will-be-at-intels-next-foundry-event-and-hes-currently-looking-for-chip-partners 15 comments
- Faulty 12V2x6 connector found with RTX 4080 Super card — it could be a one-off problem, but caution is advisable | Tom's Hardware https://www.tomshardware.com/pc-components/gpus/faulty-12v2x6-connector-found-with-rtx-4080-super-card-it-could-be-a-one-off-problem-but-caution-is-advisable 9 comments