๐Ÿ› ๏ธ Steven Gong

Search

SearchSearch

Nov 22, 2026 ยท 1 min read ยท 6,955 views ยท 203 in the last 30 days

Semianalysis

Has some really good articles to learn about GPUs

  • https://newsletter.semianalysis.com/p/nvidia-tensor-core-evolution-from-volta-to-blackwell
  • https://newsletter.semianalysis.com/p/dissecting-nvidia-blackwell-tensor

Some notes from tensor core article:

  • Issuing an instruction takes ~30pj, whereas the floating point op only takes 1.5pj
    • so thereโ€™s a lot of overhead in just issuing instructions, maybe we can combine that into a single instruction at the hardware level

Graph View

Backlinks

  • No backlinks found

Created with Quartz, ยฉ 2026

  • Blog
  • LinkedIn
  • Twitter
  • GitHub