ByteBrief
We're a portrait publication through and through. Turn your phone back and your briefing picks up right where you left it.
(We tried widescreen once. It wasn't us.)
A GPU global load traverses L1, TLB, crossbar, L2, and DRAM. On an RTX 4090, L1 hits take 15.4 ns, TLB misses 4.4 ns, and L2 hits 127 ns. The path involves 32-byte sectors, 128-byte lines, and 36 L2 slices.
Tap to vote and see what everyone thinks.
Summary by ByteBrief