GPUs can now use PCIe-attached memory or SSDs to boost VRAM capacity —Panmnesia's CXL IP claims double-digit nanosecond latency
GPUs can now use PCIe-attached memory or SSDs to boost VRAM capacity —Panmnesia's CXL IP claims double-digit nanosecond latency
www.tomshardware.com GPUs can now use PCIe-attached memory or SSDs to boost VRAM capacity —Panmnesia's CXL IP claims double-digit nanosecond latency
Provided that AMD, Intel, and Nvidia would like to support CXL for memory expansions.
5 comments
Upgradable vram would be huge.
just move to unified vram like console would be best.
Isn't double-digit nanosecond latency just microsecond latency?
Edit: I have learned thanks to helpful commenters that quadruple digit nanosecond is microseconds
Quadruple digit nanosecond latency is microseconds
100 nano seconds is 0.1 micro seconds, so not quite.
I imagine double digit is 99 ns or less