Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
latchkey
on Oct 7, 2023
|
parent
|
context
|
favorite
| on:
AMD may get across the CUDA moat
https://en.wikipedia.org/wiki/Double-precision_floating-poin...
NVIDIA fp32 (H100) has 2x more TFLOPS than AMD's fp32 (MI250) and AI doesn't need fp64 precision.
incognition
on Oct 8, 2023
[–]
Lol it was meant as I wouldn't be caught dead using fp64
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
NVIDIA fp32 (H100) has 2x more TFLOPS than AMD's fp32 (MI250) and AI doesn't need fp64 precision.