AI Weekly Malaysia

Back to items Summaries

AMD attacks Nvidia's FP64 emulation at SC26, touts native double-precision on Instinct MI430X

ID
47395
Status
summarized
Published
17 Nov 2026, 10:30 PM
Fetched
18 Nov 2026, 1:46 AM
Provider
Tom's Hardware
Category
technology
Original URL
https://www.tomshardware.com/pc-components/gpus/amd-attacks-nvidias-fp64-emulation-at-sc26-touts-native-double-precision-on-instinct-mi430x-says-ozaki-schemes-fit-less-than-10-percent-of-hpc-workloads-and-triple-memory-use-and-emulated-fp64-is-a-poor-fit-for-many-hpc-workloads
Source URL
https://www.tomshardware.com/feeds/all

Summary

Score
3.0
Created
18 Nov 2026, 1:48 AM
Tags
Audience
developersai_ml_learners

What happened

At SC26, AMD criticized Nvidia's FP64 emulation and promoted native double-precision on its Instinct MI430X. AMD claimed Ozaki-style emulation fits less than 10% of HPC workloads, roughly triples memory use, and is a poor fit for many HPC workloads.

Why it matters

If you are sizing GPU nodes for FP64-dependent HPC or scientific code, AMD's stated under-10% fit rate and 3x memory penalty are the concrete claims to test against your own kernels; do not treat emulated FP64 as a drop-in substitute without workload-specific benchmarks.

Discussion angle

What evidence would you require before choosing native FP64 on MI430X over an emulation approach: kernel-level benchmarks, memory-bandwidth measurements, or porting cost?

Top