Re: Misc: Applications of small floating point formats.

Lawrence D'Oliveiro wrote:

On Wed, 31 Jul 2024 18:31:35 -0500, BGB wrote:
 

Binary16 is useful for graphics and audio processing.

 The common format for CG work is OpenEXR, and that allows for 32-bit and
even 64-bit floats, per pixel component. So for example R-G-B-Alpha is 4
components.
 

The 8-bit formats get a bit more niche; main use-cases mostly to save
memory.

 Heavily used in AI work.

The nicest property of fp8, as seen from a GPUs point of view, is that arbitrary operations can be seen as texture map lookups. I don't think that's how they are implemented but an 8x8->16 FMUL would only need a few very small lookup tables, probably doable even on a regular CPU with 16-element permute operations.
Terje
--
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"