Hi
I didn't know about this until now.
https://software.intel.com/en-us/articles/performance-benefits-of-half-precision-floats
New intel chips support fast (vectorized) conversion between float32 and
float16. Float16 use more CPU than float32, but only half as much
cache, so there may be an advantage in some cases. Also, I guess NVIDIA
does (or will?) support half-floats.
Intel is expected to offer fuller support for half-floats in future
chips. TBD
Gerry
--
Gerald R. Harp, PhD
Director, SETI Research
SETI Institute
Mountain View, CA, USA
--
You received this message because you are subscribed to the Google Groups
"[email protected]" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
