I'm wondering if requiring double precision internal arithmetic in the
standard is not too strong a condition.
In particular, this prevents implementors from optimizing for fixed point -
an important optimization for our
DSP.
Am I misinterpreting the spec here ?
Craig Hansen-Sturm
Philips Semiconductors
This archive was generated by hypermail 2b29 : Wed May 10 2000 - 12:14:10 EDT