Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Balance

On Mon, 17 Nov 2008, Paul Zimmermann wrote:


It has been made abundantly clear to the P1788 officers by IEEE officials
that we _cannot_ mandate hardware implementation in an IEEE standard.

If we attempt to do this, then our standard _will_ be rejected at the
review stage, and P1788 will have achieved _nothing_.


of course, this does not prevent a hardware implementation of P1788, especially
if it is easy to implement in hardware.


Paul,

Quite right!
If we do our job right on P1788, then it should be a simple task to derive hardware for the standard. Indeed having a parallel specification of the required hardware is one of the things that I'd like to do as the standard emerges.
Speaking with my chip designer's hat on, I'd welcome the opportunity to provide hardware support for interval arithmetic, as the alternative is massive multi-core architectures which rely on the compiler detecting parallelism to work effectively. But that's just me... 

Cordially,

David Lester.