On Mon, 19 Jun 2006, David Berry wrote:
> On Sat, 17 Jun 2006, Malcolm J. Currie wrote:
>
>>> occassionally think about these issues but the real problem is that IEEE
>>> arithmetic is designed such that this is all irrelevant. That's the entire
>>
>> I was tempted to respond to Brad "It's a feature."
>>
>>> The only real solution is for NDF_UNMAP to go through and convert inf, nan
>>> to BAD automatically.
>>
>> We've been here before.
>
> Yes - I tried it several years ago and got "the withering look" from RFWS
> since it seriously slowed down all KAPPA applications.
>
Could it be tunable? So MATHS could have a parameter which would control
whether this was required and would call NDF_TUNE with some value to say
that IEEE -> PRM convresion was required?
--
Tim Jenness
JAC software
http://www.jach.hawaii.edu/~timj
|