Re: Motion P1788/0024.01: Rounding Mode as Operation
That's a good point. My experience is that the speed of the actual interval operations is not a
primary concern in many interval applications. However, flushing the pipelines to
change the rounding mode can be very time consuming. In any case, I personally agree that the
standard should not mandate this be implemented in hardware, for reasons already stated by
others. Along these lines, I note that 754 does not mandate hardware implementations, and I also
think we would have a hard time selling such a requirement to the IEEE committees reviewing our
(eventual) proposed standard. On the other hand, stating that such operations exist provides
a framework for hardware implementations, should they eventually be decided to be appropriate.
Baker
On 4/18/2011 14:18, Ian McIntosh wrote:
.
.
.
I understand the desire for good Interval Arithmetic performance, but performance is up to the implementers. If IA is important to them, they will do what's practical to make it fast, balancing that decision with all their other
requirements and desirables. If it is not important, they will not. If implementers have other ways to either switch rounding modes quickly or to reduce the need to switch (eg, through smarter compilers) then they may do that instead.
--
---------------------------------------------------------------
Ralph Baker Kearfott, rbk@xxxxxxxxxxxxx (337) 482-5346 (fax)
(337) 482-5270 (work) (337) 993-1827 (home)
URL: http://interval.louisiana.edu/kearfott.html
Department of Mathematics, University of Louisiana at Lafayette
(Room 217 Maxim D. Doucet Hall, 1403 Johnston Street)
Box 4-1010, Lafayette, LA 70504-1010, USA
---------------------------------------------------------------