From: Pascal J. B. <pj...@in...> - 2013-04-20 10:46:07
|
Paul Khuong <pv...@pv...> writes: > Stephan Frank wrote: >> but wouldn't a library with bindings to MPFR (using CFFI) not make more >> sense? Maxima (and I suppose matlisp as well) have to implement their >> own bigfloats data type anyway whereas the GMP integration has the >> chance of being tightly and transparently integrated with the underlying >> bignum implementation of SBCL. I could only see an advantage when >> tightly coupling it with CMUCL's quad-floats. > > We may still do something useful with long-floats. That type is > currently equivalent to double-float, but there's nothing that makes the > situation necessary. IIRC, CLISP's long floats are arbitrary precision. > I'm not completely sure what that means for standard compliance, but I'm > also somewhat open to interpreting the standard loosely if the looseness > only affects long float values. Well with respect to compliance, AFAICR, there's nothing in the Hyperspec saying that the precision of floating point numbers must be constant during program execution. But even if you expect that for conforming code, the letter of the specification is definitely that the precision can change from one implementation to another, and why not, from one invocation of a program to another. So if you set the precision of long floats once for all in a session, it should be 100% guaranteed 100% conforming (as long as it's bigger than the precision of double-float). -- __Pascal Bourguignon__ http://www.informatimago.com/ A bad day in () is better than a good day in {}. |