Status of floating point arithmetic in Java?

Hello,

Several years ago William Kahan, well-known in numerical analysis,
criticized Java's rules for floating point operations.
e.g. http://www.cs.berkeley.edu/~wkahan/JAVAhurt.pdf

What is the status of Java's (as expressed in Java 1.5 or 1.6)
rules for floating point operations? Have the rules been adjusted
since the original spec to address Kahan's remarks,
or for any other reason?

Thanks a lot for any light you can shed on this question.

Robert Dodier

.

Relevant Pages

  • Re: Rounding of the double
    ... actually, it doesn't store it in hex, it stores it in binary, with an implied high-order ... Floating point always seems to come as ... In my numerical analysis course, the professor spent 6 weeks telling us how to do ... The theory of numerical analysis assumes essentially infinite precision. ...
    (microsoft.public.vc.mfc)
  • Re: Conformance
    ... Using higher-precision floating point doesn't eliminate this ... numerical analysis expert figured out that this would only increase N by ... Sometimes the results from the automatic features are okay, ... Worse, because catastrophic failure ...
    (comp.std.c)
  • Re: Conformance
    ... Using higher-precision floating point doesn't eliminate this ... numerical analysis expert figured out that this would only increase N by ... Sometimes the results from the automatic features are okay, ... Worse, because catastrophic failure ...
    (comp.std.c)
  • Re: Floating-point arithmetic in CL
    ... > numerical analysis was in there, ... there was no required numerical analysis class. ... One of the odder things on the GRE computer science subject ... details of floating point machine representations. ...
    (comp.lang.lisp)
  • Re: Rounding v Truncation, was: Re: Platform Support vs.
    ... > It's interesting that you invoke a Forth thread in a floating point ... just did a Goggle search for ROUNDING NEAREST ... there's a whole field of Num Anal (Numerical Analysis :-) to ...
    (comp.os.vms)