Re: Is it done?
"Ben Laurie" <[email protected]>
| Newsgroups | gmane.comp.lang.e.general |
|---|---|
| Message-ID | <[email protected]> |
On Jan 1, 2008 1:14 PM, Ben Laurie <[email protected]> wrote: > On Jan 1, 2008 7:34 AM, Mark Miller <[email protected]> wrote: > > On Dec 31, 2007 9:54 PM, Kevin Reid <[email protected]> wrote: > > > > (mismatch (file #P"/Users/erights/svn-other/e-on-cl/tests/ > > > > numbers.updoc") > > > > (source "def x := 2.0 ** -0.5") > > > > (expects ("value" "0.7071067811865476")) > > > > (instead ("value" "0.7071067811865475")) (opt-backtrace > > > > nil)) > > > > > > This is the floating-point problem I mentioned. I believe that > > > 0.7071067811865475 is actually the correct answer; please verify this. > > > > E-on-Java does give 0.7071067811865475. For this case, E-on-Java > > simply delegates to Java's StrictMath.pow(), which is supposed to use > > IEEE double precision with round-to-even. I'm not sure what else to do > > to verify this. > > $ bc > bc 1.06 > Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc. > This is free software with ABSOLUTELY NO WARRANTY. > For details type `warranty'. > scale=100 > 1/sqrt(2) > .7071067811865475244008443621048490392848359376884740365883398689953\ > 662392310535194251937671638207863 Alternatively, this shows which is righter: 0.7071067811865476*0.7071067811865476-.5 .00000000000000010691335120706576 .5-0.7071067811865475*0.7071067811865475 .00000000000000003450800503024375 > > > > Since this kind of thing may be sensitive to platform issues at > > several levels of abstraction, for the record, my "rune --version" > > prints > > > > E version: 0.9.1c at /Users/erights/ehome/ > > on Java version: 1.5.0_07 at > > /System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Home > > on OS: Mac OS X version: 10.4.11 on i386 (mac/x86) > > for erights in US > > > > > > > > -- > > Text by me above is hereby placed in the public domain > > > > Cheers, > > --MarkM > > > > _______________________________________________ > > e-lang mailing list > > [email protected] > > http://www.eros-os.org/mailman/listinfo/e-lang > > >