[Numpy-discussion] Re: [OT] A weekend floating point/compiler question

[Robert Kern]

Fernando Perez wrote:

> Any ideas/comments?  Shouldn't the result be independent of the
> intermediate double var?  It is for icc, can this be considered a gcc bug?

It seems like it might be processor-specific. On my G4 Powerbook (g77 3.4.4, gcc
3.3) and AMD64 Linux desktop (g77 3.4.5, gcc 4.0.2), both programs give the
expected results. Specifically, the Intel 80-bit FPU thingy is probably a factor.

It might be worth filing a bug report against gcc. If nothing else, you might
get a better explanation of what's going on.

-- 
Robert Kern
robert.kern at gmail.com

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco