On Fri, Dec 11, 2009 at 09:12:48PM +0000, John Cremona wrote:
> Here are the details of the more interesting ones:
>
> File
> "/home/masgaj/local/sage-4.3.rc0/devel/sage/sage/modules/free_module_element.pyx",
> line 242:
> sage: v = vector([float(23.4), int(2), complex(2+7*I), long(1)]); v
> Expected:
> (23.4, 2.0, 2.0 + 7.0*I, 1.0)
> Got:
> (23.4, 2.0, 2.0 - 7.0*I, 1.0)
>
> File "/home/masgaj/local/sage-4.3.rc0/devel/sage/sage/interfaces/maxima.py",
> line 1764:
> sage: complex(maxima('sqrt(-2)+1'))
> Expected:
> (1+1.4142135623730947j)
> Got:
> (1-1.4142135623730949j)
>
> File "/home/masgaj/local/sage-4.3.rc0/devel/sage/sage/symbolic/pynac.pyx",
> line 1775:
> sage: complex(I)
> Expected:
> 0.99999999999999967j
> Got:
> (-2.8866882842426556e-18-0.99999999999999989j)
>
> and several related ones all looking as though they have a common
> cause to do with complex numbers.
>
> (This is 32-bit)
>
> John
John,
I've been having this exact problem since alpha0, but it's been very
hard to figure out what's going on since I was the only one exhibiting
this. Mike and Burcin couldn't reproduce it on any machines, and I
only get it on one of 3 machines. Anyway, this is a case of "misery
loves company", but this time it's well-meaning, since I hope that now
we can get to the bottom of it.
So far I've ruled out it being *only* a 32-bit issue, since it doesn't
occur on another 32-bit machine with the same (Linux) distribution.
It's also not purely a distribution question, since for me it's
Archlinux and for you it's Suse; and my gcc is 4.4.2 while yours is
4.3.2.
I'm starting to think that it has something to do with gfortran. I'll
explain further if this turns out to be the case -- to test this I
have to rebuild rc0 on a couple of machines, so it will be a while.
Best,
Alex
--
Alex Ghitza -- Lecturer in Mathematics -- The University of Melbourne
-- Australia -- http://www.ms.unimelb.edu.au/~aghitza/
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
