#9002: Noise on PPC Mac in parametric_surface.pyx
------------------------+---------------------------------------------------
Reporter: kcrisman | Owner: tbd
Type: defect | Status: needs_review
Priority: minor | Milestone: sage-4.5
Component: doctest | Keywords:
Author: | Upstream: N/A
Reviewer: | Merged:
Work_issues: |
------------------------+---------------------------------------------------
Comment(by kcrisman):
The expected value doesn't matter in that sense. It is just creating a
box which contains a certain graphic, and the last few digits are
completely irrelevant (try the pictures). E.g.,
{{{
sage: from sage.plot.plot3d.parametric_surface import MobiusStrip
sage: M = MobiusStrip(.0000000000007,.00000000000003,2)
sage: M.bounding_box()
((-7.3000000000000002e-13, -7.0049870840325469e-13,
-2.9940801852848143e-14), (7.3000000000000002e-13, 7.0049870840325469e-13,
2.9940801852848143e-14))
}}}
and you can also check by looking at the graphic that the box is correct
(though the formatting of the numbers could be better, which is a separate
issue).
Now, that doesn't mean that somewhere there isn't some evil bug lurking.
But tracking down where the discrepancy comes from in this code would mean
following a *lot* of branches, for no real purpose. Most likely it is the
fact that `MobiusStrip` has an `eval()` method that uses `math.sin` and
`math.cos` a lot, which presumably will be very slightly different on
different architectures.
David, I realize this is a decision regarding how far back to look for
numerical consistency, and I agree than for things like `gamma(2.3)` it
would be very good to ensure maximum consistency. But here the whole
point is a heuristic to get a box that encloses the figure, and
differences at that level of accuracy are irrelevant and take resources
away from other badly needed improvements to Sage. So it is unfortunate
that we won't track it down, but it is reasonable, I think.
Georg, when you review this, you might as well try the example in this
comment, to make sure that it looks right, but if it does then just do a
normal review. Unless you want to compare what you get from `eval()` with
this one, which is surely where the actual difference lies. I essentially
only have access to one box at a time, so it's very difficult for me to
compare them.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/9002#comment:3>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
