Bug#499565: basic matrix computations give incorrect results

Frederic Lehobey Frederic at Lehobey.net
Sat Sep 20 09:40:15 UTC 2008


Hi Sylvestre,

Sylvestre Ledru <sylvestre.ledru at inria.fr> (2008-09-20 00:39:49) :
> tags 499565 + pending
> thanks

> > u should be the 3x3 identity matrix, and so u(2,2) should be 1 not 2.
> I confirm that Scilab 4.X has some issues with 64 bits CPU.

  This is huge miscalculations. Do you consider Scilab 4.X usable on
64 bits architectures at all or should it be disabled there?

  Should this be considered serious or RC on 64 architectures?
(renders Scilab useless there, doesn't it?)

> The version 5.0 fixes most of these issues (and I confirm that this bug
> does not occurs with a Scilab 5).

  What about fixing this RC bug with the 5.0 new version
(congratulations for the release by the way)? (I know, not realistic
and such, but why not let try...)    :)

Best regards,
Frédéric





More information about the debian-science-maintainers mailing list