Hi,
I built octave 3.2.2 on my x86_64 desktop and x86 notebook a few weeks
ago.
They passed all the tests.
But I've just noticed some weird complex matrix multiplication results
on both
the machines:
octave:48> [1 0]*[i 0]'
ans = 0 - 1i %% correct
octave:49> [1 i]*[i 0]'
ans = 0 + 1i %% incorrect, 0+1i=[i
0]*[1 i]'
The problem is that for nx1 row vectors a and b both with non-zero
imaginary parts,
a*b' turns out to be b*a', which is the conjugate of the correct result.
If either a or b
are real, the result will be correct.
I also checked out Octave 3.0.3, it didn't have this problem.
So, is it a run time configuration problem, a build configuration
problem, or a bug?
Regards,
Shuhua Zhang, Oct. 16, 2009
------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev
