[Octave-bug-tracker] [bug #34334] symrcm order is not reversed

[Jordi Gutiérrez Hermoso]

URL:
  <http://savannah.gnu.org/bugs/?34334>

                 Summary: symrcm order is not reversed
                 Project: GNU Octave
            Submitted by: None
            Submitted on: Mon 19 Sep 2011 10:13:07 PM GMT
                Category: None
                Priority: 5 - Normal
                  Status: None
                 Privacy: Public
             Assigned to: None
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any

    _______________________________________________________

Details:

the sparse matrix function symrcm appears to produce results in non-reversed
Cuthill-McKee ordering rather than reverse Cuthill-McKee (compare to MATLAB
symrcm with spy). This makes a big difference for FEM solution of
Navier-Stokes with Taylor-Hood elements via gmres (Octave 3.4.2) --
non-reversed ordering produces poor matrix conditioning, reverse ordering
works properly (verified in both MATLAB and Octave). Direct solutions
(mldivide) are not impacted however. Please update symrcm (possibly with
fliplr or flipud of the result).




    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?34334>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/