Re: Speed of lsim

[John W. Eaton]

On 29-Jul-2003, Geraint Paul Bevan <address@hidden> wrote:

| I've run the same test a couple of times on the same machine with Octave
| versions 2.1.35 and 2.1.50. The newer version of Octave is much faster,
| which seems to agree with the previous results.
| 
| Best advice is probably to upgrade to version 2.1.50.

Yes, that is probably good advice.  Probably the difference for lsim
is the following change.

jwe



2002-11-12  John W. Eaton  <address@hidden>

        * control/base/lsim.m: Use approximate test for step size change.

Index: lsim.m
===================================================================
RCS file: /usr/local/cvsroot/octave/scripts/control/base/lsim.m,v
retrieving revision 1.4
retrieving revision 1.5
diff -u -r1.4 -r1.5
--- lsim.m      4 Nov 2002 22:03:05 -0000       1.4
+++ lsim.m      13 Nov 2002 00:45:23 -0000      1.5
@@ -73,10 +73,13 @@
   t(2)-t(1);
   u=u';
   n = max(size(t));
+
   for ii = 1:(n-1)
 
     ## check if step size changed
-    if (t(ii+1) - t(ii) != Ts)
+    ## XXX FIXME XXX -- this is probably not the best test, but it is
+    ## better than a test for exact equality.
+    if (abs (t(ii+1) - t(ii) - Ts) > 10 * eps)
       Ts = t(ii+1) - t(ii);
       ## [F,G] = c2d(a,b,Ts);
       dsys = c2d(sys, Ts);



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------