[R] Efficiency of local functions

[Peter Holzer]

Dear R-users

In order to speed up the performance of my program I tried to make a
function f2 (that is only called from function f1) local to f1. In that way I
pass one argument less to f2, a huge matrix. (This matrix isn't changed
inside f2, so - afaik - it is passed by reference and size doesn't matter.)
Still I would have expected that f1 would be a little faster, but the
opposit happened:

system.time(
rs <- rotateSegment(im, 272, 234, 100, 72, 51, 21)
            )

(where rotateSegment is f1)

gave

> + + [1] 12.13  0.05 12.31  0.00  0.00 ## f2 extern of f1
> + + [1] 12.29  0.02 12.46  0.00  0.00 ## f2 local to f1

(I tried several times and got always similar results. I am using Linux,
Red Hat 6.1)

Does the use of local functions generally slow down the performance? (Even
though just a little bit, whereas the advantage of elegancy in programming
might be considered quite big.)

Peter

____________________________________________________________

Peter Holzer                      phone: + 41 1 632 46 34
Seminar fuer Statistik, LEO C14   fax:   + 41 1 632 12 28
(Leonhardstr. 27)                  <holzer at stat.math.ethz.ch>
ETH (Federal Inst. Technology)
8092 Zurich                       http://www.stat.math.ethz.ch/~holzer/
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._