[R] bug in fft

[Gabriel Fricout]

I wanted to use th fft function for some convolution purposes and it seems
to have a bug:
I did the following:
sizeX<-361
sizeY<-565
depthX<-23
depthY<-23
kernel<-array(0,c(sizeX,sizeY)+c(depthX,depthY))

kernel[1:(2*depthX+1),1:(2*depthY+1)]<-1
transKern <- fft(kernel)
K <- fft(transKern,inverse=TRUE)/(dim(transKern)[1]*dim(transKern)[2])
plot(Re(transKern[1,]))

and obtain something fully different from the nice function kernel !

Is that a known bug, and is there some alternative to fft function
NB: I used this function for kernel with depthX and depthY equal to 20 to
80, and everything worked pretty well (maybe!) except for the value 23 and
31!!!

Thanks for any help!

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._