rbind and data.frame [simplified]

G?ran Brostr?m <gb at stat.umu.se> writes:

Thanks for the interest in my timing problem. I have scaled off all 
calculations in order to purify it, and it is obvious that size 
matters a lot. Also that 'matrices are faster than data frames'.

I give you the full listing here, but it is 
really the last few lines that are interesting (= slow):

...

  for (cur.row in (1:no.of.outrows)){
    dat.out[cur.row, ] <- fixed.rec
      ## cbind(fixed.rec, com.dat[1, , drop = FALSE])
    ## cat("row = ", cur.row, "\n")
  }
  ## return(dat.out)
}
------------------------------------------------------------------------

...,

Sens moral: Avoid data frames for manipulations of this kind.
(Am I right?)

Hmm. With 1.4.0 on the immediate horizon I don't want to go into
actually trying to run this stuff, but it wouldn't be the first time
that data frame code caused slowness by ensuring that the
result of operations is a data frame every single step of the way.

What happens if you do do something like this?: Replace the
  dat.out[cur.row, ] <- fixed.rec
with 
  for (i in 1:ncol) dat.out[[i]][cur.row] <- fixed.rec[[i]]

bypassing "[<-.data.frame"