[R] Fwd: probably "round()" bug --- in Guido's Win' R 0.62.3

Guido Masarotto guido@sirio.stat.unipd.it
Thu, 15 Oct 1998 19:35:01 +0100


On Thu, Oct 15, 1998 at 03:59:03PM +0200, Martin Maechler wrote:
> It must be a "architecture dependent" (maybe Windows only) bug, triggered
> by bogous  libraries / optimizations, or such things.
> 
  That's true. I have just checked and compiling src/nmath/fround.c 
  without any optimization seems to give the right `round' under Win32
  (at least on a Win3.1 system).
  I will prepare a bug fixed version of rw0623 in few days
  (this bug, the Alt-Gr keyboard bug,....).
  A question: is it sensible that I prepare the binary distribution
  using R-Release (in spite of R-0.62.3)? I would like to have
  a version with at least the (0+0i)^2 and the seq(..,..,by=) bugs
  fixed. And the easiest way it is to use the Release(ed) source.
  g.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel 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-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._