[R] Any interest in ATLAS-enabled R-1.4.0 for MSWin?

Prof Brian D Ripley ripley at stats.ox.ac.uk
Thu Dec 20 21:35:55 CET 2001

On Thu, 20 Dec 2001, Liaw, Andy wrote:

> Hi all,
> I've compiled R-1.4.0 on my NT box with link against ATLAS.  It has passed
> all the Rcmd CHECK that I could run.  If there's sufficient interest, I can
> make the SetupR.exe available (on CRAN?).  Just drop me a note if you're
> interested.  Note the following:
> o  I can only connect thru 56K dial-up until Jan. 2, 2002, so unless there's
> overwhelming demand, I won't attempt the upload until then.
> o  The ATLAS is optimized for the Pentium III with SSE1 architecture, so you
> need at least that (e.g., P4 probably works, but not PII).
> o  The pre-compiled packages made available by Prof. Ripley may not work (if
> they contain calls to BLAS).  Thus update.packages() may not work for such
> packages.  I can also provide ATLAS-enabled packages, if there's sufficient
> interest.

There are only 5 exceptions at present (KernSmooth, bqtl, fracdiff, gss,
quadprog and quantreg) that need rebuilding.

This should be much easier come R 1.5.0, when all that needs to be done is
to replace a single DLL, and then we can make various versions of that DLL
available.  For those who are interested, it might be better to use the
current R-devel, which is already set up for this.

> Now, question:  Is it possble (perhaps by adding options to
> update.packages()) to get update.packages() to download the source and
> compile, like other platforms, on Windoze?

No, it's not.  People have enough problems doing it manually!

Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

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

More information about the R-help mailing list