[Rd] Why not in Debian distro?

Prof Brian D Ripley ripley@stats.ox.ac.uk
Wed, 8 Aug 2001 08:01:07 +0100 (BST)

On Tue, 7 Aug 2001, Dirk Eddelbuettel wrote:

> [ Sorry for chiming in late, but was off email when this thread started ]
>   "Tony" == A J Rossini <rossini@blindglobe.net> writes:
>   Peter> "Peter" == Peter Dalgaard <BSA <p.dalgaard@biostat.ku.dk>> writes:
>   Peter> later. Whatever happened to the internal Debian discussions on
>   Peter> creating a faster upgrade path for applications?
> That is not the real problem. The real problem is (soon, hopefully: was) the
> dependency on Atlas in Debian's r-base package, and the lack of Atlas libs
> for architectures beyond i386, alpha and sparc. Because of the this, the last
> pre-Atlas-as-default remained in the released distribution (aka "stable") as
> release are required to contain the same package set across architectures.

Just for information, can someone tell me how Debian (which has a `i386'
port) can possibly use ATLAS properly?  I find that there are major
performance differences between versions of ATLAS tuned on different
Pentium II chips (PII-450 vs mobile PII-366, for example) so that a large
part of the performance gain goes if they are switched.

Reason for question: I might put out an ATLAS-equipped Windows version of R
if I had any idea how to produce a `i386' version of ATLAS.


Brian D. Ripley,                  ripley@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-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