[Rd] Power (^) 10x slower in R since version 1.7.1... What next?[Windows platform]

[Prof Brian Ripley]

On Thu, 27 Nov 2003, Philippe Grosjean wrote:

> Prof. Brian Ripley wrote:
> >You might like to test the enclosed.  Drop the file into .../src/gnuwin32
> >and edit the following line in Makefile there
> 
> >OBJS = $(SOURCES:.c=.o) e_pow.o
> >                       ^^^^^^^^
> >Then re-make.  This uses an existing solution, that from glibc (which MinGW
> >will not want to use for licence reasons).  It's very fast and seems
> >as accurate.
> 
> It is very fast! Thank you. This solves my problem!
> By the way, will this patch be incorporated in the next official release of
> R?

Not unless MinGW has not replaced pow in the meantime (which we expect to 
happen).

My Windows mailer added a binary attachment, so if any other list members 
want this, it is at

http://www.stats.ox.ac.uk/pub/RWin/e_pow_S

-- 
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 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595