"Miller, Raul D" <[EMAIL PROTECTED]> wrote:
> It might also be worth thinking about what this kind of
> thing would mean when the left argument to . is not +/
> or when the right argument is not *

There are already many places in the J interpreter where
there are quite a few possible optimizations, but the only
ones that are implemented are ones involving frequently-used
operations. For example, there is already some special code
in -/ .* Deteriminant and +/ .* Matrix Multiply. If symmetric
matrix operations are important, +/ .* could be specifically
optimized without necessarily having to optimize the general
case of u/ .v

-- Mark D. Niemiec <[EMAIL PROTECTED]>

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to