My apologies for this message - I should have done a bit more thinking 
before hacking...


(1) Yes, there _is_ a  speed gain by implementing evaluation 
"at compile time" - which may seem a bit strange, considering the 
tremendous amount of instruction-level parallelism in current processors, 
but you can expect roughly about 20% for a 10-th order polynomial with 
nonsparse coefficients.

(2) In many situations, this is perhaps not worth the effort.

(3) One may wonder whether the calculation may be re-arranged in such a 
way (separating odd and even coefficients, say) that more use can be made 
of the processor's multiple floatingpoint units. Again, this does not seem 
to work too well.



-- 
regards,               [EMAIL PROTECTED]              (o_
 Thomas Fischbacher -  http://www.cip.physik.uni-muenchen.de/~tf  //\
(lambda (n) ((lambda (p q r) (p p q r)) (lambda (g x y)           V_/_
(if (= x 0) y (g g (- x 1) (* x y)))) n 1))                  (Debian GNU)

Reply via email to