>
> On Mon, Feb 6, 2017 at 9:52 AM, Waldek Hebisch <hebi...@math.uni.wroc.pl>
> wrote:
> > No. I was writing about handling of 'float', especially
> > the lines:
> >
> > b := ((ll.3) pretend Integer)::SF
> > _*(a, EXPT(b, e)$Lisp)$Lisp pretend INF
> >
> > b was coverted to double, but a and e were kept as integers.
> > AFAICS this maximizes chance that the expression will be
> > computed exactly. With Lisp which does not realize that
> > e is an integer we will effectively get 'exp(e*log(b))'
> > and that introduces rounding.
>
> OK, now I understand you. But your example
> ((4*x*(x^4+1)^(1/2)+(x^4+2*x^2+1)*2^(1/2))/(x^4+(-2)*x^2+1))::Expression(Float)
> is now related with this "float" clause, right?
> BTW, how to create a EXPR with operator "float"?
((4*x*(x^4+1)^(1/2)+(x^4+2*x^2+1)*2^(1/2))/(x^4+(-2)*x^2+1))::Expression(Float)::InputForm
(15)
(/
(+
(* (* (float 4 0 2) x)
(^ (/ (+ (* (float 1 0 2) (^ x 4)) (float 1 0 2)) (float 1 0 2))
(/ 1 (/ (float 2 0 2) (float 1 0 2))))
)
(+
(+ (* (float 208701085205324515398 - 67 2) (^ x 4))
(* (float 208701085205324515398 - 66 2) (^ x 2)))
(float 208701085205324515398 - 67 2))
)
(+ (+ (* (float 1 0 2) (^ x 4)) (* (float - 2 0 2) (^ x 2))) (float 1 0 2)))
Type: InputForm
--
Waldek Hebisch
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to fricas-devel+unsubscr...@googlegroups.com.
To post to this group, send email to fricas-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.
