Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Bug or hard-to-understand feature of evalf[n] ?
10 views
Skip to first unread message
jfh
May 15, 2012, 9:53:45 PM5/15/12
to
Maple 16 gives
> Digits;
10
> evalf[20](1/9);
0.11111111111111111111
> evalf[20](1.0/9.0);
0.1111111111
Is it supposed to give only 10 digits in the second case?
-- John Harper
> Digits;
10
> evalf[20](1/9);
0.11111111111111111111
> evalf[20](1.0/9.0);
0.1111111111
Is it supposed to give only 10 digits in the second case?
-- John Harper
Joe Riel
May 16, 2012, 1:06:57 AM5/16/12
to
the floating-point fraction is done during the simplification
pass, before evalf is called. So it uses the prevailing Digits,
which is 10.
At least that's what I believe is happening.
Joe Riel
0 new messages