Re: simplify sinh/cosh to tanh
From: Kerem (uguz_at_ufl.edu)
Date: 02/24/05
- Next message: Alec Mihailovs: "Re: MuPAD bugs: series can return 6 terms only"
- Previous message: Stefan Wehmeier: "Re: MuPAD bugs: series can return 6 terms only"
- In reply to: Robert Israel: "Re: simplify sinh/cosh to tanh"
- Messages sorted by: [ date ] [ thread ]
Date: 24 Feb 2005 06:00:13 -0800
evalhf(Digits) returns 14 on my computer also. Thank you Robert for
your explanation.
Kerem
Robert Israel wrote:
> In article <1109186954.848304.239570@f14g2000cwb.googlegroups.com>,
> Kerem <uguz@ufl.edu> wrote:
>
> >I hope I am not bothering too much. But, I want to make this subject
> >clear. If the problem is related to hardware floating point, how am
I
> >able to plot (sinh(x)/cosh(x) x=100..1e6) when I increase the Digits
to
> >15? The help file of Maple states that Digits is related to control
the
> >number of digits that Maple uses when calculating with software
> >floating-point numbers. So by changing Digits, I am in fact changing
> >the software floating point. I am confused. How is then the problem
> >related to hardware floating point?
>
> The exact value may be platform-dependent, but evalhf(Digits) returns
14
> for me under Windows. Since you set Digits = 15 which is more than
that,
> plot decides that hardware floats will not provide sufficient
accuracy,
> and therefore does the calculation using software floats.
>
> Robert Israel israel@math.ubc.ca
> Department of Mathematics http://www.math.ubc.ca/~israel
> University of British Columbia Vancouver, BC, Canada
- Next message: Alec Mihailovs: "Re: MuPAD bugs: series can return 6 terms only"
- Previous message: Stefan Wehmeier: "Re: MuPAD bugs: series can return 6 terms only"
- In reply to: Robert Israel: "Re: simplify sinh/cosh to tanh"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
