Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.soft-sys.math.mathematica > #10673

Re: needed.

From awnl <awnl@gmx-topmail.de>
Newsgroups comp.soft-sys.math.mathematica
Subject Re: needed.
Date 2012-06-28 08:00 +0000
Organization Steven M. Christensen and Associates, Inc and MathTensor, Inc.
Message-ID <jsh2vp$sb6$1@smc.vnet.net> (permalink)
References <23314523.14289.1340701923445.JavaMail.root@m06> <jsefcl$i9u$1@smc.vnet.net>

Show all headers | View raw


Hi,

> Plot[Exp[8*(10 - Abs[10 - 20000*x])]*(1 -
>      Erf[(8*(10 - Abs[10 - 20000*x]))^(1/2)]), {x, 0, 1*10^(-3)},
>   PlotRange ->  Full,
>   WorkingPrecision ->  6]
>
> And then maybe someone can explain why MachinePrecision fails but a
> WorkingPrecision as low as 6 works.

I think that's because Mathematica will use "arbitrary precision 
numbers" (software emulated) instead of "machine precision numbers" 
whenever WorkingPrecision is set to something different than 
MachinePrecision, most probably using completely different algorithms 
for anything related to tracking precision.

It's probably even more surprising if you compare 
WorkingPrecision->MachinePrecision with WorkingPrecision->$MachinePrecision

hth,

albert


Back to comp.soft-sys.math.mathematica | Previous | Next — Previous in thread | Find similar | Unroll thread


Thread

Re: needed. "djmpark" <djmpark@comcast.net> - 2012-06-27 08:14 +0000
  Re: needed. awnl <awnl@gmx-topmail.de> - 2012-06-28 08:00 +0000

csiph-web