Re: Table through zero...strange
- To: mathgroup at smc.vnet.net
- Subject: [mg67037] Re: Table through zero...strange
- From: David Bailey <dave at Remove_Thisdbailey.co.uk>
- Date: Wed, 7 Jun 2006 05:10:30 -0400 (EDT)
- References: <e63mqt$kqd$1@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
leptonic wrote:
> Hello,
>
> When i type in Mathematica (version 4) Table[x, {x, -0.1, 0, 0.02}]
>
> it gives me {-0.1, -0.08, -0.06, -0.04, -0.02, -3.46945 x 10-18]}
> but for example Table[x, {x, -0.1, 0, 0.05}] will give a reasonable
> {-0.1, -0.05, 0.}.
>
> Probably a very stupid question but i can't figure out why the zero
> makes problem.
>
> Thanks for any help.
>
Floating point numbers are usually not exact - so this result is not
surprising. In fact, by default floating point numbers are stored at
machine precision using 64-bit floating point. This is a binary format,
so numbers that are not of the form a*2^n (a and n integer) can never be
exact.
The following is a more reliable way to specify such an iteration:
Table[x/100.,{x,-10,0,2}]
David Bailey
http://www.dbaileyconsultancy.co.uk