MathGroup Archive 2006

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: LessSame

  • To: mathgroup at smc.vnet.net
  • Subject: [mg70307] Re: LessSame
  • From: "Chris Chiasson" <chris at chiasson.name>
  • Date: Thu, 12 Oct 2006 05:37:38 -0400 (EDT)
  • References: <acbec1a40610111851u633c0d25vf611318569a3d1f8@mail.gmail.com>

So, setting $EqualTolerance to zero did solve my problem. However, I
have some questions.

What exactly is this tolerance?

Where they say there can be x number of digits difference in the
binary representation of the number, I don't really know what they
mean.

When I want to know if two numbers are "the same" to a certain
(additive) accuracy and precision, I use a comparison like this:

Abs[num2-num1]<10^(-acc)+Abs[num1]*10^-(prec)

How does that relate to the above "tolerance"?

On 10/11/06, Chris Chiasson <chris at chiasson.name> wrote:
> I think
>
> -0.9999999999999919`<=-1.`
>
> giving True is toasting a numerical algorithm I'm trying to write.
>
> What I need is LessSame or something
>
> What's the best way to proceed?
>
> Block $EqualTolerance?
>
>
> --
> http://chris.chiasson.name/
>


-- 
http://chris.chiasson.name/


  • Prev by Date: Re: recursive blues :(
  • Next by Date: RelaTime3D
  • Previous by thread: Re: sample programs
  • Next by thread: RelaTime3D