See also the Wikipedia entry
http://en.wikipedia.org/wiki/IEEE_754#Double-precision_64_bit
----- Original Message -----
From: Roger Hui <[EMAIL PROTECTED]>
Date: Tuesday, April 10, 2007 8:53 am
Subject: Re: [Jprogramming] limit constants
> t=: ('3fe',13$'f'),('3ff',13$'0'),:'3ff',(12$'0'),'1'
> t
> 3fefffffffffffff
> 3ff0000000000000
> 3ff0000000000001
> ":!.18 cons"1 t
> 0.99999999999999989 1 1.0000000000000002
> ":!.18 ] 1 - cons"1 t
> 1.1102230246251565e_16 0 _2.2204460492503131e_16
>
>
>
> ----- Original Message -----
> From: Roger Hui <[EMAIL PROTECTED]>
> Date: Tuesday, April 10, 2007 8:31 am
> Subject: Re: [Jprogramming] limit constants
>
> > I usually solve such problems by working with the
> > hex representation. Thus:
> >
> > unhex=: 3!:2
> > hex =: 2&(3!:3)
> > cons =: unhex @ ((}: hex 0.5)&,)
> >
> > ":!.18 cons (15$'0'),'1' NB. just larger than 0
> > 4.9406564584124654e_324
> > ":!.18 cons '3fe',13$'f' NB. just smaller than 1
> > 0.99999999999999989
> > ":!.18 cons '7fe',13$'f' NB. largest finite positive
> > 1.7976931348623157e308
> >
> > Regarding the middle one, the following expressions
> > show that the answer is correct:
> >
> > ":!.18 ] 1 - cons '3fe',13$'f'
> > 1.1102230246251565e_16
> > ":!.18 ] 1 - cons '3ff',13$'0'
> > 0
> >
> >
> >
> > ----- Original Message -----
> > From: bill lam <[EMAIL PROTECTED]>
> > Date: Tuesday, April 10, 2007 3:05 am
> > Subject: [Jprogramming] limit constants
> >
> > > Hi all,
> > >
> > > With respect to J's IEEE double, how to get the numbers that
> > > a. just larger than 0 (smallest positive number)
> > > b. just smaller than 1
> > > c. largest positive number
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
