Re: bug conc. number_codes

[Remy Haemmerle]

hi, 

It's the RIGHT values.
In fact GNU-prolog prints 16 significant digits and so doesn't round
floatting point imprecision.

Remenber that the binary codings of 1.4 and 1.56 and 0.2 have an
infinity of numbers after the point.

example: 0.2 = 1/5 = 1b/101b = 0.00110011001100110011....b
   

Rémy.


Johannes Middeke a écrit :
> 
> hi,
> 
> there seems to be a bug within number_codes/2 as follows:
> 
> > | ?- number_codes(N,"1.4").
> >
> > N = 1.3999999999999999
> >
> > yes
> > | ?- number_codes(N,"1.56").
> >
> > N = 1.5600000000000001
> >
> > yes
> 
> is there a way to get the right values?
> 
> _______________________________________________
> Users-prolog mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/users-prolog