1. Re: Peculiar IEEE Bug...
"Carl R. White" <C.R.White at SCM.BRAD.AC.UK> wrote:
>QB can handle double length floats. My guess is that is's internally
>converting to these to avoid the 1.0..034 inaccuracy that creeps in with
>the shorter IEEE format Euphoria uses.
I don't think this is the case. I *created* the number in QB, true, but I
read it into Euphoria with float64_to_atom(), and Euphoria printed it out as
exactly -1.000000000000000e-308. So I don't think it's a problem with the
shorter IEEE format, else Euphoria wouldn't be able to import the value with
such 100% accuracy. It's only when the value is *created* in Euphoria that
the -1.0..034 problem pops up. Euphoria *can* represent -1e-308 more
accurately, but it doesn't.
So I still say it's an interpreter/Watcom C issue.
Gabriel Boehme
