On 16/06/2010 09:39, J.Fine wrote:
> From: Joseph Wright [[log in to unmask]]
>
> In terms of input range, I've gone for 999999999.999999999 to
> +999999999.999999999. This should be sufficient for a wide range of
> purposes (for example, if you convert a dimen into a floating point
> number in units pt then you are safe for any dimen TeX can handle).
>
> ====
>
> Please would you clarify for me, Joseph. Does this package provide floating point numbers, as described in:
> http://en.wikipedia.org/wiki/Floating_point
>
>> From what you've written above it seems that your implementing fixed point.
> http://en.wikipedia.org/wiki/Fixedpoint_arithmetic
> namely "a fixed number of digits after (and sometimes also before) the radix point".
Currently the number of dp is fixed, so formally "fixed point
arithmetic", but following Peter's suggestion I will be adding the
facility to deal with floating point numbers (range yet to be decided).
Quite how to deal with the overlap at the low end I'm not yet quite
sure. At least for division it makes sense to convert small numbers
(e.g. 0.0001) into their exponent form. I'll probably take the approach
a pocket calculator does and convert 0.001 downward into the exponent
form, for output for add/sub/mul (etc.) and at the input stage for div.

Joseph Wright
