Infinite-precision arithmetic

[Dar Scott]

On May 24, 2005, at 3:55 AM, Cubist at aol.com wrote:

>    Naturally, there's also the option of splitting N1 and N2 up into
> 7-character chunks, doing all the multiplies for the relevant 
> chunk-pairs, and
> combining the results of said multiplies. That looks like it'd be a 
> pain to code, so
> I am going to try the other algorithm first. Depending on how fast it 
> is, I
> may or may not want to give "7-char chunks" a try as well.

I have experimented with this and like it better than the ASCII 
approach.  However, if speed is not an issue, then the ASCII approach 
may be easier.

Folks have given a lot of thought to multiplication and division, so it 
might be good to stand on their shoulders.

You might want to consider DC and shell, if speed is not important.

When I get back into this, I will likely go with Victor's suggestion 
and go with a well-established C library (or build my own) and make an 
external.  I want interval arithmetic, too, so I might have to build my 
own.

I've considered making a general number that would encompass all that 
but I'm not sure what to do with very small numbers.  Floating point 
would handle those, so I might have to go that way:  indefinite 
precision decimal floating point interval with small vectors.

Dar

-- 
**********************************************
     DSC (Dar Scott Consulting & Dar's Lab)
     http://www.swcp.com/dsc/
     Programming and software
**********************************************