BigInt memory usage

[kennytm]

Jay Norwood <jayn at prismnet.com> wrote:
> I haven't looked at the std implementation, but I liked this guy's
> package in c++ .  He uses strings to hold the data, but you choose how
> the string packs the digits, with one of the choices being base 256 for
> the most efficient storage.    It might be a nice fit for a D port.
> 

D's big-int *is* using base-256. (actually, base-(size_t.max+1))

> http://www.hvks.com/Numerical/arbitrary_precision.html
> 
> bearophile Wrote: