Subject: Re: [boost] [Review Request] Multiprecision Arithmetic Library
From: Christopher Kormanyos (e_float_at_[hidden])
Date: 2012-04-09 17:50:40
See previous long e-mail.
> I experimented with FFTs.
> My program crashed for a million digit square-root calculation.
Was stack overflow, as expected.
* 10^6 digit10, naive FFT, base-10: time =3.0s
* 10^6 digit10, MPFR: time = 0.23sLot's of information here. But in short, big digits require allocator,
as in your other back-end classes.
Best regards, Chris.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk