Re: Rate the following Cryptolibraries from fastest to slowest

From: Phil Carmody (thefatphil_demunge_at_yahoo.co.uk)
Date: 07/31/03 

Date: Thu, 31 Jul 2003 11:43:49 +0300

On Thu, 31 Jul 2003 10:01:43 +0200, George Joseph wrote:

> Please rate the following libraries from fastest to slowest (the pure
> C-implemented version - not assembler)?
> (a) MIRACL
> (b) LibTomMath
> (c) GMP
> (d) LIP
>
> If there are any others which you think are faster, please inform of them
> and where I can obtain them.

They probably won't be faster than GMP, but you're missing bn, Piologie,
and NTL, for a start. If GMP does what you want natively, then I'd stake
my that on GMP being faster than anything else.

Why don't you just download them, compile them, and speed test them yourself?

Phil

Relevant Pages

  • Re: Tuning Karatsuba in LibTomMath
    ... Certainly GMP ... > My comparison is based on the fact that LibTomMath uses very similar ... The cutover to FFT method was huge, thousands of digits. ...
    (sci.crypt)
  • Re: C90 long long
    ... super-fast bignum library would be well advised to use GMP or Miracl. ... 64 bit math like Chess programming (uses a 64 bit integer to hold the ...
    (comp.lang.c)
  • Re: C90 long long
    ... super-fast bignum library would be well advised to use GMP or Miracl. ... Financial calculations and reporting by governments and ...
    (comp.lang.c)
  • Re: Tuning Karatsuba in LibTomMath
    ... > is performance oriented code and the algorithms are often quite complex. ... Certainly GMP ... My comparison is based on the fact that LibTomMath uses very similar ... algorithms use the LibTomMath API and I can say it worked out well :-) ...
    (sci.crypt)
  • Re: Places to purchase PowerBASIC?
    ... into GMP or Miracl. ... But there's no intrinsic reason why it *couldn't* have such a ... Google users: ...
    (comp.programming)