MathGroup Archive 2006

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Version 5.2 Pi computation speed is simply amazing.

  • To: mathgroup at smc.vnet.net
  • Subject: [mg71568] Re: Version 5.2 Pi computation speed is simply amazing.
  • From: snoofly <snoofly at snoofly.com>
  • Date: Thu, 23 Nov 2006 05:41:39 -0500 (EST)
  • References: <ejur4h$i7r$1@smc.vnet.net>

Zeno wrote:
> Version 5.2 computes Pi to a million places more than 11 times faster
> than version 4.2 does on my same machine. (Computed in 37.32 seconds on
> a G4 933 Mhz OS X machine, actual time to show full result on screen.)
> This speed is simply amazing, considering it is almost as fast as the
> fastest AGM-Borwein programs to compute pi. However, those slightly
> faster AGM-Borwein programs require the number of digits computed to be
> a power of 2, where in Mathematica one can stipulate a completely
> arbitray number of digits for output.
> This performance in version 5.2 is nothing short of amazing. Great work
> there. :)
> 
perhaps mathematica is using the borwein algorithms and simply 
truncating the result of the next highest power of two to that inputted 
value
do timings in the locale of a power of two seem smooth?


  • Prev by Date: Re: general form of a n-derivative
  • Next by Date: Re: 1`2 == 1*^-10
  • Previous by thread: Re: Version 5.2 Pi computation speed is simply amazing.
  • Next by thread: number not fitting in