On Fri, Feb 14, 2025 at 07:30:59AM -0800, Lindsay Lawrence wrote:
> On Thu, Feb 13, 2025 at 9:49 PM Lindsay Lawrence <
> lawrence.lindsayj...@gmail.com> wrote:
>
>
> > A bit of searching came up with the paper with the haskell version
> >
> > https://www.cs.ox.ac.uk/people/jeremy.gibbons/publicat
>
> How about a coroutine version?
>
> Neat! You have options over what to do with the digits with the coroutine.
And negligible difference in performance, if it matters, either. I didn't
expect that.
/Lindsay
: (bench (out "pi-digits.1.txt" (makePi 1)))
3.150 sec
-> NIL
: (bench (out "pi-dig
On Fri, Feb 14, 2025 at 05:21:46PM +0100, Alexander Burger wrote:
> Wow, that's cool! I'll measure it now :)
The old algorithm needed for 10 digits 8369 sec (02:19 h), the new
one (I use the coroutine version) just 726 sec. That's about 12 times as
fast!
☺/ A!ex
--
UNSUBSCRIBE: mailto:picol
___
Heiko Henrich
Kirchenmusik, Jazz und Feldenkrais
Britzer Straße 58
12109 Berlin
01522 8776573
heiko.henr...@gmail.com
On Thu, Feb 13, 2025 at 9:49 PM Lindsay Lawrence <
lawrence.lindsayj...@gmail.com> wrote:
> A bit of searching came up with the paper with the haskell version
>
> https://www.cs.ox.ac.uk/people/jeremy.gibbons/publications/spigot.pdf
>
> Interestingly enough there is another version of PI in there
On Fri, Feb 14, 2025 at 8:58 AM Alexander Burger
wrote:
> The old algorithm needed for 10 digits 8369 sec (02:19 h), the new
> one (I use the coroutine version) just 726 sec. That's about 12 times as
> fast!
>
👍 👍 :)
/Lindsay
