Re: Repeated digits in Pi -- the Feynman point

On Thu, 03 Jul 2025 15:30:02 +0300, Phil Carmody wrote:

By hand, I think I'd favour the spigot method. I know Matt Parker
(/Stand-Up Maths/ on youtube) likes to organise the manual calculation
of pi on "pi day", and I'm a little disappointed he's never tried
spigot. I think its Big-Oh is superior to any of the algorithms that
effectively do arbitrary precision arithmetic, as you never need to deal
with numbers much bigger than a hundred.

Yes, I've just got to that in Chapter 6 of pi Unleashed. The book includes
a faster variant that calculates 1000pi and includes C source code. It
looks a little opaque, at first reading, but I'm sure I can get something
to work. I'll give it a go shortly.

I've implemented a solution based on the Leibniz series and a variation of
that which convergences somewhat faster.

Pi Unleashed is an interesting read. It starts of on a bit of a ramble,
but includes lots of fascinating details. The array of formula to
calculate pi is quite something.

--
David Entwistle