Not in Python, but one I investigated last year for calculating π on an obscure BASIC system. It didn't have it as a constant and only used degrees, so the convenient 4*ATN(1) calculation wasn't useful:In the end, though, I used the area of unit n-gon approximation A = ½n⋅sin(360°÷n), which with n = 36000 was more than good enough for the system's single-precision floating point. I used it later on in PostScript, too:No Pi Day would be complete without a mention of Ramanujan, whose brilliantly opaque π ≅ 9801÷(2206⋅√2) is good to 7 decimal places.
In other Pi day news, the Toronto Raspberry Pi meetup reconvened after a very long hiatus
Code:
DEFDBL a-zepsilon=1e-15a=1b=1/sqr(2)t=1/4x=1DO y=a a=(a+b)/2 b=sqr(b*y) t=t-x*(a-y)*(a-y) x=2*x IF a-b<epsilon THEN EXIT DO END IFLOOPPRINT (a+b)*(a+b)/(4*t)Code:
/pi 0.001 sin 2 div 360000 mul defIn other Pi day news, the Toronto Raspberry Pi meetup reconvened after a very long hiatus
Statistics: Posted by scruss — Sun Mar 17, 2024 3:41 pm