Wow, good thing I check my spam folder every so often -- this was in it!
Thanks for the kind offer Dale but I'm not local (Edmonton, AB, Canada).
Not a math major but will give that pi algo a try :D
On Wed, Mar 24, 2021 at 5:08 PM <dalescie...@aol.com> wrote:
> Brad & all,
> If you are in the Los Angeles, Calif area I could loan you a dual chan
> O'scope (20 or 30 MHz)
>
> I live in Torrance (SW of LA proper adjacent to Redondo & Hermosa Beach):
> dalescie...@aol.com
>
>
> For you PC-8201a owners here is a short 1 line pgm to iterate to 17
> significant digits of PI, in 25 mins 25sec for 4.9MHz clock.
>
> Why use 22/7 or 355/113 =6sd, 355/113# <8sd
>
> 4 powerCONT: defDBLd,i: Time$="000000"::for J=2 to 46378 step4: i=J: d=d
> +4/(i*(i+1)*(i+2)) -4/((i+2)*(i+3)*i+4)): next:
> ?Time$" 3"chr$(27)d;(i-2)/4;:beep
>
> algorithm:
>
> pi = 3 +4/(2*3*4)-4/(4*5*6) +4/(6*7*8)-4/(8*9*10)....repeat sequence
> until reach limit of 8201a double precision math routines
>
> =16 sig digits, squeezed out 17th by cutting iteration just before it
> breaks down - result begins slow downward plunge (decreases)
> (46378-2)/4 = 11594 actual iterations Assumes 1st digit of PI is 3
>
> Result: 3.1415926535897932 (25 min, no spaces between pgm cmds) {does
> 12 sig digits in 3min}
>
> Any motivated Math majors able to design an algorithm that expands beyond
> 17sd independent of double prec assy lang limit?? :-)
>
> if there is interest, I could post more:
>
> 1..array of 17 for/next targets for each add'l digit displayed on separate
> lines in two columns taking up the whole display area.
> (displays time & expanding PI, current For/Next tgt) adds 25 sec for
> extra displays (25:50)
>
> 2. creative way to save same to a file, formatted as was displayed.
> (ordered: 1st - 8th, 2nd - 9th....) in lieu of print screen to file!!! TBD
>
> Dale... K3MNN /6 since 1967 retired TRW Space & Electronics 2006
> Redondo Beach, Calif
>
--
--
Brad Grier
