93379723 / 29723689 is pretty good. Can anyone get better? :) Ryan On 29 May 2015 at 16:35, RussellMc wrote: > On 29 May 2015 at 15:05, Harold Hallikainen wrot= e: > >> That formula for pi is in the manual that came with my HP-35 calculator = in >> 1972. >> > > My brain told me that it came via HP (which I thought about mentioning) > but I did not remember beyond that. A friend bought an HP35 in 1973 - so > that may well have been the source. > > I imagine that formulae that come with modern calculators will not last > anywhere near as long as those which were provided 40+ years ago. > > The traditional 22/7 returns > > 3.142 857 142 857 =3D 22/7 > 3.141 592 653 979 =3D Pi > 3.141 592 920 353 =3D 355/113 > > 22/7 Pi x 1.004024... > > ___________________________________ > > I made a spreadsheet with 65,530 calculation lines. > Very simplistically it starts with 22/7 and advances numerator if a/b< Pi > and denominator if > Pi. Yes, I too can think of better ways to do this. > It found 355/113 (as you'd hope) and all multiples. > Interestingly, very little else got very close. > > 1 / (Pi - x) > > 2.68E-7 355/113 > 8.27E-7 49722/15827 > > Other peaks are down around 20% as good but improving linearly with size > of numbers. Something better than 355/113 looks due to pop out around > 248610 / 79135 (which is a multiple of 355/113) > > ie 355/113 is astoundingly good and a very lucky find and fact OR an > artefact of something that is (to me) inobvious. > > >> We haven't yet dealt with >> > > > >> >> > Watts ~=3D Torque in kg.m x RPM >> > :-) > > > > >> Russell >> > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .