My data does not contradict Skip Cave's data, but they refute his conclusion: "The closed form solution is slow AND uses lots of space". Actually the closed form solution is fast and uses less space.
closed =: [: (,.3r4*<:) 5&^ NB. closed form verb definition closed 15x NB. test 30517578125 22888183593 6!:2'closed 5000x' NB. timing 0.0181154 6!:2'closed 5000x' NB. timing 0.0360412 6!:2'closed 5000x' NB. timing 0.0292909 6!:2'closed 5000x' NB. timing 0.0266407 6!:2'closed 5000x' NB. timing 0.0186728 closed 50 NB. using standard arithmetic 8.88178e34 6.66134e34 6!:2'closed 50' NB. timing 0.000110246 - Bo Fra:Boyko Bantchev <boyk...@gmail.com> Til: programm...@jsoftware.com >Sendt: 14:52 søndag den 23. december 2012 >Emne: Re: [Jprogramming] arithmetic sequence > >On 23 December 2012 02:59, Bo Jacoby <bojac...@yahoo.dk> wrote: > >> Note that the closed form solution doesn't have to compute the intermediate >> values. >> .................... >> Here the closed form solution was 14 times faster and used 29 per cent less >> space. > >Bo, > >You are comparing here two solutions for computing S(5000). >What Skip was discussing was computing S(i) for i=1,…,5000. >These are different problems, so your data does not contradict his. >---------------------------------------------------------------------- >For information about J forums see http://www.jsoftware.com/forums.htm > > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
