So... here's a collection of some of the definitions seen here, with some timing runs:
NB. Xiao-Yong Jin's implementation as revised by Pascal Jasmin NB. http://jsoftware.com/pipermail/programming/2017-July/047881.html revised=: ''1 :(0 :0-.LF) /:~@( -&1`]@.(=&0)"0 (#/.~@] + ~.@]{[) `(~.@]) `[ } >&0 # ( ?@#~@#@]`]`[} ] I.@:= i.@# )@?@#~@#) ) NB. Mike Day's implementation NB. http://jsoftware.com/pipermail/programming/2017-July/047992.html new=: (, (# | (+ ?&.<:@#~@#)))@(i.@#) +//. ((- , ]) 1&<.) NB. Louis de Forcrand's version NB. http://jsoftware.com/pipermail/programming/2017-July/047996.html RUN=: UPD rec UPD=: 4 : 0 b -~ x u}~ (x {~ u=. ~.y) + y +//. b=. x >: 1 ) rec=: ((] + ] >: i.@[) (?@$ <:))@# NB. Raul Miller's revised version NB. http://jsoftware.com/pipermail/programming/2017-July/048000.html tock=: ((0 >. <:) + _1+i.@# #/.~@, I.@:* (] + <:) +/@:* ?@# <:@#) NB. Skip Cave's version: NB. http://jsoftware.com/pipermail/programming/2017-July/048003.html hist =: <: @ (#/.~) @ (i.@#@[ , I.) NB. Histogram iter =: 4 :0 nx =: y$y NB. initialize money while. x>0 do. sg =: +/nx>0 NB. How many still giving? nx =: nx-nx~:0 NB. Those that have, give nx =: nx+(i.y) hist ?sg$y NB. distribute dollars x =. x-1 end. ) NB. timing ts=: timespacex try=:3 :0 echo y echo ".y ) testrun=:3 :0 'n g d'=. ":&.> y subst=. rplc&('NN';n;'GG';g;'DD';d;' 1 ts';' ts') try subst ' NN ts ''revised^:GG #~DD''' try subst ' NN ts ''new^:GG #~DD''' try subst ' NN ts ''RUN^:GG #~DD''' try subst ' NN ts ''tock^:GG #~DD''' try subst ' NN ts ''GG iter DD''' echo '' ) testrun 1 5e3 200 testrun 1 5e3 10 testrun 10 1e4 100 testrun 100 1e4 20 testrun 100 1e4 200 NB. ............. test runs follow: 1 ts 'revised^:5000 #~200' 0.752696 30208 1 ts 'new^:5000 #~200' 0.261853 20992 1 ts 'RUN^:5000 #~200' 0.060494 25344 1 ts 'tock^:5000 #~200' 0.04234 18432 1 ts '5000 iter 200' 0.049315 18560 1 ts 'revised^:5000 #~10' 0.057119 6144 1 ts 'new^:5000 #~10' 0.023318 4992 1 ts 'RUN^:5000 #~10' 0.017986 7040 1 ts 'tock^:5000 #~10' 0.010964 3840 1 ts '5000 iter 10' 0.017491 2176 10 ts 'revised^:10000 #~100' 0.753988 16384 10 ts 'new^:10000 #~100' 0.266346 11520 10 ts 'RUN^:10000 #~100' 0.0737156 14848 10 ts 'tock^:10000 #~100' 0.0521216 10240 10 ts '10000 iter 100' 0.0603638 11136 100 ts 'revised^:10000 #~20' 0.198089 6144 100 ts 'new^:10000 #~20' 0.0712385 5248 100 ts 'RUN^:10000 #~20' 0.0434272 7040 100 ts 'tock^:10000 #~20' 0.027364 4096 100 ts '10000 iter 20' 0.034847 3456 100 ts 'revised^:10000 #~200' 1.53964 30208 100 ts 'new^:10000 #~200' 0.57214 20992 100 ts 'RUN^:10000 #~200' 0.140146 25344 100 ts 'tock^:10000 #~200' 0.095298 18432 100 ts '10000 iter 200' 0.100683 20352 FYI, -- Raul On Thu, Jul 20, 2017 at 2:26 PM, Skip Cave <s...@caveconsulting.com> wrote: > My explicit, looping version: > > NB. y is number of people and dollars > > NB. x is iteration count > > hist =: <: @ (#/.~) @ (i.@#@[ , I.) NB. Histogram > > > iter =: 4 :0 > > nx =: y$y NB. initialize money > > while. x>0 > > do. sg =: +/nx>0 NB. How many still giving? > > nx =: nx-nx~:0 NB. Those that have, give > > nx =: nx+(i.y) hist ?sg$y NB. distribute dollars > > x =. x-1 > > end. > > ) > > ts =: 6!:2 , 7!:2@] > > 1e2 ts '1e4 iter 20' > > 0.0262932 4224 > > > 1e2 ts '1e4 iter 200' > > 0.0915201 20352 > > > 10 ts '1e4 iter 100' > > 0.0528825 9344 > > Skip Cave > Cave Consulting LLC > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
