With J504 on a 64-bit Linux system v1=:(1:+])@] v2=:%:@] v1 +--------+-+-+ |+--+-+-+|@|]| ||1:|+|]|| | | |+--+-+-+| | | +--------+-+-+ v2 +--+-+-+ |%:|@|]| +--+-+-+ (v1&0)^:(0) 256 256 (v1&0)^:(1) 256 1 (v1&0)^:(2) 256 1 (v1&0)^:(i.3) 256 256 1 1 (v2&0)^:(0) 256 256 (v2&0)^:(1) 256 0 (v2&0)^:(2) 256 0 (v2&0)^:(i.3) 256 |domain error | (v2&0)^:(i.3)256 9!:14 '' j504/2005-03-30/13:35
SUSE Linux Professional 9.2 (64-bit) On Sunday 16 April 2006 09:40, June Kim wrote: > Oh, now I get it. > > (<3) ((*:^:_1)@[)&0 ] 256 > 256 16 4 > > Bond has a special feature of powering! > > Now I am almost close to full understanding of your solution. > > What is the logic behind ~:/\^:_1 ? > > On 4/16/06, June Kim <[EMAIL PROTECTED]> wrote: > > On 4/16/06, Dan Bron <[EMAIL PROTECTED]> wrote: > > > June, > > > > > > I find (@:[)(^:]) unappealing. I prefer the dyadic invocation of a > > > bonded monad, ie. x m&v y or x u&n y , particularly when the > > > primary argument will be the number of iterations, and some seed data > > > is required for the intial iteration. > > > > > > See my earlier responses to this thread for an example. > > > > > > -Dan > > > > Thank you for the direction but I could hardly get it. > > > > I suppose the essence of what you are suggesting is: > > > > (100&(,~ *:^:_1)&256) <3 > > 256 0 0 > > 16 100 0 > > 4 10 100 > > > > However, I couldn't get the hang of it. > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
