I see - I guess I didn't annotate the follow-up to that comparison in which
I (silently) demonstrated that the two are really the same:
>./|xx-xx2 NB. Largest difference
9.536725e_7
(] i. >./) |xx-xx2 NB. Where is largest difference?
20
20{&>xx;xx2 NB. This looks like a rounding error.
0.50000095 0.5
On Fri, Mar 12, 2010 at 2:44 AM, Aai <[email protected]> wrote:
> Sorry Devon it's not for the timing I pointed to that. It's just for the
> comparison of both results.
>
> > xx-:xx2
> > 0
> >
>
>
> Hallo Devon McCormick, je schreef op 11-03-10 21:25:
> > That's two out of a million - I doubt the timer is precise enough to
> capture
> > the difference between the two cases.
> >
> > On Thu, Mar 11, 2010 at 2:23 PM, Aai <[email protected]> wrote:
> >
> >
> >>
> >> Hallo Devon McCormick, je schreef op 11-03-10 20:11:
> >>
> >>> It's a pretty good improvement already.
> >>>
> >>> hc10k =: , ] +/@:{~ (,&<: -.)@{:
> >>> AnN =: % 1+i.@:# NB. a(n)/n
> >>> MxAnN =: >./;.1~ 2 (=<.)@:^. 1...@# NB. Max a(n)/n btw successive
> >>>
> >> 2^
> >>
> >>> HCF=. 3 : '1 1,(#2...@#);(,<@;@(|.@(#;. 1) (],~1$~[)&.>
> >>> <@:>:;.1)@>@{:)^:(y-2) <2'
> >>>
> >>> 6!:2 'xx=. mx...@ann 1 1 hc10k @]^:[~ 2^20'
> >>> 2485.4382
> >>> 6!:2 'xx2=. mx...@ann HCF 20'
> >>> 0.38284575
> >>> xx-:xx2
> >>> 0
> >>>
> >>>
> >>>
> >> be careful in choosing comparable arguments:
> >> in this case hc10k needs _2+2^20
> >>
> >> --
> >> Met vriendelijke groet,
> >> =@@i
> >>
> >> ----------------------------------------------------------------------
> >> For information about J forums see http://www.jsoftware.com/forums.htm
> >>
> >>
> >
> >
> >
>
> --
> Met vriendelijke groet,
> =@@i
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
--
Devon McCormick, CFA
^me^ at acm.
org is my
preferred e-mail
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
