Maybe you are looking for parametric representation
plot ;/((rot 0.1)mp(,:~ 0 1{.~#) p) (p."1) 0.1*i:90
> From: Raul Miller <[EMAIL PROTECTED]>
>
> If I take a polynomial
> require'plot'
> p=: ($&0 1 0 _1 % [EMAIL PROTECTED]) 30
> plot ;/(,: p&p.) 0.1*i:90
> and rotate it sufficiently
> rot=: (1 _1,:1 1) * (2 1,:1 2)&[EMAIL PROTECTED]
> plot ;/(rot 0.375)+/ .*(,: p&p.) 0.1*i:90
> it is no longer a function. I can, however, derive a function
> from this
> plot ;/(#"1~ (= >./\)@{.)(rot 0.375)+/ .*(,: p&p.) 0.1*i:90
>
> My question is: how do I find the rank 0 functions which
> correspond to general cases of that plot?
>
> Obviously, this derived function can not be a polynomial
> (because of the discontinuities).
>
> Also, I want this derived function to be defined in a fashion
> which is independent of the values represented in my
> "x domain".
>
> Finally, I am interested in cases where the polynomial is
> fixed, but the rotation is unknown (though it's reasonable
> to constrain its range -- for example, maybe I should
> restrict myself to the rotations corresponding to rot _0.5
> through rot 0.5.
>
> Can anyone offer me any suggestions on how to treat this
> problem?
>
> Also, I have a stylistic question. Does anyone have any
> preferences between
> plot ;/(,: p&p.) 0.1*i:90
> and
> plot j./(,: p&p.) 0.1*i:90
> or have any other opinions on how I should perhaps represent
> this question differently?
>
> Thanks,
>
> --
> Raul
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
