|
|
clipka <ano### [at] anonymousorg> wrote:
> Am 09.02.2018 um 04:43 schrieb mathzhaoliang:
>
> > Hi clipka: Thanks for your help! But my problem is still there: my tiling shapes
> > are not among these built-in ones, it's like the penrose tilings but the
> > rhombus are arranged in other way (each rhombi is computed and specified by its
> > 4 corners, in fact it's called aperiodic tilings), and I have to set the texture
> > for each rhombi manually. So the normal type identifier **tiling Tiling** does
> > not work in my problem.
>
> To be nitpicking, the Penrose tilings /are/ aperiodic tilings. They're
> /quasicrystals/ though, which not all aperiodic tilings are.
>
> If your tiling is not a Penrose tiling, then you're out of luck with the
> `tiling` pattern: If it's not in the repertoire, you can't get it this way.
>
> It may be possible to achieve your goal with a function pattern, but I'd
> expect it to be non-trivial, and also potentially slow to render.
>
> The inbuilt Penrose tiling is not implemented via precomputed rhombus
> corners, but rather via an iterative algorithm based on substitution
> rules, so this approach cannot be extended to non-regular aperiodic tilings.
Updated work: it's very closed to your image now
https://photos.app.goo.gl/eVfKrhS46Au1vsay1
By aperiodic tilings I mean a large class of generalized Penrose style
tilings consist of rhombus such that any two of them are not locally isomorphic
with each other, it's generated by de
Brujin's penta grid method. The usual Penrose tiling is just a special case of
it.
Post a reply to this message
|
|