David,
Thanks so much. You're right - I should look at modeling the columns
more accurately and similarly to how they were originally modeled out
of plaster... If I have time to get some sleep in, I'm going to go for
it. In the meantime, I'll work with the expression. Thanks so much!
On Oct 14, 3:35 pm, David Rutten <[EMAIL PROTECTED]> wrote:
> Hi Noah,
>
> the meshes computed by Grasshopper are very coarse. I need to keep the
> settings low because a lot of meshing has to happen pretty much in
> real-time. Once I'm comfortable running the mesher in a thread, I can
> progressively increase mesh quality without affecting the response
> time of the Grasshopper interface.
>
> Gaudi columns are actually region booleans of multiple curves, they
> have sharp edges in some sections, yet are completely smooth in others
> where the regions start to overlap. At any rate, you need to create a
> set of components which makes cross sections.
>
> If you want to wrap a Sine function around a curve, and you're
> comfortable using mathematics then making an expression is easiest:
>
> Point = {Origin.X + Radius * Cos(a) + SmallRadius * Cos(freq*a),
> Origin.y + Radius * Sin(a) + SmallRadius * Sin(freq*a), Origin.z}
>
> where Origin is the center point of the column, a is the angle (goes
> from zero to 2pi, in small steps so you have an accurate curve),
> Radius is the radius of the column, SmallRadius is the radius of the
> waves on the circle and freq is an integer which controls the number
> of waves around the column.
>
> If you don't like expressions, you'll have to do something visually
> intense using Graph mappers, vector multipliers etc etc.
>
> --
> David Rutten
> Robert McNeel & Associates
>
> On Oct 15, 12:23 am, noah <[EMAIL PROTECTED]> wrote:
>
> > New to grasshopper and attempting to model gaudi's section-morphing
> > columns. Tried searching prior posts, to no avail...
>
> > I'd like to create a column section comprising a sin curve along a
> > circle. The basic idea is that I would be able to adjust the frequency
> > and amplitude of the sine curve, as well as the size and location of
> > the circle. By lofting a whole series of different sections - voila! -
> > column. Any ideas on drawing the curve I'd like?
>
> > Also, the loft function gives me some funny twisted and broken lofts,
> > unless I simply loft the baked geometry in Rhino. Any idea how I might
> > be able to avoid this?
>
> > Thank you all very much. We're lucky to have such a lively discussion
> > board.
