Two variable function in local axis

Replies to This Discussion

Permalink Reply by Mar Ferrer on November 20, 2010 at 1:32pm

ok, let s see if I find a better way to describe my problem...I have a series of tilted walls and each of those walls have an hypar (hyerbolic paraboloid) as a surface...so it will be crazy to do that by hand and I think grasshopper would be extremely powerful if it can do the transformation!
Let me know!

Permalink Reply by Evert Amador on November 20, 2010 at 1:57pm

Hi Mar,

From what I understand from your post something you can do is to perform a mapping(as in mapping the domain of your function that defines the original surface with the domain of your second function) so you can perform a translation/scaling/rotation depending on your needs from your original coordinate system to your new coordinate system.

another option will be to use an equation for your f(x,y) function to solve the points and then apply the same equation on the domain of the original function with the mapped values or rather with a function that scale/rotate/translate these values on the new coordinate system.

Maybe I'm not been very clear, sorry.

I would say f(x,y) is not a valid expression, maybe f(x)=somethingX where f(x)=y. That is to say y is the image set of f(x), by doing this you only depend on one variable that makes solving its equations easier.

It may be better if you post a screen shot of what you have and/or the actual functions to better understand the problem.

Cheers


EDIT: your response didn't appear until after I wrote my reply, there is a plug in for rhino called rhino math, where you enter the equations and gives you the surface.

http://www.rhino3.de/album/math/index.shtml

I hope this helps.

Cheers again

Evert

Permalink Reply by David Rutten on November 20, 2010 at 5:09pm

Hi Mar,

what's the function? I attached a file which creates a Nurbs surface from scratch from an f(x,y) function, and also a way to modify an existing mesh using that same function.

--
David Rutten
david@mcneel.com
Seattle, WA

Attachments:

• fxy.ghx, 120 KB

Permalink Reply by Mar Ferrer on November 21, 2010 at 2:47am

Thanks everyone for all that info...yes, what I need is to map the surface afterwards and this way I can achieve this change of plane...However I am trying to get the points on the borders of the mesh_surface and can not achieve, I do not see any dupborder...actually what I need is to define the points at the border so that I can join the diagonals since it is a revolution surface after all and that gives the possibility to build it easily (I mean the hypar...z=(x^2)-(y^2)....)
is there a quick way to do that?

Permalink Reply by Mar Ferrer on November 21, 2010 at 3:10am

also, I have another question...actually what I am trying to do is to model the Philips pavillion which is a building designed by xenakis for brussels exhibition...well anyways, all the shapes are hypars...and as I really need to get it done as soon as possible (hopefully tomorrow) I am trying many strategies...my other strategy is to build up on the projections that I have traced from an ancient drawing...and those are supposed to be hyperboles...is it a way that I can approximate a curve_polyline that I have traced in rhino with an hyperbole????
thanks a lot...as it is still work in progress is really a mess so when I do sthg nice, I ll post it!
see you all tomorrow, I hope

Permalink Reply by maria.llerena.inesta on August 3, 2011 at 3:29am

Hola Mar,

He localizado este hilo sobre un tema que me interesa, lo que no sé es si aún andarás por aquí.

Estoy liada con un trabajo Fin de Máster para la Universidad de Sevilla sobre el Pabellón Philips y el análisis de su generación formal. Me gustaría realizar las comprobaciones de su geometría con Grasshopper y he visto que tú has estado trabajando sobre el tema... podrías faciliarme algún scrip del modelo o algo que me ayude a definir su geometría?

En fin, me sería de gran ayuda!

 

Gracias!