algorithmic modeling for Rhino
Hi folks,
I've found a tutorial where a component is used to panelize a surface. The component can be open or closed based on a certain sun light angle.
I was wondering if the same concept can be used but with an image: for the parts that are white I will have the closed component and for the parts that are black I will have the open component.
A variation I would like to have, for the open components, is to have a middle surface, working as a glass.
Can anyone help, what do you think?
Tags:
This is certainly possible. If you haven't used the Image sampler before the I would suggest this approach.
1) insert a Picture Frame of your chosen image into Rhino.
2) Reference Surface from Rhino.
3) Reparameterise the Referenced Surface.
4) Use the SDivide Component to generate you Points.
5) Use the UV output of the SDivide component
6) Have the Image sampler Domain as 0 to 1.
This way allows you to apply your "Manipulation" directly to the SDivide points so that you can see what is happening in the Rhino Document on top of your chosen image.. This quickly highlights any irregular or unexpected behaviour.
Well,
I've been trying, the last couple of days, and I'm sure that the theory behind all this it's not that complicated, but the 'doing it' part it's quite tricky, for me at least.
This is what I got [see the .gh file], wich is what I had in mind BUT the only thing is that i want my image to work as a hole on the surface instead of a repeated image [in this case I want the words 'the iqons' to be the ones with the open component, and therefore to be visible through the gaps on the surface] so i tried to change the propeties of the image sampler from 'tile' to 'clamp'.
Just to be clear enough, I would like my surface to work as this one http://www.grasshopper3d.com/photo/01-does-anybody-know-how-to-crea... but I will have closed components in areas corresponding to the white color, in my image sampler, and open componets in areas corresponding to the black filling.
But something it's not working, clearly. What do you suggest? It's kinda making me go nuts!!! I know I'm nearly there!!!
Thank you guys!
as Danny said, if you properly reperameterize and use the image sampler at domain 0 to 1, then no mater what the image will fit your surface. http://formularch.blogspot.com/2012/04/gh-image-sample.html
I've checked it hundreds of times, but the result it's not right: all i get is 'closed components' even though I can see that my image is read correctly [i have data from 0 to 1].
Can you have a look, please?
Three things you should have a look at.
Firstly
Reparameterise the input surface.
Secondly
10 x 10 SDivide = 121 values
this is because the divisions are actually the spaces between not the UV co-ords.
Where as 10 by 10 Domain Divide = 100 values
To rectify simply increase Domain Divide inputs by 1 to get 11 x 11 = 121
Thirdly
The Data Matching in the Dispatch Component is for 121 values on 1 branch to 121 values on 11 Branches.
I've flattened the pattern and now have 121 values on 1 branch for each input, which produces a result of open and closed panels in the same shape as your image.
Welcome to
Grasshopper
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
© 2024 Created by Scott Davidson. Powered by