Loft

Replies to This Discussion

Permalink Reply by Danny Boyes on July 10, 2014 at 8:22am

Hi Laura,

Can you change the location of this to the Discussion forum (see here for help with that)

And also it might help to upload you GH definition to get a better understanding of the data structures involved with your 3 curves.

Permalink Reply by Laura on July 10, 2014 at 8:41am

Hi Danny,

Thanks I will

Laura

Permalink Reply by Damon on July 10, 2014 at 8:57am

I downloaded your definition, but it doesn't have any input values. Could you internalize the data in the surface and two curve components?

Permalink Reply by Laura on July 10, 2014 at 10:29am

Hi Damon,

I am attaching the Rhino and Grasshopper files.

Thank you

Attachments:

• Loft.gh, 11 KB
• Loft.3dm, 353 KB

Permalink Reply by Damon on July 10, 2014 at 1:51pm

OK, so I don't think you can do what you want to do in easily. The problem is the list organization in the curve components. I did a few tests that helped clarify this for me:

1. Baked the contents of the two curve components. The top curve component with the separate polygons is mostly closed polygons That would be good. But the ones along the boundaries are open. OK, a little harder to deal with. The bottom curve component with the honeycomb pattern have hexagons and connecting lines. So if you try to make a relationship between the polygons in the top component and the polygons and curves in the bottom curve component, it is hard to tell a computer to find those relationships.

2. I exploded the contents of the curves and looked at the order of the list in the first branch of each. They are not the same cell.

So, you are trying to match lines based on their spatial relationship, not on the order they show up in the lists. 

My recommendation would be to back up one step and look at the original grid and surface. It looks like you created the grids, baked them, sliced the grid and the surface, and are now trying to turn them into a grid shell structure. If you generated the "Hexa Grid 1" with Grasshopper, I bet those lines--and possibly the the offset ones--are better organized at the end of your grid script. Do you have that? Could we look at that definition it if exists?

Permalink Reply by Laura on July 11, 2014 at 2:38am

Hi,

thank you very much to look at my script. Yes, lists are the big problem.

I actually designed the original pattern flat and then projected to the surface, then manually I offset it on the surface. Do you reckon that if I use just one set of curves, let's say the mid one, and then parametrically do the offset and the loft, do you think I might sort the problem?

Permalink Reply by Laura on July 11, 2014 at 2:39am

Thank you!!

I will test it right now :)

Permalink Reply by Laura on July 11, 2014 at 2:52am

Hi Tom,

sorry to bother again

for some reason it fails to find the closest point in the second set.

I am trying to figure out why. Any idea?

Thank you again

Permalink Reply by Laura on July 11, 2014 at 2:56am

These are the print screens.

Attachments:

• Screen_03.jpg, 855 KB
• Screen_04.jpg, 798 KB

Permalink Reply by Laura on July 11, 2014 at 3:25am

Little update, still working on lofting

Attachments:

• Screen_05.jpg, 779 KB

Permalink Reply by Laura on July 22, 2014 at 5:33am

Hi Tom,

sorry late reply. I updated Rhino and Grasshopper and it work perfectly thank you very much, great work. Just a question, which is the name of the update List +1? I had to copy your script to get it, as I couldn't find it.

Thank you again

Laura

Permalink Reply by Damon on July 11, 2014 at 7:45am

Tom's done a fantastic job working with what you have. I'm just going to jump back in here and suggest again that we go back to the beginning. I've created a new file and definition, attached, that starts with a 2d grid on the xy plane and the design surface. Note that the 2d grid is made up of all polygons, the regular hexagons as well as the triangular-shaped hexagons. This makes offsetting much easier.

Then I offset the grid and offset the surface.

Project the offset grid to the upper (original) surface.

Project the original grid to the lower (offset) surface.

Because the two grids, original and offset, are created in Grasshopper, I know that the data structure of the polygons is the same in both cases. So when I go to loft them, it is a one-to-one relationship.

I think this way you are dealing with a very simple and straightforward definition. There are really only two bits of monkey-business:

1. Separating the closed and open curves so the Loft component is happier.

2. Creating the top surface.

These are explained in the definition with notes and are quite simple.

Just another take on the problem. Hope it helps.

Attachments:

• HexaGrid.gh, 23 KB
• Loft.3dm, 500 KB

• ‹ Previous
• 1
• 2
• Next ›
• Page