Pattern

GH Community,

Seeking help for patterning problem i having. Thank you in advance for your help.

I am trying to cull a grid of points with the following pattern:

true

false

The first image (Capture1) shows the pattern in the correct way with the exception that it only has three (3) "false" inputs. When I add the fourth "false" input, the result is the second image (Capture2) which highlights entire rows. This is not the desired result. The pattern I want to see when selected, is the first point (0) green, then four red (1-4) points, then the fifth point (5) green. Any advice would be greatly appreciated. Thank you.

Replies are closed for this discussion.

Replies to This Discussion

Permalink Reply by Alyne Rankin on October 19, 2017 at 6:24am

Joseph, i especially like the diamond pattern. Thank you so much for sharing these. Could a phyllotaxis spiral be used for the grid?

Permalink Reply by Joseph Oster on October 19, 2017 at 7:02am

I like the diamond pattern too. And yes, a phyllotaxis spiral would make a fine "pixel pattern" for the 'Image Sampler'.

http://www.grasshopper3d.com/forum/topics/phyllotaxis-tutorial

Permalink Reply by Joseph Oster on October 19, 2017 at 7:29am

Here is a very quick adaptation of phyllotaxis pattern code from Jesus Galvez:

Attachments:

sortlist_2017Oct19b.gh, 23 KB

Permalink Reply by Joseph Oster on October 19, 2017 at 7:56am

Added blue group to clip spurious circles outside the rectangle, which reduces compute time for the disabled group - keeping the total number of holes to 3707:

Attachments:

sortlist_2017Oct19c.gh, 26 KB

Permalink Reply by Joseph Oster on October 19, 2017 at 8:34am

Added an 'MD Slider' to pick a center point for the pattern (makes a BIG difference!); added another blue group to clip pattern points outside the rectangle before passing them through 'Image Sampler', in part because moving the center point requires increasing the 'N:0-100' slider to 14000 to cover the image, though most of those are outside the rectangle. 3720 holes at these settings:

Attachments:

sortlist_2017Oct19d.gh, 27 KB

Permalink Reply by Joseph Oster on October 19, 2017 at 6:43am

This version is better. Simpler. Doesn't need 'Project' and 'Srf CP'.

One of the keys to success later in creating a perforated surface (the disabled group) is keeping the number of holes to a minimum. 'resolution' and 'size filter' make a BIG difference. In this case, 2619 holes takes 11.1 minutes.

But it still has one major flaw; no ability to scale the output independent of the image size while keeping aspect ratio.

Attachments:

sortlist_2017Oct19a.gh, 21 KB

Permalink Reply by Joseph Oster on October 19, 2017 at 12:47pm

Scaling of pattern using image aspect ration, then using 'Srf CP' on reparameterized surface to "translate" to zero to one domain - which 'Image Sampler' has been re-configured to use instead of the image pixel dimensions.

Attachments:

image_sampler_2017Oct19e.gh, 23 KB

Permalink Reply by Joseph Oster on October 19, 2017 at 1:54pm

Similar scaling applied to the phyllotaxis pattern, except the pattern has its own controls and is not affected by the rectangle sizing. As above, this 'Image Sampler' expects input values in the '0 To 1' domain.

Attachments:

image_sampler_2017Oct19b.gh, 30 KB

Permalink Reply by Alyne Rankin on October 28, 2017 at 12:26pm

Joseph, Is it possible to control sampled resolution of the image around specific areas?

I have attached another image i am working on and i am trying to find a good balance for resolution and max radius and the filter. The definition uses a higher resolution image of the world and i would like to control the max radius but obtain more detail around the coast line (specifically the great lakes area).

I would like to control the larger radii with a slider that would cover the bulk of the mapped area. The target radius would be 0.75" (for my final output size)
I would like the radii to fall off and get smaller closer to the coast line in turn providing more detail - controlling the number of steps in that they get smaller would be ideal
and also having the ability to control the radii of each incremental step would be great too.

As i type this, would circle packing be an option?

Attachments:

World_Image102817_01.gh, 291 KB

Permalink Reply by Joseph Oster on October 28, 2017 at 12:41pm

I would like the radii to fall off and get smaller closer to the coast line in turn providing more detail

What's a coast line? I mean that seriously, in the sense that how do you determine the edges of the surfaces? And even if I had those coast lines, I'm not sure how to use them to increase resolution near them, sorry.

Permalink Reply by Alyne Rankin on October 28, 2017 at 12:49pm

I see because its an image made of pixels with no defined edge. ok, back to the drawing board....thank you for the quick reply.

Permalink Reply by Alyne Rankin on October 28, 2017 at 1:47pm

Joseph,

Could i ask for you help again - I am having trouble sampling the points on the surface normals as it relates to a curved surface. Could you assist?

Attachments: