DesignReForm

Here are the source files.

Someone on the Grasshopper Google Group wanted a version with five points so I'll post it here too.

Cheers,
taz

Thanks Taz!

here's my attempt with attractor points.  more controls of the gradient effect and also somewhat more "intuitive" in the search for the minimum distance between a given point grid to the closest attractor point.  thoughts?

Hey tnarch,

Let me take a look at your file... 

In the meanwhile I made a slight improvement to the previous 5 pt file I posted (based on a Google Groups discussion).  It might be similar to what you're up to but now the influence of each point is defined by a radius distance.

taz

tnarch,

I took a look at your file.  Influence values for each point is an interesting idea.  Also specifing a minimum circle size within the influence area is a good control.

It would be my preference to get rid of the "Min" tree and replace it with a 5 or 6 variable function, but that's just me.

I think I've seen you over at Google Groups, but if you haven't already found it you might be interested in this discussion:

http://groups.google.com/group/grasshopper3d/browse_thread/thread/100322ed39026776?hl=en#

Towards the end of the discussion someone modified the 5 pt file to draw lines from the center of each point to the circles that each are influencing.  It's an interesting visualization.

taz

taz

My concern with the n-variable function is that it's "too much math".  I'm trying to develop some course work and the example I posted was an attempt to minimize the use of "scripting logic".  Although, your example with the cull patterns is brilliant.  Also, I was developing the example with a commnent David Rutten has made about GH tagetting "non-programmers" in mind.  Thanks for the feedback.

Tuan

How would I create a pattern with say a half-circle randomly mixed in with the full-circles shown in the tutorial?

mschuepbach,

I *think* this could be done with arcs.  You could set up a routine to randomly feed arcs either a 180 degree sweep (half-circle) or a 360 degree sweep (circle).

I can try to mock something up a little later...

taz

mschuepbach,

I didn't modify Dave's file but I modified my own.  The arcs are created at the end of the stream so it doesn't really make a difference.

A couple things to note:

You can change the "random" pattern by changing the seed value on the random component.  I think the same seed number will always produce the same "random" pattern which is why it's pseudo-random number generation.

The input value for the arc sweep is looking for an interval, so when I feed it a list of pi or 2*pi it assumes the interval starts with zero.  This is a bit of trick.  I was also going to say you could use an expression to convert from degrees to radians (if you wanted to use degrees), but you would have to add a single variable function to do this before feeding data to the arc component.  Apparently you can't use expressions on intervals.

That's probably more info than you wanted, but good luck!

taz

P.S. I also added a more updated version of the pure circle routine below.

I am setting up a custom perforation pattern for a project at work.  The file is set up and working properly per this tutorial, but how can i give a min AND a max radius for the perforations?  I am having a difficult time controlling the pattern.  Maybe i missed something somewhere.  Also, is it possible to create a curve to affect the circles in the same manner as the point?  Thanks for any replies!

Matt

mschuepbach,

I'll try to take a crack at adding in a control for minimum circle size such that there is a linear radius gradient from MIN to MAX within the point radius of influence.

I'm not sure what you mean by "is it possible to create a curve to affect the circles in the same manner as the point."  You'll have to spell that out a little better.

taz

Alright, it took some expression work but by adding a 3 variable function downstream the radii of the circles now vary linearly from a MIN to MAX radius within the radius of influence.

The screenshot may be pretty subtle but the circles are spaced at 10 units with a MIN radius of 2 and a MAX radius of 5.  The radius of influence is 100 (or 10 circle positions).

Cheers,
taz

Hi yall
I'm very, very new to parametric modelisation, and absolutely not knowledgeable in scripting of any form, and I'm French so please excuse in advance the mispelling or misunderstanding I may have.
However, I've been working as an independent architect and 3rendering artist for more than ten year, so I'm not completely new to all this.
My question is fairly simple: considering the use of the "function" command for the 5 point generator, I was wandering if this example was limited to eight generator points since that appears to be the limit of the function command in grasshopper.
Say, I have a facade for a building; I want to generate vertical bricks, with a constant spacing (like the basic grid of this example), a constant height but a variable width, generated by the radius of the circle.
Since the scale of my bricks is quite small in proportion compared to my 16 storey building, is it possible to modify this example for say, 16 or 32 generator points?
And to completely ask the question, how can I use the data of the center and radius to assign them to a corner position and width for a rectangle?
Sorry if it's not the right place for that

Your's truly

Suitaloon

Ok
Thanks to taz on the google group, I've found a way to genrate boxes (the bricks) from the coordinates of the grid shifted in Y and Z axis (first corner)and after dividing the circle with a value of one, I get the other corner.
Thanks Taz.
I still wander if there's a way to use more than 8 attractors?