I'm still a little unsure about what I want to do, but my mood/idea board is to the left. I'm thinking of something that is similar to the project from earlier, but smaller and with a rose shape instead of whatever the shape I did last time was. Regardless of what I end up doing, I think it'll probably end up looking organic.
My original idea was a rose shape. It sort-of evolved from that which you'll see below. The sketch wasn't great, but it was enough for me to decide that it was worth the time to do in rhino. I originally wanted it to be jewelry, but the scale makes that a little too difficult. I eventually decided on just making a desk decoration that is in the shape of a flower. My desk is currently pretty bland, and I thought it'd be cool to add a very unique shape that isn't seen anywhere else in my room.
I'll upload the grasshopper script and rhino stuff later once it's finalized, but on the left is the model that I'm doing my prototype on an Ultimaker with. It's larger than the final scale because I wanted to play it safe with the ultimaker, but it should be fine to print it at a smaller scale at shapeways.
Tutorials I followed:
https://www.youtube.com/watch?v=_3_7sDce6ME (to use graph mapper to make the shapes I wanted)
https://parametrichouse.com/proximity-chair/ (to figure out why my multipipe wasn't working).
I messed up the ultimaker settings and accidentally started printing the supports with the regular PLA (which would've been a disaster, but I'm lucky to have caught it early on)
This time, it actually worked! But as great as that is, this file has weird geometry and too many poly's to send to ultimaker.
While making the Grasshopper script, I have been (and continue to be) running into issues. Firstly, with multipipe, it made many smaller objects instead of a single large one. This was a problem, as when I reduced the poly count, it messed up the shapes. I resolved this by flattening the output of the proximity component. After that, certain pipes started jutting out of the shape, which I learned was due to the sharp angle that lines intersect. The only solution I found was to identify the index of those components in a giant list and cull them. Then, because of intersecting mesh parts, I remeshed them in blender and decimated the mesh. But now, it's too expensive! We'll see how the rest of it goes. Unfortunately, I don't have any screenshots because i was too focused on changing the sliders to try and get it to work (and it took like 1 minute to update every time a single component was changed).
Not much more to say. This model has been reduced and remeshed to be sent to shapeways.
Not much else to say. The features were too thin for metal (even after it passed the automated checks), so now I'm on plan B, which is just using fine detail plastics.