I am creating a new topic to hopefully get a response and in case my question was not relevant to the topic that I posted it on. My question is somewhat relevant to this topic, but I want to subdivide four surfaces. I have reparameterize one surface as a test, but some panels that are supposed to be flat and rectalinear are not rectalinear and do not have the same lengths on their parallel edges.
Can you offer an explanation on why this is occuring or how to resolve this portion of my script.
Disclosure - The design intent is to subdivide the surface to get flat panels and further subdivide the curvalinear and use radial faceting to make straight panels. (Seems likely more development will be needed) I will also subdivide the flat panels to be similar dimension to the radial faceted panels. Please see the image and files below.
I think there might be a misconception about the Reparametrize
Reparametrize does not “straighten/stretch/shrink” the curve function in such a way parameters t travel linearly, but just remaps the numeric values of the Curve Domain in such a way its Bounds are 0 to 1
the behavior (you can think at it like the speed and accelerations at which the t parameter moves) stays the same, Reparametrized or not
for instance, a straight line with 2 control points (top) and a straight Curve with several control points unevenly distributed along its length, both evaluated by the same amount of points equally distributed along their [0,1] Domain:
you can interpret those curves as if they were the top edge of your surface, and the middle curve in your surface: they do not travel at the same speed
there are a couple of very interesting articles from David Rutten’s blog:
[edit: replaced GH file because curves were not internalized]
Thank you for the timely reply, and please excuse for the late reply. I came to a similar concept while I was waiting for someone to answer my post.
After skimming the article, I used the bottom curve as a curve to create planes at my desired panel width, and the panels are almost the same size with a tolerance of .0001. Did you know of a way to get the panels to be a dimension at whole inches?
I can get them close but the both bottom and top dimensions have fractional inches at straight panels of less than 1/256". In the script file I am uploading, the rectangular panels are at whole inches but in my personal file that has the geometry (surface and point) as part of the script file I get very small fractional dimensions. It may be something with my rhino file. Do you happen to know why. I will continue thinking and researching on why that is the case. 20240923 - Facade Script - Forum_Post.gh (200.8 KB)
you are shortening the bottom curve in such a way you have a short trim on the left side and on the right side, and divide the remaining part into 70 identical smaller portions of 72 units of length each
that works well for the bottom curve on which the whole definition is based on: the bottom curve is for sure divided into 70 parts of 72 units each (the thing of having some 72.00004 and some 71.999999 is just a matter of Precision that you can enhance on Rhino Options → Units → Tolerance and the press F5 to update the GH file)
your final Surface is not a perfect vertical extrusion along Z-axis of your bottom curve, so I think I would not expect a perfect divisions for the other curves -which are pretty different from the bottom one-
Those mid and top curves have different length, different shape, and I guess they would also have much different Perpendicular Planes orientations on the very same locations…
I have not understood exactly what is your final goal, could you please make some sort of fast sketch, even paint on a screen grab the final result you want to achieve?
Thank you for the reply. I have posted an image of the building mass that I am working on for a school project that I am redesigning to get better at Grasshopper for a position that includes parametric projects in their work.
I want to have a glazing system at the vertical parts of the mass along the long edges of the building, and maybe a double-height glazing at the short edge of the building. Where the surface curves can be a cladding system.
Secondly, I want to use the unitized cladding panel system based on a Think Parametric tutorial series that uses the logic of the Soumaya Museum on the part where the surface curves vertically. I will deal with the soumaya cladding logic part when I get to that part. I don’t want it to be hexagonal cladding, but maybe skewed rectangles.
The original brep was made with two lofts with loose as the type of lofting method. So the loose loft may be creating the non-vertical surface that I sent you.
I will work and experiment with nurbs curves or regular lofting method.
Please excuse me for an additional post, but the second image was supposed to show something similar to my desired glazing system. I am working towards a glazing system that is in the same plane as the backing for the cladding system, and where it is located below the vertically curved cladding, the cladding backing sits on top of the 8’ - 10’ glazing system. Thank you in advance, and please share with me what I should be doing if the post is asking too much of you.
@inno I did not think that you would think that. That picture is meant to be a reference to a similar type of glazing approach.
I should have sent an image of a horizontally curved facade with unitized (somewhat equal) panels. The panels that have vertical curvature will only have straight glass on the lower end of the panels.
The two-direction curvature will have straight cladding panels, panels that are straight. I was thinking of using the soumaya museum method for the cladding.
Please excuse me, but good news. The precision suggestion you mentioned worked so now the straight panels are reading as straight, and probably have corner angles of 90 degrees.
A new problem is one panel does not have a curve that splits it. I am thinking of how I can answer my own question which would be to get it in its own list using a length equality group of components. After splitting it into two panels using the lowest point to make a surface and split the panel, and then insert them back into the list. Do you have a reason of why the curve is not whole? (I researched a bit, and it may have to do the with curvature of the surface at that point.)
Also If I could ask you, how would you create straight panels at the parts that have curvatuve. I was thinking of joining the portions that have curvature then using the sa
20240923 - Facade Script 02 - Forum_Post.gh (649.8 KB)
me group of components to create horizontally straight panels that curve vertically and then split them when they begin to curve as increasing in height?
Please excuse me, but when I have been working on the file, the precision of the script or my geometry may have changed, because my the straight panels are reading as .000001 and similar values. Do you know how to confirm that precision has been changed in grasshopper?
@Joseph_Oster Hi, I had different questions, and I forgot something a topics and then wanted to ask another question. I did not have an intentions to be rude. It felt similar to a conversation. Thank you for the concern.
@scottd Reaching out to you Scott to see if you can move my learning about facades through a personal portfolio project toward completion. I have gotten to the point similar to Raffles City by UNStudio - Planar facade paneling post
and the post about
I was able to get planar panels, but like the doubly curved surface post, some of my panels are
deviation too high
Is there a way you can explain to replace similar types with a unit that fits in the model (if possible) meaning water-tight solid and imported into Revit with Rhino Inside Revit as curtain wall or does it need to be an adaptive component?
I do not know how to write the part of the script that decreases the deviation. How do people rebuild the curves? I can attempt to sketch out a method and work on rebuilding the curve if that would work. 240930 - Facade Script Post.gh (2.0 MB)
decrease the deviation and get panels within tolerance to be glass panels near the floor level until I hear from you. Please see the attached model and script file.
Thank you for the potential answer, and I will continue looking for the answer.
