ive 'correct' normal.
Non-normalized cross products is effectively weighting face normals by area, and is fast and simple, so we put that one as the default.
In some cases normalizing the cross-products improves the result, but not always.
Another option is to weight by angles, though this is computationally slightly more expensive, so might not be ideal for real-time updates on large meshes.
As an example, here is a mesh with a 90° corner, and uneven meshing on the 2 sides.
The arrows show:
0- Area weighted (non-normalized cross products)
1- Angle weighted
2- Normalized cross-products
Here the angle-weighted normal is the one at 45°, which is intuitively the 'best' one in this case.
These 3 seem to be the most commonly used, but there are many other possible definitions of normals - such as inverse-area weighted, mean curvature, etc...
I think really what would be best would be to put a few of these into Plankton, and include an optional argument in GetNormal for selecting which one you need for a particular application.
Pull requests welcome if you feel inspired to add this!
http://meshlabstuff.blogspot.co.uk/2009/04/on-computation-of-vertex-normals.html
http://steve.hollasch.net/cgindex/geometry/surfnorm.html…
us allows Grasshopper authors to stream geometry to the web in real time. It works like a chatroom for parametric geometry, and allows for on-the-fly 3D model mashups in the web browser. Multiple [Grasshopper] authors can stream geometry into a shared 3D environment on the web – a Platypus Session – and multiple viewers can join that session on 3dplatyp.us to interact with the 3D model. Platypus can be used to present parametric 3D models to a remote audience, to quickly collaborate with other Grasshopper users, or both!
You can down load the Grasshopper plugin at food4rhino, and visit 3dplatyp.us to view your geometry on the web. This first round of Alpha testing will run for two weeks, until April 24 2014, after which the Grasshopper components will not solve.
We are very interested in hearing feedback from the community while the project is still in the prototyping stages of development. Please use the comments on this discussion to ask questions, suggest ideas, report bugs, etc. We are planning on rolling out another public alpha release or two this Spring, depending on how this first one goes, in advance of our Technology Symposium and Hackathon in New York.
Check out our getting started video below, and enjoy!
…
e rod with circular section (no goals allow for controlling torsion for what I know). The rods are set with two options, with straight rest position or the (initial) bent one. The calibration integrated with the model is more about giving a scale between the forces rather than the will to accurately simulate them (at the moment). Anyway, I am trying to do it on a macro scale, instead of a micro, with elements which are rather thin.
The system at the moment is not stable. In fact, besides the rods' characteristics is quite fundamental to keep them planar when they intersect. I am lacking something but also probably missing some parameters. In the script, there are two goals to define this: impose 90° between vertical and horizontal, as well as between these and a normal to their intersection. For my understanding, angle goal works tri-dimensionally without a preferred plane and this (hopefully) should address it.
Just wondering if anyone can give me a hint on this. After this step, it would be great to understand if the system can get out of its plane (through a pull force out of its plane, simulated in the script through point loads in the joints). I am still not entirely sure about the possibility of doing this. By looking at how other auxetic patterns have been used to generate freeform surfaces, I am giving it a try.
Thank you
Claudio
PS: I noticed also this post and this, really interesting. I see the problematic over the stability and the necessity to separate the states with an energetic hill in the first, as well as some potential in using auxetics in the latter.…
opening a simple file with 30 curves being lofted took like 2 minutes to complete and Rhino crashed afterwards saying:"Windows is out of memory and Rhino will close after you click ok."evethough I still had 7GB of free physical memory and my page file is set also to 16 GB just to be shure...I then switched to Rhino 5.0 Version 5 SR14 64-bit (5.14.522.8390, 05/22/2017) which also had big problems to display the lofted surface. It was unresponsive after loading the file for a minute and a half and then it normally displayed the lofted surface. Every move of camera takes at least 10 seconds to update, but at least it runs. GH profiler says the loft took only 12 ms (90%).
So I'm suspected my graphics card, because the Windows are just three weeks from a clean install. I've also updated my Graphics Driver from the stock Windows one to Intel HD one, but nothing changed.Is there something I'm missing??? What can I try next?My specs:CPU: i5-3320M @ 2.60 GHzRAM: 16 GBGPU: Intel HD Graphics 4000, driver: 07.04. 2017, version 10.18.10.4653
…
Added by Šimon Prokop at 10:39am on October 21, 2017
up structural systems in the parametric environment of Grasshopper. Participants will be guided through the basics of analysing and interpreting structural models, to optimisation processes and how to integrate Karamba3D into C# scripts.
This workshop is aimed towards beginner to intermediate users of Karamba3D however advanced users are also encouraged to apply. It is open to both professional and academic users.
Earlybird (until May 10):
Professional EUR 750 (+VAT)
Educational EUR 375 (+VAT)
Course Fee:
Professional EUR 825 (+VAT)
Educational EUR 415 (+VAT)
Course Outline
Introduction & Presentation of project examples
Optimization of cross sections of line based and surface based elements
Geometric Optimization
Topological Optimization
Structural Performance Informed Form Finding
Understanding analysis algorithms embedded in Karamba and visualising results
Complex Workflow processes in Rhino3d, Grasshopper3d and Karamba3d
Places are limited to a maximum of 10 participants with limited educational places. A minimum of 4 places are required for the workshop to take place.
The workshop will be cancelled should this quota not be filled by May 31st.
The workshop will be taught in English. Basic Rhino and Grasshopper knowledge is recommended. No knowledge of Karamba3D is needed.
Participants should bring their own laptops with either Rhino5/Rhino6 and Grasshopper3d installed. A 90 day trial version of Rhino can be downloaded from Rhino3d.
Karamba3D ½ year licenses for non-commercial use will be provided to all participants.
…
e and i get it. If you have time check the attached papers we published a while ago in relation to the contribution of thermal mass in the reduction of temperature in residential buildings. See the nice contribution of the heavy TM or the lower one for light TM.
As for the solarHeatCapacity, your description (of the 50W) is derived on a 1 Facade/Floor ratio and fully glazed. The only way to reduce it is to increase the ratio (bigger facade area). Which is not recommended (energy losses), but this is a different issue. So, roughly, we can say that 50 is the lower value. If i have less glazing area this number will be higher (right?)
I want to define a value list of "architectural situations", so it is easy to explain and understand. One situation can be:
"Ratio facade/floor 1 & Fully glazed" = 50
"Ratio facade/floor 1 & Half glazed" = 75
"Ratio facade/floor 1.5 & Fully glazed" = 30
"Ratio facade/floor 1.5 & Half glazed" = 50
"Ratio facade/floor 0.75 & Fully glazed" = 70
"Ratio facade/floor 0.75 & Half glazed" = 90
Makes sense for you something like this?
I also defined a value list for the timeConstant like this:
Light Building (Mobile home) = 1Medium-light building (Cement tiles on floor) = 4Semi Heavy Building (Concrete floor + Tiles) = 8Heavy Building (Concrete floors/ceilings + Heavy external and internal walls) = 12
As for the first 5-10 cm effective TM in general my assumption is that you take half of the mass to your space and half to the space above/below you. Will be interesting to do a parametric study on just the thermal mass, uninsulated and insulated to see what the depth limits effectivity will be. Interested in doing such a study together? Can be a nice work even for publishing.
Thanks a lot ... again,
-A.…
393&xg_source=activity
In this case we see a geometrically approach, which doesn’t works efficient, because it required knowing how they behaviors together before, and I think it is not the ‘really behaves’.
To make the structure ‘really behaves’ I tried use kangaroo and the result works very well! As you can see I simply give the 2-set reverse UForce, and then they start to rotate until they found their equilibrium. That means 90 degree rotation. I was wondering what we can do to make a endless-rotation. I am mean 360 degree or more like this:
https://www.youtube.com/watch?v=4owFczeqqMQ
By the way, I try to give supports which allow a horizontal movement only (Just curious how we could keep the anchor-movement horizontally and in the same layer, for example like usual supports for compression ring…). I use the AnchorXYZ, but Kangaroo-Engine seems don’t accept its output.
So maybe some one knows a better solution?
…
Added by Jon to Kangaroo at 7:40am on March 11, 2014
problem later) to fit more shapes that are otherwise won't fit in.
On the example below horizontal rectangle couldn't fit in but its rotated analog could and thus was placed in.
Later, when placed shapes are used to generate frames, because of this rotation, the position of the starting points changes and because of the approach I use to generate the frames some angle values are attached to the wrong corners, this brakes the frame shape and looks like this (on the left the frame of sick shape and on the right the frame of the healthy shape):
Again, this happens because the angle values are assigned to the specific corners (points) and previously rotated shapes get these all messed up:
Easy fix, don't rotate the shapes, problem is, I've already baked a good number of them for later use. I'd like to avoid regeneration because it takes a lot of time and without rotation I constrict the algorithm even more.
Better fix, use a different approach, this is where I'd like to hear suggestions and kicks in a right direction. Please take a look at my definition. It works but I have a feeling like giving an amputee a job of sweeping the floor.
…
creating the structural frame, finding the endpoints, linking these endpoints with curves and afterwards lofting the surfaces between the curves.
The results were quite nice, however, the procedure is very time consuming and inefficient. There is just too much copy-pasting involved.
(see attached file: "Old Attempts.zip" )
Mesh relaxation:
I have later on used Daniel Piker's tutorials on Mesh Relaxation and realized that this might be the way to go.
The link to these online tutorials on wewanttolearn.net is:
https://wewanttolearn.wordpress.com/2011/10/22/mesh-relaxation-kangaroo-tutorial/
His tutorials, however, only deal with mesh boxes which are ideal cubes. He then joins them together in various directions, but it is under 90 degrees angle.
( see attached file: "Daniel Pikers Examples" )
What I would like to achieve:
I want my bridges to go in all directions and angles, not just under 90 degree angle.
Ideally I would like to make a square (polygon) follow a curve (which moves in all axis) at certain number of division points. I would then loft these squares into a mesh and use that shape as a mesh box. I would later use this mesh box and relax it the same way as Daniel Piker used the cubes in his tutorial. The anchor points are only the vertices of the squares which create the lofted mesh box.
( see attached file: "New Attempts" )
As you can see below this procedure works even if the curve is moving in all directions not only along xy axis. There are, however, many problems connected to it.
The problem:
Despite all the effort I cannot seem to come up with a design where I would be able to draw a random curve which would be the guideline for my mesh box and then apply this box to one definition in order to relax the mesh and create the shape that I want. Without this I am again forced into a lot of copy pasting as the final mesh box is made out of several sections.
Also is there any way I could make the final resulting mesh a bit smoother? Increasing the number of mesh faces is probably the only way, right?
Thank you guys so much for any potential help.
All best,
Luka
…