this was about some boring building I wouldn't respond ... but here we are talking sardines.
Here's my take on that matter:
1. The 4 C# first create/use a nurbs, then define some random planes (and transformations) and then (a) either they place some humble stripes or ... er ... (b) sardines as instance definitions (NOTE: Load Rhino file first).
2. All important decisions are the ones in yellow groups.
3. You control what you get via this (priority on stripes or sardines? that's the 1M Q):
4. If you decide for sardines (the right thing to do) then you must ENABLE the Sardiniser(C)(tm)(US patent pending) as follows:
5. The vodkaFactor on that Sardiniser C# adds some spice in the sardine placement (it does that by altering the priority on the "composite" transformation in use: first randomly rotate then planeToPlane .... or the other thing?).
6. Only the finest Da Morgada sardines are used in this definition:
7. Spot the WARNING in the filter related with what sardine to choose > do it wrong and no hard disk on your workstation > no risk no fun > sorry Amigos, he he.
8. 1M question for you all: why placing sardines (it's real-time you know) is WAY faster than creating these humble stripes?
9. Although the sardines are placed in real time as regards your CPU ... the critical factor is your GPU (display mode: rendered).
10.Still WIP (dancing sardines in the next update).
have some sardine fun, best, Lord of SardineLand…
housing for an LED PCB. The object is a parametric series of discs with an opening inside made up of a mirrored curve [drawn in Rhino, mirrored in GH]
It is madde up of N number of discs which can be varied through the distance between the circular outline using a divided curve [straight line in GH]. The length of the object can be varied using a length parameter, and the shape using a graph mapper.
I've chosen to cap the end two discs by creating two sets of outlines. One set has the central aperture cutout for the PCB, whereas the other set is a trimmed circle [achieved using the "trim box" layer profile in Rhino]
I then cull the outer two curves from one array, and the inner N-2 curves from the inner array.
The final outcome I am after is to create the housing as both an STL and a 2d template for laser cutting. This is a learning exercise for me as well as a cool project.
I had it working OK, but then I adjusted the profile for the PCB and joined it and now it is giving me some grief. I am sure the answer is obvious. The problem is the PCB profile is made up of 3 polycurves, whereas the disc profile is one planar curve. I have no problem using the flatten option so there are only N sets of curves coming out of the "Join Curves" component. However when I cull the curves, the planar curves making up the exterior edge cull fine, but the interior curves [the joined, PCB profile] culls in a different [irrational?] order to what I would expect. If I connect the single planar curve to the culls section, it works fine, but the joined line section just won't play.
In the instance uploaded N = 10, based on spacings. And index white it appears that 0 and 9 are diagonal to what I'd expect, although if you fiddle with the values they go all over the place.
Can someone please help me and explain what I did wrong? Files are attached... I have screen grabbed the relevant section, but it is grouped in red and labelled as "problem child" :)
Many thanks for your help, sorry if this looks like a clusterf**k, first time for everything... any advice very much appreciated, not just relating to my problem.
All the best
Nick…
ns about them.
It's a direction for Kangaroo I very much intend to continue developing - and I am still getting to grips with the possibilities and experimenting with how different optimization and fairing forces work in combination with one another, so I would value your input and experience.
For those interested in some background reading material -
[1] http://www.cs.caltech.edu/~mmeyer/Research/FairMesh/implicitFairing.pdf
[2] http://mesh.brown.edu/taubin/pdfs/taubin-eg00star.pdf
[3] http://www.pmp-book.org/download/slides/Smoothing.pdf
[4] http://graphics.stanford.edu/courses/cs468-05-fall/slides/daniel_willmore_flow_fall_05.pdf
[5] http://www.evolute.at/technology/scientific-publications.html
[6] http://www.math.tu-berlin.de/~bobenko/recentpapers.html
[7] http://spacesymmetrystructure.wordpress.com/2011/05/18/pseudo-physical-materials/
[8] http://www.evolute.at/technology/scientific-publications/34.html
[9] http://www.evolute.at/software/forum/topic.html?id=18
At the moment the Laplacian smoothing is uniformly weighted, which tends to even out the edge lengths as well as smoothing the form, which is sometimes desirable, and sometimes not. It also tends to significantly shrink meshes when the edges are not fixed.
I plan to try some of the other weighting possibilities, such as Fujiwara or cotangent weighting (see [1] and [3]), as well as other fairing approaches, such as Taubin smoothing [2], Willmore flow[4], and so on. This also has applications in the simulation of bending of thin shells.
Planar quad panels are often desirable, but I'm finding that planarization forces alone are sometimes unstable, or cause undesirable crumpling, so need to be combined with some sort of fairing/smoothing, but the different types have quite different effects, and the balance is sometimes tricky.
There's also the whole issue of meshes which are circular (I posted a demo of circularization on the examples page), or conical (this one still isn't working quite right yet), and their relationship with principal curvature grids and placement of irregular vertices, all of which is rather different when the whole form is up for change, rather than having a fixed target surface [7].
I'm also trying to get to grips with ways of making surfaces of planar hexagons, which need to become concave in regions of negative Gaussian curvature (see this discussion)
and I hope to release soon a component for calculating CP meshes, as described in [8], which I think could have many exciting construction implications.
While there are a number of well developed smoothing algorithms, their main area of application so far seems to be in processing and improving 3D scan data, so using them in design in this way is somewhat new territory. There can be structural, fabrication or performance reasons for certain types of smoothness, but of course the aesthetic reasons are also often important, and I think there are some interesting discussions to be had here about the aesthetics of smoothness.
Anyway, that's enough rambling from me, hopefully something there triggers some discussion - I'm really keen to hear about how all of you envision these tools might be used and developed.
…
ow the steps of the successful run when step 1.2 is bypassed (note that the and OpenFOAM session is open in the background while running the Butterfly demo file):
1. create wind tunnel, and use different parameters of (4,4) for _globalRefLevel_ as suggested by Theodoro in this post
2. run blockMesh:
3. run snappyHexMesh:
4. run checkMesh:
5. connect the case from checkMesh to simpleFOAM and run the simulation:
6. the simulation converged at 1865 iteration, but the results visualization part has some problem:
7. so I revised this part according to suggestions from Hagit:
8. and the results can be visualized for P and U values:
The GH file used for the successful run shown above is attached here.
Now, the following is the error I got when the case from the update fvScheme component is used for simpleFOAM simulation:
the warning message on the simpleFOAM component is:
1. Solution exception: --> OpenFOAM command Failed!#0 Foam::error::printStack(Foam::Ostream&) in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so" #1 Foam::sigFpe::sigHandler(int) in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so" #2 ? in "/lib64/libc.so.6" #3 double Foam::sumProd<double>(Foam::UList<double> const&, Foam::UList<double> const&) in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so" #4 Foam::PCG::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so" #5 Foam::GAMGSolver::solveCoarsestLevel(Foam::Field<double>&, Foam::Field<double> const&) const in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so" #6 Foam::GAMGSolver::Vcycle(Foam::PtrList<Foam::lduMatrix::smoother> const&, Foam::Field<double>&, Foam::Field<double> const&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::PtrList<Foam::Field<double> >&, Foam::PtrList<Foam::Field<double> >&, unsigned char) const in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so" #7 Foam::GAMGSolver::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so" #8 Foam::fvMatrix<double>::solveSegregated(Foam::dictionary const&) in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/lib/libfiniteVolume.so" #9 Foam::fvMatrix<double>::solve(Foam::dictionary const&) in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/bin/simpleFoam" #10 Foam::fvMatrix<double>::solve() in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/bin/simpleFoam" #11 ? in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/bin/simpleFoam" #12 __libc_start_main in "/lib64/libc.so.6" #13 ? in "/opt/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/bin/simpleFoam"
The error message from the readMe! output node is attached below as a text file.
Hope you can kindly advise what the important steps or parameters I might have missed here. I assume it might be related to OpenFOAM rather than with the Butterfly workflow...
Thank you very much!
- Ji
…
mbre de 9:00 am a 8:00 pm Este taller está dirigido principalmente a arquitectos y diseñadores interesados en el aprendizaje del diseño paramétrico y generativo aplicados a la generación y racionalización de geometrías complejas para su implementación en diferentes procesos de diseño. En el curso se abordarán los conceptos básicos y metodología para hacer frente a diversas problemáticas del diseño mediante el desarrollo de herramientas algorítmicas a través de un lenguaje de programación visual y el desarrollo de esquemas de fabricación digital. No se requieren conocimientos previos de Rhinoceros 3D ni de programación, conocimientos previos de CAD deseables. Estudiantes: 2,500 MXN Profesionales: 3,000 MXN
CONCURSO DE RENDERS - BECA DEL 100% - Parametric & Generative Architecture & Design Grasshopper Workshop.
- Publica tu render en www.facebook.com/3dmetrica - El render con más likes será el ganador. - Fecha límite de votaciones 15 de septiembre del 2012.
Informes e Inscripciones: workshop@3dmetrica.com 04455 28790084 www.3dmetrica.com www.facebook.com/3dmetrica
…
imes. Your loop should go to y.Count - 1. Or, you could use a For...Each loop, circumventing the problem altogether:
Dim shortLines As New List(Of Line)
For Each segment As Line in y
If (segment.Length < x) Then
shortLines.Add(segment)
End If
Next
A = shortLines
--------------------------------
Another problem is this line of code:
New_Lines.Add(New_Line)
It is located inside the loop but outside the If statement, meaning it gets run every single iteration. This fills up the short line list with duplicates.
-------------------------------
Here's something else which is redundant:
Dim Input_Line As New Line
Apart from the fact that you don't need a special variable for this at all, you also don't need to add a New keyword. The type Line in RhinoCommon (just like Point3d, Vector3d, Plane, BoundingBox etc. etc.) are Structures, not Classes. Structures always exist when they are defined, whereas Classes can be null ("Nothing" in VB).
-------------------------------
Some more advice:
Dim i As Integer
For i = 0 To y.Count()
You can merge these two lines into one. VB.NET allows you to declare your iteration variable inside the loop:
For i As Integer = 0 To y.Count - 1
--------------------------------
If you don't like the For...Each approach at the top of this answer, here's how to write this using a For...To loop:
Dim shortLines As New List(Of Line)
For i As Integer = 0 To y.Count - 1
If (y(i).Length < x) Then
shortLines.Add(y(i))
End If
Next
A = shortLines
ps. A personal preference of mine is that I always encase the expressions inside If...Then statements in brackets. You technically don't need to do this, but I find it makes the code more readable.
--
David Rutten
david@mcneel.com
Poprad, Slovakia…
taTree.
2. Since GH is acyclic by design we can't pick individually (without code, that is) our "picks" for the iceberg ... thus we need a global policy applied to ALL grid points at once.
3. This is what the next part does: it picks randomly some iceberg stuff and modifies their Z by a random value. If the Z is always "above" the grid or not it depends upon the domain of values to operate. Seed means "roll the bones again" (meaning another collection).
4. So we have the modified points Data Tree (that are steady - acting as the tips of the iceberg). Let's call them Anchors.
5. If we subtract set 4 from 1 we have the points prone to vary according some manipulation. Kangaroo does that manipulation (this is the best add-on that GH has to offer by 1M miles made by a very clever fella).
6. But if we instruct Kangaroo to do the job... he makes chaos since the points in 4 are not sufficient: we need perimeter steady points that act as Anchors as well. So we manage some logic to pick a variable set of perimeter points and we "merge" 4 and 6 and we have the final set of Anchors on hand - whilst all the rest are points willing to change.
7. Kangaroo is a physics engine meaning that the only thing that understands is ... er ... points and their relation (the "line" connecting them, that is). Kinda like a CPU that understands 0 and 1 and nothing else.
8. So we provide Kangaroo info about all the lines involved: how "stiff" they are and what is the expected/desired final length.
9. By double clicking the Kangaroo component ... the "simulation" starts running (in some kind of "loops") and goes towards an "equilibrium" where all our desires are satisfied - or the solution's entropy is the minimum possible (well up to some level, he he). Kangaroo displays a small control dialog that allows you to halt the process or reset it (meaning: start again).
10. If the instructions are "good"/"proper" the "loops" (iterations) are relatively few: if K does 1M "loops" ... this means that your instructions are silly or not well thought.
After stopping Kangaroo ... we have (hopefully) a "well" distorted collection of points (and their equivalent mesh) to proceed further via components usually found in the WB add-on
PS: If all the above sound Greek to you ... it's because I'm Greek, he he.
Moral: Get the gist of Kangaroo ASAP - worth spending some time I recon. If you do that and you need examples (other than the ones available at download time) ... well I have more than 300 (from simple to ultra paranoid).…
d of interpenetrating surfaces somewhere:
Now all links (except a possible single ball on the very end of odd numbered ball series) are four balls long, including the jostled ones. Without that step, those items simply don't appear in the output, leaving way too big of gaps to ignore, eventually leaving huge gaps at later stages of segment doubling:
So if I turn the jostling multiplication factor way down it should work imperceptibly:
Ta-dah! The jostling strategy WORKS! Granted, only in this special case where I know I'm dealing with adjacent pairs of worms along a curve, not generic objects arranged in space by some artist.
Now I just need to wrap the multiple Python script components I'm stringing together into one script.
How long does the full 2400 balls take, finally? It took 12 Python scripts that merge pairs, to achieve this breakdown: 2400 -> 1200 -> 600 -> 300 -> 150 -> 75 -> 38 -> 19 -> 9 -> 5 -> 3 -> 2 -> 1. Time was 2 minutes 50 seconds, so there is some extra struggle for 2X as many balls as 1200 that took 1 minute 20 seconds, but only ten more seconds.
…
Added by Nik Willmore at 9:06pm on February 17, 2016
Angeles, which has 12% of the year made comfortable, and Shiraz, Iran, which also has 12% comfortable (assuming default parameters).
Jerusalem also makes sense to me. There is only a maximum possible 9% of the year that is inside the polygon (you'll see this if you set the timeConstant to a very high number). The default strategyPar makes 6% of these hours comfortable and 3% without cool enough temperatures in the previous hours. This seems reasonable to me.
I could be convinced to change the default time constant to 12 hours (instead of 8) as I know that 12 is the default of climate consultant but that seemed really idealized in my opinion. You'll need really high exposed mass and insulation without much internal heat gain to make conditions stable for more than 8 hours in my opinion.
As for the solarHeatCapacity, I get changes when I drop it down to 10 W/m2 or boost it up to 100 W/m2. It's definitely a parameter that operates on an "order of magnitude" scale and little tweaks to it won't change it too much. You can think of this number as representative of a lot of other physical properties: most notably the depth of the space being passively heated and the thermal mass of that space's materials that participate in heat exchange over the time constant. Climate consultant uses a default assumption of 30 W/m2 but, from my calculations, this is likely assuming a space that has a facade to floor area ratio that is greater than 1. If we say that we need to raise the temperature of 10 cm of an exposed concrete floor for passive heating purposes, and we have a facade-to-floor area ratio of 1:
Required solar flux = ((1 facade-to-floor ratio) x (0.1 m3 of concrete) x (2400 kg/m3 concrete density) x (880 J/kg-K concrete specific heat capacity)) / 3600 seconds/hour
This lands you with a required solar flux of 58 W, which is almost twice the 30 W climate consultant default. While me might say that not all 10 cm of concrete participates over the course of a default 8-hour time constant (most of the action is probably within the first 5 cm), we also have to account for things like transmittance of solar though the window, which, for triple pane, is probably only half of the incident solar. So 50 W seemed to be a more reasonable rule of thumb from my perspective, essentially assuming a facade-to-floor ratio of roughly 1 with 5 cm of concrete participating in an 8 hour heat exchange and a little more than half of solar heat getting through a fully glazed window.
Let me know if that makes sense or if you have any suggestions,
-Chris…