h tubes are redundant so surfaces overlap instead of interpenetrate, so it is not a good system.
Cocoon is the best answer these days unless you can get Exowire/Exoskelton to work. If you want more control over shape, feed your uncapped tubes into Cocoon as meta-surfaces and delete any and all of the inner meshes to just keep the outer single closed one, but this is just duplicate-culled lines used as meta-lines:
Turn down the CS input to 0.005 for this result, from 0.02 used for faster preview. In fact bake the lines and only test Cocoon on a few of them in order to get the result you want before doing the whole thing.
Whole thing at 0.005 cell size takes 5 minutes for Cocoon and 2 minutes for refinement to a smooth and even mesh.
Actually, seems like 0.005 is way too fine, giving a 600MB STL file.
So, 0.01 cell size at less than a minute total:
159MB STL which is still a bit too big for places like Shapeways. Wow. OK then 0.02 cell size, but I have to increase diameter or my two smoothing steps in refine collapse things too much, an in fact I set it to no smoothing, getting more volume and a reasonable 46MB STL file:
Alas, now it's more frail and overly organic rather than mechanical. Increasing diameter just merges it into perforated plates too much. File size is simply an issue with this complexity level, so different 3D printing services will have different file size limits.
Exowire/Exoskeleton would work but your original mesh hasn't been MeshMachine remeshed to be regular, so short segments ruin it. Here is just a corner:
I think that's why more wires fails, at least. Pretty temperamental component.
Switching to MeshMachine is needed, I guess, instead of Cocoon refine, to remesh away so many small triangles along the boring tubes. Crucial for good remeshing was to set Flip to 0 or I failed to get a rough enough mesh.
It's an adaptive mesh so I can retain good detail while roughing out the tubes.
MeshMachine is terribly slow for this whole thing, like 6 minutes, and blows up for this overly rough setting, 20 steps, so less rough, ugh, I'm out of time. I think free Autocad Meshmixer is the way to make a better smaller mesh, after a refined output from Cocoon. MeshMachine is just too slow to tweak and when it blows up, creating massive triangles jutting out, it hangs too when you change settings.
Starting with a Cocoon refined mesh certainly helped Meshmixer. Using triangle budget lets me have full control. Here is 150K triangles instead of 200K:
STL file size down to 40MB. I think Shapeways is 70 or 100MB limit? So it can be even finer. Here is the Cocoon output versus the Meshmixer reduction:
To use Meshmixer, turn on View > Show Wireframe, Command-S to select all and use Edit > Reduce from the palette that appears.
Cocoon can end up making a few inner meshes where things get weird in your uneven original mesh with small holes so fish out the main mesh by adding a List Item node.
The best strategy for Cocoon is indeed to make an overly fine STL so you avoid any need to tweak forever in Grasshopper, but then you can achieve a smaller mesh file size while preserving shape instead of things turning all smearly organic in Grasshopper.…
presentar Digital Process: Generative Design Technologies Workshop; Taller especializado que se llevara a cabo en 4 de las ciudades mas importantes de la republica mexicana [Puebla] [Mexico DF] [Guadalajara] [Leon] en Enero y Febrero de 2012.http://gendesigntech.wordpress.com/
Enfocado principalmente a arquitectos, diseñadores industriales, diseñadores de interiores, Urbanistas, Artistas digitales, estudiantes y profesionistas afines al diseño; este Workshop tiene como objetivo proporcionar a los participantes los conocimientos y recursos tecnológicos que les permitan desarrollar los elementos de un proyecto desde la concepción hasta su aplicación de manera completa.Apoyándose en un conjunto potente y flexible de plataformas, los participantes aprenderán a generar, analizar y racionalizar morfologías complejas, formas orgánicas libres y algoritmos computacionales avanzados así como a producir visualizaciones fotorealístas aplicables en diversos proyectos de Diseño.A lo largo de 5 dias de intenso trabajo, exploración y retroalimentación los participantes seran guiados en el desarrollo de un flujo de trabajo mas dinamico, que les permitira explotar al maximo el potencial de las herramientas y potencializar sus habilidades, aptitudes y capacidades.Instructores:Leonardo Nuevo Arenas [Complex Geometry]José Eduardo Sánchez [DesignNest]Daniel Camiro/Luis de la Parra [Chido Studio]http://issuu.com/chidostudiodiseno/docs/digproworkConoce el programa aquí.http://gendesigntech.wordpress.com/program/Para registrarte por favor visita.http://gendesigntech.wordpress.com/registro…
tar Digital Process: Generative Design Technologies Workshop; Taller especializado que se llevara a cabo en 4 de las ciudades mas importantes de la republica mexicana [Puebla] [Mexico DF] [Guadalajara] [Leon] en Enero y Febrero de 2012.http://gendesigntech.wordpress.com/
Enfocado principalmente a arquitectos, diseñadores industriales, diseñadores de interiores, Urbanistas, Artistas digitales, estudiantes y profesionistas afines al diseño; este Workshop tiene como objetivo proporcionar a los participantes los conocimientos y recursos tecnológicos que les permitan desarrollar los elementos de un proyecto desde la concepción hasta su aplicación de manera completa.Apoyándose en un conjunto potente y flexible de plataformas, los participantes aprenderán a generar, analizar y racionalizar morfologías complejas, formas orgánicas libres y algoritmos computacionales avanzados así como a producir visualizaciones fotorealístas aplicables en diversos proyectos de Diseño.A lo largo de 5 dias de intenso trabajo, exploración y retroalimentación los participantes seran guiados en el desarrollo de un flujo de trabajo mas dinamico, que les permitira explotar al maximo el potencial de las herramientas y potencializar sus habilidades, aptitudes y capacidades.Instructores:Leonardo Nuevo Arenas [Complex Geometry]José Eduardo Sánchez [DesignNest]Daniel Camiro/Luis de la Parra [Chido Studio]http://issuu.com/chidostudiodiseno/docs/digproworkConoce el programa aquí.http://gendesigntech.wordpress.com/program/Para registrarte por favor visita.http://gendesigntech.wordpress.com/registro…
h, and using the BScale and BDistance are creating havoc somehow too. I've simplified first, and used the Kangaroo Frames component along with setting internal iterations, to make MeshMachine act like a normal component, along with releasing the FixC and FixV. The FixV didn't make any sense anyway. I've also set Pull to 0 to speed it up during testing, since much less calculation is involved to just let the meshes collapse, prevented from disappearing altogether by using a mere 15 iterations.
Also, your breps are open so that allows much more chaos and then collapse, though they did manage to close themselves too at times. Here is closed breps with a full 45 iterations:
So now that it's working, lets re-Fix the curves, and the problem arises that there is an extra seam line that is getting fixed too, running along the cylinder, stopping the mesh from pulling tight under tension wherever a vertex happens to be near that line:
So lets grab only the naked edge curves instead:
And what happens if we lose the end caps, now that we don't have an extra line skewing the result?:
There is no real curvature differences since it's not a curvy brep so the Adapt at full 1 setting has little to do. Now what does the BScale and BDist do? Nothing! Why? Your scale is out of whack, 99 mm high cylinders but only a falloff maximum of about 5, so let's make the falloff be 25 instead, but I must restore the end caps or the meshes collapse away for some reason and freezes Rhino for a minute or so the first time I try it:
It's a start.
If I intersect the cylinders, nothing changes, since they are being treated as separate runs. MeshMachine outputs a sequence of two outputs though, due to Frames being set to a bare minimum of 2 needed to get it to work, so I filter out the original run, which is just the unmodified initial mesh it creates.
The lesson so far is that closed meshes are much less prone to collapse and glitches leading to screw ups.
A Boolean union of the cylinders is when it gets funner, here show with and without the fixed curves that seem to define boundaries too where really there are just polysurface edges:
…
input orientation of the objects. I can see that you've already done this with Vec2pt. Doing it with a sun vector is a little easier, because you are working with one vector, not a bunch of different vectors. you probably know a lot of this already, but I wanted to write a comment that is helpful to anyone coming across the discussion, because it is a common design task.
To orient a bunch objects towards a sun vector:
1. you need a vector to represent the sun's rays. You can either use an existing definition from the web (definitely look at Ted Ngai's amazing work on this), or just make a single adjustable vector as a stand in. I've often simply made a vector using azimuth and altitude angles as inputs, since those are common ways of describing the location of the sun, and makes it easy to look up a sun angle and put it in to your definition.
2. assuming you have some vector to represent the sun's rays, make a plane that is perpendicular to this vector. But Why, Precisely?, you're already familiar with some of the quirks of making a plane perpendicular to a vector, just keep those quirks in mind.
3. next, create reference planes for your panels. If your panels are flat (i.e. planar) this is really simple, just make a list of their planes, using whatever you like (check planarity, evaluate surface, whatever). If your panels are not planar, then you need to decide on a plane you can make from each one that you would like to use as a reference plane. plane from 3 points might be a good method here.
4.take your single plane that is perpendicular to your single sun vector, and place it at the origins of all of your reference planes. Now you have a sun-oriented plane for each panel.
5. Using the orient component, input your reference planes as the reference planes, and input your sun-oriented planes as the target planes, and input your panels as the geometry to transform. You should now have a bunch of panels oriented to the sun vector.
6. In this method, I've assumed that you want your panels perpendicular to a solar vector, to face the vector, but if you want a different relationship to the sun vector, you just need to change the relationship of that single first sun-oriented plane to whatever relationship you would like to make.
One thing to think of when designing for sun angles is just that at any given point in time, for any given point on the earth's surface, the rays of the sun are basically parallel. the angle of these rays changes over time, but at any other time, the rays are still parallel to each other, and can therefore be described by a single vector for each moment in time.…
diseño computacional.
La Visiting School digitalMed 2014, promovida por Medaarch y Emwesoft Sevilla S.L.N.E, se celebrará en la ciudad de Sevilla, y tendrá como tema central la Smart City y el estudio de la interacción entre las personas y su entorno a través de objetos, dispositivos e infraestructuras.
Fecha limite de inscripción: 16/01/2014
info@emwesoft.com
OBJECTIVOS Adquirir la capacidad de gestionar flujos de datos en los que las ciudades están sumergidas, para insertar proyectos que sean útiles, contextualizados, poco invasivos y aptos a establecer un intercambio de informaciones con los usuarios.
El objetivo final es redactar un catálogo de proyectos que puedan formar parte de un contexto urbano y puedan delinear el perfil de las ciudades en las que viviremos en el futuro próximo.
METODOLOGÍA Metodología basada en el aprendizaje activo, en la puesta en práctica de métodos activos que estimulan y facilitan el intercambio de experiencias y puntos de vista entre el alumnado: Buscando la participación del alumno, planteando todas las cuestiones que considere necesarias a la hora de aclarar conceptos.
Fomentando el debate y la colaboración entre los participantes.
Dando respuesta a las dudas planteadas.
La metodología será presencial, lo cual permite un mayor acercamiento entre profesor y alumno, y en consecuencia una mayor asimilación de los conceptos.
PROGRAMA Los primeros días del taller serán dedicados a establecer definiciones comunes que nos permitan trabajar a partir de significados compartidos. En esta fase se tratarán temáticas que recurren a menudo en la práctica arquitectónica contemporánea, es decir el diseño computacional, la fabricación digital y los data driven. Los alumnos tendrán la posibilidad de aprender a usar software para el diseño paramétrico, como Rhinoceros y el plug-in Grasshopper, a través del conocimiento de dichos software, el alumno conseguirá competencias teóricas y técnicas, para un enfoque al diseño computacional.
PROFESORADO La formación será impartida por profesionales con amplio conocimiento y experiencia en el ámbito. Los tutores serán los arquitectos Amleto Picerno Ceraso y Francesca Viglione.
DURACIÓN TOTAL DEL TALLER
40 horas
QUIÉN PUEDE PARTICIPAR?
. Funcionarios con una actitud proactiva hacia la construcción de ciudades inteligentes;
. Académicos y estudiantes en áreas relacionadas con el desarrollo de proyectos y soluciones tecnológicas para ciudades digitales y ciudades inteligentes;
. Arquitectos;
. Ingenieros;
. Diseñadores;
. Profesionales de las tecnologías de información y con relación a el área de tecnología.
REQUISITOS BÁSICOS
- Conocimiento básico de Rhinoceros
- Inglés medio
*Disponibilidad de un intérprete español.
PRECIO y Tarifa especial
El cuesto del taller es de 500€.
También hay facilitacióno en caso de Inscripciones de grupo: para cada grupo formado por 5 inscriptos, que paguen en un única solución, el costo total será de 4 miembros y no 5 (una persona gratis)
DONDE
Emwesoft Sevilla S.L.N.E C/ Monte Carmelo 21, 41011 – Sevilla (España)
Teléfono: +34 (955) 224 524
Email: info@emwesoft.com
Internet: www.emwesoft.com …
Data matching is a problem without a clean solution. It occurs when a component has access to differently sized inputs. Imagine a component which creates line segments between points. It will have…
ll geometry.
The difference with programs like Inventor is that they are made for production, regardless of the fabrication method. I won't go into detail about that, and instead focus on the modeling process.
In this little model, the starting point actually is a bit obvious, the foundation.
The only contents in the 3dm file are 27 lines. These indicate the location of each footing, and the direction of the tilt of each column. Everything else is defined in GH with the use of numbers as input parameters.
Needless to say, instead of those lines you could obviously generate lines and control the number of columns and panels, hence establish their layout, with any algorithmic or non-algorithmic criteria you please. That marks a major difference between GH and Inventor.
You can generate geometry with Inventor via scripting/customization (beyond iLogic), with transient graphics for visual feedback similar to GH's red-default previews. However Inventor's modeling functions are not set to input and output data trees. I won't go into detail on that, but suffice to say that the data tree associativity of GH was for me the first major difference I noticed. I've used other apps with node diagram interfaces like digital fusion for non-linear video editing since the late 90's, so the canvas did not call my attention when I first started using GH.
Anyways, here's a screen capture of the foundational lines:
In the first group of components, the centerlines of the rear columns are modeled:
And the locations in elevation for connection points are set. Those elevations were just numbers I copied from Excel, but you can obviously control that any way you please. I was just trying to model this quickly.
The same was done for the rear columns:
The above, believe it or not, took me the first 5 hours to get.
Here's a screen capture of what the model and definition looked like after 4 hours, not much:
If you're interested, next post I can get into the sketching part you mentioned, which is a bit cumbersome with GH, but not really.
I wouldn't say that using GH to do this little model was cumbersome, it just needed some thinking at the beginning. You do similar initial thinking when working with a feature-based modeler.…
Added by Santiago Diaz at 12:44am on February 24, 2011
tives for low-dimensional, or highly continuous problems. Having a somewhat faster way to trigger a galapagos run would also be beneficial."
I found a post on the 'hoopsnake/forum' describing the very same problem I am trying to solve, and looked into using HoopSnake (without satisfaction so far):
Double loop and hydrostatics?
I don't want to wait until G2 so will re-state some of what I posted earlier, then offer a template for an ideal "fast solver" component ('B-Solve') that could be widely useful. I am ready to accept that it might be written in Python, C, or VB - as long as it's open source or built in to standard GH. If there is a GH plugin that will do this, I'd like to know that too, though prefer a lightweight solution rather than a big toolbox.
QUESTION: Is there a FAST (binary search speed) GH way to "solve" toward a goal by "moving" a single slider?
CONTEXT: I have a boat hull of a given displacement at rest. I rotate the hull to an arbitrary angle ("heel" caused by wind in the sails) and want to adjust a 'Z-offset' slider so the displacement is the same as it was at rest.
I can adjust the slider manually, zooming in for better control, and with a dozen tries or so, in a very short time, narrow in with a binary search method and get very close to matching the value.
When I hook up Galapagos, it runs on and on forever, trying values that are "obviously" further away instead of closer to the goal. When I can solve it manually faster than Galapagos, a different solution is needed.
OBJECTIVE:
I want a FAST solution that doesn't need any manual input. Ideally, it would respond like any other component and re-calc whenever its inputs changed. At worst, a 'start/reset' trigger, "soft input" so it can be used inside a cluster.
It doesn't need to control a slider, they just happen to be handy for defining a range and precision of values.
Using Galapagos: HydroSolve_2015_Sep8a.gh (attached)
An extremely stripped down version of the problem using Grasshopper.
NOTE: One obvious problem here is that by using absolute value ('abs()') for the 'difference' here, Galapagos doesn't know whether it's too high or too low!
Instructions:
Start with 'Roll=0', 'Volume=1543.943'
Adjust 'Roll' to ~35 degrees
"Solve" 'Z-offset' value to return to 'target' (original) volume of 1543.943
Using 'B-Solve': HydroSolve_2015_Sep8b.gh (attached)
'B-Solve' is the proposed fast solver component. Its 'solution' output is always in the range of zero to one, which is remapped by the green group as -5 to 5 and used as the 'Z-offset' for 'Pitch-Roll-Z'.
Starting value ('Reset') for 'solution' is 0.5, and 'B-Solve' tries different 'solution' values to make 'result' (the 'Volume') and 'goal' match. An efficient uphill(?) or binary searcher could be very fast.
Does this sound feasible? Can anyone implement 'B-Solve'?
Two at once?
The post noted earlier, Double loop and hydrostatics?, brings up a complication that's worth considering from the start... Depending on hull shape, the center of buoyancy may move fore and aft, away from the center of gravity, as the hull rolls. This induces a change in pitch so a second 'B-Solve' component could be used in the same model to adjust pitch, which of course changes 'Volume' again... Not quite sure how the two would get along?
Thanks.
Note: the hull in these examples is a really poor shape!…
Added by Joseph Oster at 1:30pm on September 9, 2015
, Engineer and Researcher from France with broad programming experience. He is the author of the City in 3D Rhinoceros plugin for creation of buildings according to geojson file and with real elevation. Guillaume already created a new component: "Address to Location". It enables getting latitude and longitude values for the given address:
2) Support of Bathymetry data: automatic creation of underwater (sea/river/lake floor) terrain. This feature is now available through new source_ input of the "Terrain generator" component. Here is an example of terrain of the Loihi underwater volcano, of the coast of Hawaii:
3) A new terrain source has been added: ALOS World 3D 30m. ALOS is a Japanese global terrain data. Gismo "Terrain Generator" component has been using SRTM 30m terrain data, which hasn't been global and was limited to -56 to +60 latitude range. With this addition, it is possible to switch between SRTM and ALOS World 3D 30m models with the use of source_ input.
4) 9 new components have been added:
"Address To Location" - finds latitude and longitude coordinates for the given address.
"XY To Location" - finds latitude and longitude coordinates for the given Rhino XY coordinates. "Location To XY" - vice versa from the previous component: finds Rhino XY coordinates for the given latitude longitude coordinates. "Z To Elevation" - finds elevation for particular Rhino point. "Rhino text to number" - convert numeric text from Rhino to grasshopper number. "Rhino unit to meters" - convert Rhino units to meters. "Deconstruct location" - deconstructs .epw location. "New Component Example" - this component explains how to make a new Gismo component, in case you are interested to make one. We welcome new developers, even if you contribute a single component to Gismo! "Support Gismo" - gives some suggestions on how to make Gismo better, how to improve it and support it.
5) Ladybug "Terrain Generator" component now supports all units, not only Meters. So any Gismo example file which uses this component, can now use Rhino units other than Meters as well. Thank you Antonello Di Nunzio for making this happen!!
Basically just forget about this yellow panel:
This panel is not valid anymore, so just use any unit you want.
6) A number of bugs have been fixed, reported in topics for the last couple of weeks. We would like to thank members in the community who invested their time in testing, finding these bugs and reporting them: Rafat Ahmed, Peter Zatko, Mathieu Venot, Abraham Yezioro, Rafael Alonso. Thank you guys!!! Apologies if we forgot to mention someone.
The version 0.0.2 can be downloaded from here:
https://github.com/stgeorges/gismo/zipball/master
And example files from here:
https://github.com/stgeorges/gismo/tree/master/examples
Any new suggestions, testing and bug reports are welcome!!…
Added by djordje to Gismo at 5:13pm on March 1, 2017