Hello blender community, i am an architecture student from greece, age 24.
So I am here to make a proposal for a future blender edition or plugin.
an introduction :
The architecture community is a great software consumer. Software in architecture was introduced around mid 80s, with tools simulating paper drafting, named CAD programs. This kind of software was eventually developed at a good level, to take over almost every architecture office by around mid 90s.
(lame video, but just to get an idea of precision needed)
The same period some people gained very good familiarity, and started drafting in the 3d space, and rendering, making previsualizations of their buildings. Today, 3d rendering is a must, to all arch. projects.
Later on there were some avant guarde movements, stating that computer as tools, were much more than that, and so proposing, a shift to the next level of architecture design, now called parametric architecture. Implementing scripts (a first for architecture) they managed to design complex geometries, that until then were very tough to design and compute in static drafting.
This is basicly, modeling relationships between objects in the 3d space, and beeing able at any time to change this relationships, producing at every time something else based on the same core formula. So now you have the process (the core formula), that accepts the parameters as inputs, and eventually producing an output. Then the output can be processed as a list, categorized and organized, and produce exact fabrication plans, that can go straight to the factory for production (File to factory, also check CNC , laser cutting, and 3d printing).
an example : instead of modeling a static cube, design 6 points (vertices), and tell the software to make faces between them, so to have a cube. And so parametric design, refers to this process of building relations, so that at anytime we can move these points, and so make a different shape, with real time results. ex, move the top 4 points, 2 units upwards etc, and then take all the faces, and apply a circular hole to them, with lets say the parameter of the radius to be proportional to the distance of the center of the face to an external point, and moving the cube points, soft recomputes both the shape and the holes and we areable to see that in real time, and at the end to take all these faces, make blueprints (plans) and email them to the factory, for a machine to cut.
(this is an actual practice in architecture now, and some factories, for example those who fabricate metal do this, but this is also extended to wood cutting, and even smaller model production in 3d printers etc)
Upon this (very simplified) theory, stands parametric design, wich gains more and more architects as time goes by.
In any architecture office, the basic software is Autodesk Autocad, a program that enables drafting in a vectorial environment (note VE), and costs around 2-4 grands. This is a static drafting tool. then you need 3d rendering soft, most use 3ds max (also by autodesk) that also costs much.
Some years ago, the arch software industry, realized the need for the above and invented BIM CAD soft. that enables some relations between objects, though far away from parametric design.
In need for parametric design software, the worlds bigest arch. firms, tried to invent their own soft. so that they could actually design and build the ideas that they had in mind.
Namely, Foster and Partners of London, collaborated with Bentley (car industry), and smart geometry group, and based on Bentley microstation, they designed Generative Components. For it's greater life, GC was closed, untill now when it became free (who knows until when), but with scarce user base, and very limited educational sources.
The 2nd office is Gehry and partners, of US, that created Digital Project (not a freeware), based on Dessault's Airplane soft Catia 5 (on catia boeing 747 was designed). This has even less user base and support.
Another visionary architect, Kas Oosterhuis (ONL), of Holland, begun from the mid 90s to use the Game production soft. Virtools by 3dVia. The advantage of Virtools is the ability to use game production based relationships, triggers, etc, making the real time change
The next best thing, was Grasshopper, an open source plug in for Rhino (created 4-5 years ago), which enabled, free and easy, parametric design, with the userbase that comes with open source soft. GH gains more and more friends over time, and has been collaboratively developed with users producing modules (called components), in C# or Vb that the base software supports. The disadvantage of GH is that being a plug in, the model it creates, is not a Rhino object, before it's been baked, and thus being static. Also rhino from itself cannot animate. It can animate the camera with tweekings, and with fur more tweeking animate the GH model. Also a physics engine called Kangaroo is now making it's way in GH, in it's first beta version. So in conclusion GH can be used in parametric design, but it's difficult to animate it's model, and has a long way ahead to go.
(example of grasshopper design)
(the prototype physics engine)
In the lacking of every soft's defects, some schools and people resort to Autodesk (again) Maya,that enables both animation and also script, but these are also tweeks, though maya was made for other stuff and not for buildings, making realization of the design difficult.
So my conclusion is, that observating Blender, it has a bunch of capabilities, that others dream of, and in a very compact, though difficult environment.
I think that you programmers, can with some way, implement parametric relations between obects in blender, and extend blender to make the most powerfull parametric design software on the market.
PD, is the present and the future of architecture. If you could win architect users blender user and developer base would double easily.
The basic needs, is the ability to draft in a vectorial background, on global coordinates, and real time line and curve modeling, with precision. Then relations could be easily programmed, to control and create geometry.
Seeing the current blender, i think that it's not so far from what i described because it has all it needs, it just lacks precision drafting modes.
Thanks for hearing, and for any considerations.
if anyone want to contact me do that at pantelisadiavastos [@] hotmail.com
hope you find my proposal interresting.