Generative Design

Last post

2010-10-18T03:00:00.001-07:00

This is the last post on this blog - as the name is no-longer relevant.

Will now be posting from www.digitalmorphogenesis.blogspot.com

Stadium experiment

2010-09-30T10:55:00.000-07:00

Finally done with that time-consuming presentation.

Going to work on a little experiment over the next two weeks, investigating the Revit SKD, as the work in progress, will try to model this form parametrically, hopefully leading to an interesting stadium

4.A habit that you wish you didn’t have

2010-09-19T21:15:00.000-07:00

Staying up too late.
Eating junk food.
Getting caught up in side things.
Getting too frantic about things.

30 days, Q2

2010-09-19T04:39:00.001-07:00

What is the meaning behind the blog name?

The first part is my name, the second part was a modelling systems that I was interested in. Never meant for it to be my main blog - it was supposed to be just a record of modelling with L-systems which are generally used to model plant growth.

Have been adapting them for my own design thesis, so the name stuck.

2010-09-18T04:09:00.000-07:00

15 Intestesting Facts about myself.

1. Born in the same city as Nelson Mandela.
2. The country (Transkei) no longer exists.
3. Spent 10 years at uni.
4. Watch way to much anime.
5. Interested in 3d CAD modelling.
6. Played badminton and soccer in high school.
7. Captain of the chess team in high schhol.
8. Do programming as a hobby, fav language is c#.
9. Favourite architect is Santiago Calatrava.
10. Favourite game - Star Craft.
11. Favourite chess opening, French Defense.
12. Been to all continents except south america and antarctica.
13. Interested in the philosophy of Gilles Deleuze.
14. Played the violin for awhile.
15. Favourite food - Japanese & Italian

30 Days of Me

2010-09-14T21:31:00.000-07:00

Well this blog was all about science experiments - but have been talked into answering these questions, over 30 days or so.

A recent picture of you and 15 interesting facts about yourself
The meaning behind your Blog name
A picture of you and your friends
A habit that you wish you didn’t have
A picture of somewhere you’ve been to
Favorite super hero and why
A picture of someone/something that has the biggest impact on you
Short term goals for this month and why
Something you’re proud of in the past few days
Songs you listen to when you are Happy, Sad, Bored, Hyped, Mad
Another picture of you and your friends
How you found out about Blogger and why you made one
A letter to someone who has hurt you recently
A picture of you and your family
Put your iPod on shuffle: First 10 songs that play
Another picture of yourself
Someone you would want to switch lives with for one day and why
Plans/dreams/goals you have
Nicknames you have; why do you have them
Someone you see yourself marrying/being with in the future
A picture of something that makes you happy
What makes you different from everyone else
Something you crave for a lot
A letter to your parents
What I would find in your bag
What you think about your friends
Why are you doing this 30 day challenge
A picture of you last year and now, how have you changed since then?
In this past month, what have you learned
Your favorite song

Grass

2010-09-14T20:39:00.000-07:00

Playing around with MAX to show some students how to render grass courtesy of Jeff Patton.

BSP Trees Part 3

2010-09-03T21:42:00.003-07:00

Been on hold for more than half a year - continuation of using BSP trees to hold triangles.

Using a Binary Space Partitioning tree to classify points inside an icosahedron. One of the preliminary steps before doing solid boolean operations. On the 2D side of things, have been thinking that since BSP trees point location method only works when the normals to the polygon are correctly working, need to implement a method that corrects this based on an alternate method where you cast rays.

Rectangles

2010-08-26T09:35:00.000-07:00

Following the recreation of the Reptile experiment, which I thought was quite successful, I added rectangles in the same manner that I did for tetrahedra except that the reader has to look for 3 recorded points instead of two - will make some modifications later so that the rectangle can be non-orthogonal. Next step is to bring the overlaid grid back in as with the posts from about a month ago.

Skyscraper

2010-08-22T07:50:00.000-07:00

Slowly coming together

photon mapping

2010-08-22T07:43:00.001-07:00

Might have gotten photon mapping working - kinda hard to tell - need better filtering and colour correction and scaling before it really shows through. Will be putting this on hold for awhile, while I play with the skyscraper model.

sphere packing 2

2010-08-16T03:26:00.000-07:00

Gave Maria a hand modelling one of the structures in 3dsmax, attached a quick render. Nice thing about cg rendering is how easy it is to do metals (and plastics) especially on spheres.

Reptile revisited

2010-08-15T08:05:00.000-07:00

As there seems to be a lot of playing with triangular grids and sphere packing... it seemed a good idea to finish up a project that I've been interested in for awhile - to translate John Frazer's Reptile (repetitive tiling) project into the grammar I've been working on. Its fun to watch the structure grow - but slows down quite a bit after just 5 iterations. Not that my code is in any way efficient.

Anywho, happy for now - one big hurdle ahead - reverse engineering plans into grammars and then I'll be in a good position to stop experimenting and start planning out the main points of the thesis.

Cornell Box

2010-08-11T00:42:00.000-07:00

Constructed a small version of the Cornell box - this seems as good as I can get it with Raytacing alone. Going to try photon mapping next.

walls

2010-08-07T08:02:00.000-07:00

An ongoing problem is to turn ifcwalls into centrelines. Tricky because the wall is usually defined as the extrusion of a polyline - which isn't always a rectangle. Got most of it working - the next part is to extend lines so they intersect with other lines and we can get a graph out of it. Using a search minimal cycles, they can then be divided into rooms.

Raytracer 4

2010-08-07T04:16:00.000-07:00

Cleaned up the code by adding in interfaces on the shape classes. The algorithm looks quite concise now - just about 30 lines or so. I'd still like to know more about the science of light and color. On my things-to-do list was to read up on BRDF.

Diffuse and Specular

2010-08-05T05:29:00.001-07:00

Added Diffuse and Specularity to the Raytracer - can't believe how simple those are to implement - the diffuse factor is the dot product of the lights normal with the surface normal (which is the same as the Cosine of the angle) and the specularity is the dot product of the reflection of the lightvector and the vector of the camera to the point raised to the power of the 'shininess' value used to describe the material.

When I first saw these values in studio max about 2004, they seemed so complex. Thats why raytracing is fun - it can start out very simply and gradually build up complexity.

The image on the blue screen is made of points (what digital image isn't) though in this case they are OpenGL 3D points.

Fun with IFCs

2010-08-03T21:31:00.000-07:00

Still need to extract properties and metadata

Raytracer part 2

2010-08-01T05:12:00.000-07:00

Added triangles for the raytracing program. Whats missing before something relatively decent can come out are some photometry calculations and secondary bounces.

Raytracer

2010-07-31T06:42:00.001-07:00

Took a break from the architecture models. One thing I've always wanted to do since I first played with POVRAY (about 1998?) was to understand raytracing better. In 2nd year of uni, I learnt Radiance and the first time I rendered a room, it was like magic seeing something photo-realistic made from scratch. In this case, I'll be using my own libraries and classes - and also have the bonus that via OpenGL I can visualize the scene before rendering!

IFC SPACE AND WALLS

2010-07-26T05:26:00.001-07:00

aaaargh... spent ages on the problem of triangulating a polygon just because I didn't realise I was sending the wrong list to the Binary Space Partition. A few letters difference wasted so much time :'(

Anyway, am happy with how this looks, have tried it with many different configurations and they all seem to work.

Delaunay

2010-07-24T06:18:00.001-07:00

Continuation of the previous post - triangulating a concave polygon. There are more triangles than what I'd like. Probably got some tweaking and cleaning to do, but for the most part am happy that its this far at least.

Actually with BSP trees choosing the root node is quite important for efficiency, in the case below I've just set it as the first line of the list of edges.

Point in Polygon

2010-07-24T03:12:00.001-07:00

For a very long time have been meaning to clean up this code. It uses a binary space partitioning tree to represent a polygon with the added use that it can classify whether a point is in the polygon very quickly. Below is 100,000 points classified in about 0.1s.

Now just need to adapt this for 3D polyhedra which is a stepping stone to 3d boolean operations on meshes.

The reason this actually came about though, was that I needed to cap the 'Extruded solids" that was coming from the IFC files in the post below. For some reason could not get the OpenGl Glu's Tessalotor working and am not sure if its another one of those 64-bit problems as I can't get an debugging errors.

Plan representation

2010-07-22T05:54:00.000-07:00

The state of the representation scheme that I'll be using to describe plans. Combines a lot of the experiments previously blogged.

Next job is to get a merge function and clean up so that it look more like below:

Finally

2010-07-16T21:57:00.001-07:00

Finally got a set of walls out of Revit via IFC into my viewer. Was stuck on a problem for ages for the simple reason that I didn't realise in IFCAXIS2PLACEMENT3D(#67,#9,#7), the second attribute given is the Z axis, not the X, the first is the origin point and the second will be the Xaxis. After making that small adjustment everything fell into place. Still have to do something about polylines and curves, but for the most part am happy that I can say a mock-up of the system is complete and I can now start working on the unit which begins (eeeek) next week which I'll be lecturing in. The rest of this weekend can be dedicated to extracting images and notes from all the HVAC books that have been accumulating at my desk.

More Plan Generation

2010-07-14T00:50:00.000-07:00

A few small adjustments to the l-systems shown in the previous post. Later tonight will add the grid and space coloring. And after that merging of colors and checking the graphs.

Lsystems again

2010-07-07T07:47:00.000-07:00

Moving back towards l-systems :D

This week am combining a reworked version of the l-system code in the initial experiment with the subdivision and cellular code. Hoping that this new genotype-phenotype matching will be rich enough to represent plans decently.

2010-07-02T09:48:00.000-07:00

Sent from my iPad

Termites

2010-06-30T04:11:00.000-07:00

Still at early stages - trying to get some termites to build their mound.

Some distortion

2010-06-29T07:29:00.000-07:00

Moving one control point on the cylinder, not as smooth as I'd like it to be - though I guess you have to adjust the weight with the point moving to get a smoother result. Just tried the same in rhino, and the cylinder does the same after its just made, but when rebuilt, the control points are moved closer to the surface.

NURBS Cylinder

2010-06-28T14:05:00.000-07:00

Again, one of those things that took longer than it should have, just always mixing up indexes of arrays and such, and hard to track bugs.

Next step will be to add the surface in the draw function. Would also be nice to be able to get the surface normals and to test for inside/out.

Will have to start working on the BREP structure soon.

Circle Part II

2010-06-27T23:46:00.000-07:00

Just moved the code to my Tao GL template. Feeling quite happy about this. Threw in a Torus for good measure :D.

Circle

2010-06-27T23:18:00.000-07:00

Slow weekend - got stuck on a problem which shouldn't have taken as long as it did.

Finally got it working - ta da!

So I suppose you're wondering whats so special about the circle. This one was made as a NURBS curve - so finally I have in my library a curve general enough to be any conic section as well as act line a Spline. Its also not that big a step to start building NURB surfaces. I'm sure I had NURBS last year sometime, but was using OpenGL's evaluate function.

Next step, have to find out, whats the best way to intersect two Nurb surfaces and to get resultant Curve.

Also found an interesting program called Ayam yesterday, will take a look at it later.

First CFD simulation

2010-06-20T10:23:00.001-07:00

Writing post from my new iMac :)
For a long time, I didn't like Macs, cos I had difficulty using the one-button mouse and a new system, but like with anything, you get used to it over time. Am still nowhere near as familiar with UNIX commands as cmd line ones.
Ran first simulation from OpenFOAM - took awhile to get used to the system, and was glad when it was working in the end.
I think moving away from the operating system, I was using was good for me, for starters ended the clutter that I had created!
At the moment am downloading the free version of Ensight

Computer got taken away

2010-05-11T21:46:00.001-07:00

2010-04-28T08:45:00.000-07:00

Query from a student about modelling soft forms, we came to the solution by trial and error.

Sweep

2010-04-24T05:22:00.000-07:00

Pretty much working now, sweeping along a bspline path. Learnt a bit about projecting points onto planes. In the first try, the profile was translated and orientated to the coordinate system at the division points, in this version, the polygon is projected from each plane to the next - that'll help when the path consists straight lines at sharp angles - have to now write a bit to check whether the path has any segments in it that are sharp say close to 90 degrees and then to add a plane between them at 45.

Frames

2010-04-20T02:28:00.001-07:00

Coordinate systems along a curve. Not sure why this took so long as the maths is quite simple. Was first getting plans along the curve which was relatively easy as you have a normal at any point and an origin. The part I couldn't figure out was then putting points on that plane. Getting a normal vector (from the infinite amount present i.e. the plane at normal to the vector) could be done in pretty much the same way its done in 2D and having the y-value as zero. Which is where I got caught out last night when I used the get 2D normal method from some old code - as in that the z-value was set at zero.

Also one of the stupidest errors ever:

Line lv = new Line(p1s[i], p1s[1 + 1]);

vector v1 = new vector(p1s[i], p1s[1 + 1]);

it took a good hour or two to see that that 'i' was '1'.

Ball of yarn

Translating planes

2010-04-16T08:50:00.000-07:00

Got back from Hong Kong two days ago and its been hectic catching up with all the tutoring.

Outcome for today was moving the points on the base Coordinate system to that of another plane. Tomorrow will have another go at getting some curves, then their normals, then the planes at normal to the curves, to recreate some basic parametric exercises.

viewer

2010-04-04T05:54:00.001-07:00

Working on a viewer based on a video I saw about Ken Yeang, the idea is to separate out the model into different parts.

contintes

2010-04-03T00:35:00.001-07:00

So far so good, splitting an icosahedron with a plane.

Booleans part III

2010-04-02T22:26:00.001-07:00

Going to attempt solid booleans again, this time using BSP trees. First task is to divide triangles by a cutting plane. It seems that this is much simpler than the last attempt as in every case where splitting is appropriate, the result is 3 triangles.

OBJ

2010-03-27T09:03:00.000-07:00

A task that I've been meaning to do for a long time - read in OBJ files, its a very simple format - a list of vertices and a list of faces.

The nice thing is that its a very common format within CAD software, so can read in mesh data into my apps.

Was stuck on a horribly simple problem for ages - trying to figure out a way to look at things from a calm and rational point of view than just grinding away at problems mindlessly.

Cellular Automata

2010-03-13T07:38:00.000-08:00

is it just me, or does this CA look like an Alien head ?

E+

2010-03-09T04:24:00.001-08:00

Working on a dotnet wrapper and simple interface for E+

Pavilion

2010-03-05T07:52:00.001-08:00

Working on confirmation of candidature document, looking for stuff to put in to make it look like I didn't spend the whole year fiddling around with unrelated experiments and watching anime.

Series was for a competition for a pavilion in Korea. The actual images were suggestions for part of the ceiling.

Graph to plan

2010-02-25T21:07:00.000-08:00

Probably one of the algorithms that I have been happiest to see in awhile! Have been thinking for awhile about graph layout algorithms - first in final year when I was looking at l-systems and trying to get nodes connected. It was only recently that I found force-directed layout algorithms. This one uses the Fruchterman-Reingold algorithm. I then connected it to the rectangle dissection shown in an earlier post. Next steps will b to connect this up to the cell matrix as well as getting some basic symbols in - doors, windows, stairs, furniture.

Another problem that just kept going

2010-02-21T01:35:00.001-08:00

An image, I'm really happy to see after working on it for so long. Doing tests for Point in Polygon testing, but using BSP trees as the structure to hold them which added extra layer of complexity.

Still got some bugs, but I think its time to take a break from this!

Mosque

2010-02-17T04:35:00.000-08:00

Made some adjustments to the appearance of the draw class. Was happy with the result as had been dealing entirely with regular grids last week. Next step is to extract something more 'declarative' from the plan so that it can be altered more freely.

Hyperboloid of one sheet

2010-02-11T05:40:00.000-08:00

Looked at OpenCascade today, I had downloaded the previous version last year and was having trouble compiling it, but the latest download works well. There is also java and C# source code available.

Wrote some code for a hyperboloid of one sheet, as a challenge. Might come in useful when I model some of Gaudi's work later but will have to optimize the solid boolean code before that can happen. Other than that, will be working on a very simple sketch program to create some floor plans so I can analyse what geometric relations need to be captured to give sensible results when it recombines.

Solar heat gain

2010-02-10T02:12:00.000-08:00

Not a particularly productive day, though did find that a good friend is going to be a mummy! So thats exciting news.

Continued on the GA with some basic fitness by the CLTD, linearly interpolating between the N,S,W,E cardinal directions. Results are making sense, like if I turn off the minimum size and proportion control, it starts to reduce itself to a line, like bottom image.

Pavilion

2010-02-09T00:03:00.000-08:00

Supervisor 2 put a pavilion onto my list of things to do:

Am basing in on John Frasier's original reptile scheme.

Grayson, 1970

2010-02-08T23:49:00.000-08:00

Looking at some old floor plan generating algorithms from the 70's. One that gets cited a lot is Grason's Thesis : "Methods for the computer-implemented solution of a class of "Floor Plan" design problems"

A summary of it is in his paper:

http://portal.acm.org/ft_gateway.cfm?id=805070&type=pdf

Carried out my own interpretation of the work using an adjacency matrix and a grid which is sized by the number of nodes connected to the direction (N, S, E, W) notes.

Strange to think that back then, it took over an hour to generate the solution, while now this can be calculated in real-time.

Some more subdivision

2010-01-27T17:10:00.001-08:00

Did two more versions of subdividing polygon orthogonally. The basic algorithm goes:

Start with rectangle,

insert a point if there are no points in the rectangle,

if there are, put the point in, and divide the rectangle into two so that each is in a new rectangle, delete the old one from the list.

Satolas

2010-01-27T06:34:00.001-08:00

2 Weeks without a post?? How did that happen? Can't remember what I've been doing since then - reading up on thermal performance, playing with graphs, working on pavilion, writing papers, playing super mario bros. Still.... two weeks ??

Got a paper accepted for caadria, just need to tidy it up a bit.

One of my holiday projects was to make a model of Satolas TGV and later the Milwaukee Art Gallery, how fast Grasshopper is, never ceases to amaze me. Did a quick GH model of the roof and made a start on the larger curtain walls. Am trying to give some thought to generalizing the solution.

3D CA

2010-01-13T19:29:00.000-08:00

Made a small adjustment, replacing the points with a GluSphere

CA

2010-01-12T17:36:00.000-08:00

Had a got a 3D Cellular automata set. Couldn't find a nice set of rules like with the 2D Game of life. Yesterday supervisor showed me some work from way back of interesting 3D CA formations.

Actually

2010-01-11T21:21:00.001-08:00

Actually results arn't so bad when the shapes are subdivided enough that one triangle on Object 1doesn't intersect multiple triangles on the other object.

yay

2010-01-11T20:16:00.000-08:00

Yay, though still having trouble when triangles get too small

trouble with attempting sphere

Getting better

2010-01-11T07:47:00.001-08:00

Getting a little better

Long and frustrating weekend

2010-01-10T07:54:00.001-08:00

Still having trouble with solid booleans. Looks like its making some progress though, might take awhile before the resutls are decent.

DIFFERENCE

UNION

INTERSECT

Intersecting rays and triangles

2010-01-07T22:14:00.000-08:00

Working on two things, intersecting triangles with each other, and intersecting rays with triangles. One step closer to solid boolean operations. Haven't given any thought to optimising at the moment, would like to see it working with a brute force calculation first.

new viewer

2010-01-06T18:48:00.000-08:00

While in Welly was working on a new look for the 3d environment. Huge improvment over the last one. Also can overlay 2d lines over the 3d screen.

Offsetting polygon

2010-01-04T21:28:00.001-08:00

Spend weekend working on graphs. A nagging thing on the to-do list has been to turn a graph into something closer to a floor-plan, so was working on offsetting polygons. Only really useful in 2D and will spend the rest of the afternoon working on implementing walls in 3D.

Inverse kinematics

2010-01-04T21:12:00.000-08:00

After a challenge from Kris, it became apparent that we needed an IK library to solve some folding problems. Algorithm not very stable, largely because it relies on taking the inverse of a matrix, which can produce some weird results. Will have to update it later to take the psueudo-inverse instead, if anyone has that as an algorithm please post it.

Unconstrained GA

2009-12-28T02:21:00.000-08:00

Working on a GA to solve a simple geomtry problem - make a polygon that approximates a circle - without ever refering to the analytical formula for a circle.
Selection was based on number of line intersections and the even-ness of the length from the centroid. Had to fiddle around with the rates of mutation and population size to stop it converging too quickly on a local maximum.

RGBE files

2009-12-14T04:44:00.001-08:00

Adapted some java code to read in Radiance Pic files, to C#. Was quite happy with the results as I learnt about binary operations as well as RLE compression in this small project.

More Subdivision, checking connections

2009-10-10T06:06:00.001-07:00

More work on the subidivion code - now also keeping track of what shape is connected to what. Easy way to do this is to see if the points from each rectangle lie within the others. A (O(n^2)) approach, but I don't expect the number of rectangles to go beyond a few dozen so its alright for now.

Easy way to check if a point is in a rectangle:

P is the point
P1 is a corner of the rectangle (the other corders are P2, p3, p4)
V1 and V2 are the two vectors that define the sides (i.e vector from P1 to P2, and from P4 to p1)
V = is the vector P to P1

Then P is in the rectangle if
0<=dot_product(v,v1)<=dot_product(v1,v1) and 0<=dot_product(v,v2)<=dot_product(v2,v2)

Subdivision

2009-10-09T06:28:00.000-07:00

Yay! Finally got something decent from the subdivision code. Spent quite awhile fussing about how to connect rooms up. Then it just struck me that its quite simple to do that after the geometric processes that trying to do that during the process.

Fixed block sizes

2009-09-10T01:53:00.000-07:00

Working with a design that uses blocks of fixed size, so had used Grasshopper to constrain the size of each hexagon, and instead increase the radius of the stack, in relation to more blocks being used.

Circle Collision

2009-08-02T23:21:00.001-07:00

Did some collision detection algorithms for 2D circles and will extend them to 3D spheres. The algorithm uses a hill-climbing algorithm and thus usually misses the 'true' optimal solution.

Tried combining the grammar system I did last week with my viewer code from ages ago. Still works remarkable quickly. Next step, I'm going to look at some of Wittkower/Rowe's analysi and see what are the smallest number of functions needed to create 'parti' models of Palladio's plans.

One more

2009-07-23T03:45:00.001-07:00

Thursday night

2009-07-23T03:23:00.000-07:00

Been a long day, brain too tired to continue - so blobs shall have to suffice!

Follow up

2009-07-21T23:23:00.001-07:00

Aiming for a softer finish:

Next step is to look at process at different scales.

Fun with spheres

2009-07-21T06:30:00.000-07:00

Got send an image of a project looking at the void space of closely packed spheres and did a quick recreation of it, might produce some interesting results especially when the forms are more dynamic or more streamlined or when there is a range of different sphere sizes.

Alcoy community hall opening

2009-07-17T03:17:00.001-07:00

Have always been fascinated with the work of Calatrava - thought I'd start on a tutorial on the opening on the Alcoy Community Hall as a Grasshopper tutorial.

Grasshopper

2009-07-15T01:11:00.000-07:00

Doing up some grasshopper transactions for the class. One observation is that a lot of Calatrava geomety can be abstracted to a a handful of replication types - multiplying along a curve while scaling, multiplying and scaling to two curves etc.
Will try to build up enough transactions to make a basic model of the TGV at Lyons.

The modelling was done in Sketch-up and exported as STL to Rhino. Geometry pretty much stays in place - so editing the stl file and reloading it makes making changes easy.

Polyps

2009-07-14T02:50:00.000-07:00

Again nothing substantial - just some experiments on polyps for a pavilion competition. Had forgotten how long parti volume rendering takes - computer is moving at snails pace at the moment.

Random-ness

2009-07-13T04:46:00.000-07:00

Haven't posted in a week due to not having internet at home - don't have anything usefull to add so will post some random things made while waiting for a download.
Okay, will have to start being more serious. Next week a workshop with Carlo Ratti starts, have to do some prep.

Skyscraper

2009-07-04T02:42:00.002-07:00

An entry for Evolo that I never eneded up finishing - will try again next year, hopefully with better time management but also managing to consolidate all the bits of code that are floating aroung my computer.

Patterns from old project

2009-07-03T06:20:00.000-07:00

Another image from an old experiment. Looking back on it now, I'm wondering how better to formalise the nested hierachies - to put some circles as 'children' of different centres, which in turn are from a different centre...but also to desolve these relations based on some sort of proximity rule - so that in future work a 'cell' can cause reactions in the cells around it.

Recreation of JF's Reptile WIP

2009-07-02T03:30:00.000-07:00

Recreating an old evolutionary design project - still a work in progress.

Snowflakes

2009-07-01T02:41:00.000-07:00

One of laste years experiments. Animated them in MAX - next step is to try the same thing with some physical properties such as the surface tension of water and effects from the rate its cooled at.

http://www.youtube.com/watch?v=Rr6EZf486pI

Situate Sculpture Competition

2009-06-30T01:36:00.000-07:00

Entered a sculpture competition with some colleagues at school. The idea was a bit site specific in that the sculpture was pointing towards the station across a highway. Had some fun modelling playing off organic and rigid forms.

SmartGeometry 2009

2009-06-29T04:26:00.000-07:00

Got to go to San Francisco for the SG conference - learnt a few tricks in GC and spent the workshop time trying to connect it with C# express edition. The final project combined some experiments from my final year project 2008 - with some notes on Gaudi's parametric rules.
Where to go next? Would like to put in the rules for columns and vaults and it would also be cool to be able to animate the growth sequence.

My first tree that looks like a tree!

2007-12-26T02:21:00.000-08:00