Sunday, April 20, 2014

Final project - alsarc

The plan:


So I started with two multi-section surfaces controlled by curves


and projected a grid of points onto the surface.

This was my module, driven by eight non-planar points as inputs.





I couldn't get the power copies to work because my forms were in different geometrical sets. Instead, I made a product file with global parameters and a document template. I imported the template into the part, and the inputs were the 8 points and matching parameters, so I could change them after. I could also go into the sketch, changing the multi-section surface and the spacing/shape of the projected grid. TA DA.

Sunday, April 6, 2014

P009/10 - scripting

Struggles...
I'm at the last 2 minutes of the tutorial. And getting this error.

I've checked that it matches the script custom file and it's still messy... Will resolve soon.

Update: solved through re-making udf file again and again and again... I had checked that the script matched the Cat script file, but that didn't solve things. When in doubt, start over!

Look, I can move around the point that the surface passes through, change the number of curves and change the height of the plane that the passing point is set at:







Part 2
Look I broke it.



Self intersecting panels? Clean the wire?





A better view of the panels

Less bulbous


Sunday, March 23, 2014

P08 Knowledge patterns - alsarc


After lots of greif with the power copy - note: don't copy the equations, cut and paste, so there's not more than one instance of it.






This used an old part from the beginning of the semester, and I used it in a much more efficient way now that we know how to powercopy! I powercopied a rule:

/*Rule created by alsarc 3/23/2014*/

If InnerCircle.1   < .5in
{Depth  =.2in}

else if 
InnerCircle.1   >.5in and InnerCircle.1   <.8in
{Depth   =.5in}

else if 
InnerCircle.1  >.8in and InnerCircle.1   <5in
{Depth   =1in}

*My computer unfortunately blue screened of death before I could take a dinner break and color code the height changes. GAH *

If the file hadn't vaporized, I would have made 3 separate parts, with global parameters based on each of the three circles so that they change height individually and color code based on the change. Ah well. Now to fix my computer...

Algorithmic Architecture - Reading Response

This reading discusses the interesting relationship between human design and computing, providing some interesting theories as to why computational design is seemingly misunderstood. Architects are slow to adapt to the increasing use of technology, but it is because we see algorithmic architecture as too many defined variables when we are so used to "vague, indefinite, and uncertain" design. We don't understand the complx nature of the processes, and "cannot implement them in a meaningful way in design." Our ongoing struggle with working with engineers is much more obvious, for they look for rational consistency when we fight for creativity and resourcefulness. We need to have an open-ended design process, so that any changes in design are not hard to adapt. Architects think that computer aided design impedes this fluidity in the design process, when we are really just understanding how it works all wrong. It is possible for the world of science and the arts to coexist, as long as we understand the principles of design and how to appropriately apply them. We should give algorithmic architecture a chance, because when used properly, we will "open a portal into another world new to the human mind" and explore a much larger range of design possibilities.

Sunday, March 16, 2014

P007 - unfolding

Here's my unfolding, along with small variations to the height factor and original framework






Mark Burry - Models, Prototypes and Archetypes

"Specifically, just as we begin to resolve the model's diverse role as part of the architect's design repertoire, we are confronted with new dilemmas: the model as model, as prototype, and as archetype."

This reading discusses the relationship of digital modeling, prototyping, and models, specifically as they are applied to Gaudi's Sagrada Familia. Having the ability to make full-scale prototypes is extremely valuable, although smaller scale projects might not necessarily have that kind of budget. Models like the classic hanging chains are perfect for these situations, as they can inform the physical nature of the building without a full-scale construction. While it is not as common for models to inform us as much as a full-scaled prototype, they can't be written off as invaluable in the process of fabrication from a digital model.

Sunday, March 9, 2014

P006 - powercopies - annie locke

FJKDSLFDS:

This is the story of all the errors...

Everything going alone fine... then:

FIRST: didn't make input and framework different geometrical sets. Bad news bears.

Second: was selecting the multi-section mesh as powercopy instead of the whole geometrical set. Rookie mistake. This wasted a good hour or two completely rebuilding. and rebuilding. and rebuilding.

Then: powercopies: success! yay happy AL.


Then... oh wait.
 The evil sweep operator came along. Had paused the tutorial so I didn't continue watching to see that you had the EXACT same errors, but I basically went though and rebuilt that one sketch/fixed it blah blah.


SUCCESS

but oh wait. more errors:

by this point I figured out that I needed to watch the rest of the tutorial and found how to reverse the offset. Easy enough.

Pretty!


This is how far I got: all the other powercopies kept erroring but I know how to fix one so I figured it wasn't necessary to fix them all. Yay learning through messing up...

P006 - Spuybroek - Textile Tectonics response

This is a great reading to setup power copies in DP. The reading discusses the relationship of parametrics and craft, relating the process of making by carving vs. modeling. He compares the variable modulation of parametrics to that found in nature (a la leaves/biology), suggesting that parametric modeling keeps design from being too arbitrary, while also having a naturally informed logic. While it is not as simple as translating thoughts to paper via pencil, it does allow "variable modulation," allowing copying and mutations to be regulated and informed. The notion that all craft needs to move to robotics is a little far-fetched, even though there is some beauty in controlled variation. However, I'll still a big proponent of not leaving the computer to do everything, and finding the beauty of accidental discoveries that are naturally inherent in human craft and design.

Sunday, February 23, 2014

Assign 005 - Framework/STAAD

Here is the analysis done on the members at .01, .1, .2, .3, and .4 % down the original member. I didn't do an animation of the deflection since it isn't noticeable.




Thursday, February 20, 2014

P005 - Loukissass_How do simulations know? response

This reading discusses bridging the gap between speculation, simulation, and built architecture. It discusses the implications of the inaccuracies of simulations, because of their simplification and inability to account for unpredictable variables.

As simulations become more trustworthy, we can challenge the traditional basis of design knowledge. CATIA is one of these tools where many different kinds of knowledge meet, including information from consultants, architects, and fabricators.

Even once simulations have passed all the tests, it is difficult for them to have precedence over trusted regulations. This is not only because of the variables that bring change into an environment but also that varying assumptions can produce virtually infinite results. What use are these simulations to us now? Designing with multiple simulations that attack a problem from different angles is a good way to gain a holistic perspective on building performance. As long as we understand that simulations do not give us a whole or accurate picture of what will actually happen, they do provide valuable information to influence architect's design decisions.

Sunday, February 16, 2014

Assign 004 - Folding paper

So I tried to model this paper fold to make a basic parabola. Based on the tutorial, I was able to get half of it together, but joining them was a bit problematic because it's not planar. I'm not sure how to get this working - maybe make it out of four products to a larger product, instead of two. Also, I'm not sure how to link the dimensions of one part to another - I ended up making 6 individual parts knowing how big they were supposed to be. I'm sure there's an easier way to make a formula for it though...



P 004 - Fabricating Architecture reading response

As parametricism becomes increasingly popular, it is extremely important that architects have a better understanding of fabrication techniques and processes. As the reading points out, a significant number of architects do not, and therefore the accountability and responsibility for construction. Being a CNC router tech myself, it is far easier to design having personal experience and understanding of the limitations of the machine. Unfortunately, we have a lot of students come in who want to make the router do something that it just isn't designed to do. This reading reminded me of this general mentality, and how architects don't actually make anything, they just design. Needless to say, we don't need to all go put our hard hats on and put up framing, but it is increasingly crucial to have a good base understanding of how non-traditional construction methods work, in order to understand budgets and a projects feasibility. With this workflow understanding, we can design more ambitious projects and best understand the advantages of CAD/CAM.

Sunday, February 9, 2014

P 003 - coincident faces and points, changing angles

Part 1:

Part 2:

Part 3:




This week I made three components (re: the tutorial) and associated the faces and center points. The angle was set with a constraint, so that I could change the orientation between components without changing every degree. I had some issues with turning the planes off in the components, because sometimes the whole part would disappear... Resolved eventually with lots of SAVING.