The word "genetic" gets thrown around a bit too easily!

take a look at the Emtech program at AA, although i consider what they call "intelligent" or "genetic" just a form-oriented imitation, nothing systematic to start with...

nomadzilla - I disagree. I think it works at a structural/material performance level. Having said that I'm not a fan of the work.
Jayness - you could perhaps start out by looking at the EmTech work in this book: http://www.amazon.co.uk/Morpho-ecologies-Towards-Heterogeneous-Architectural-Design/dp/190290253X

Perhaps also of pop interest would be the S. Wolfram tome here: http://www.amazon.co.uk/New-Kind-Science-Stephen-Wolfram/dp/1579550088/ref=sr_1_1?ie=UTF8&s=books&qid=1198240473&sr=1-1

try and avoid the architecture distortion field: go to the buzz word sources and make up your own mind how these ideas might be applicable or not:

this is a good starter -
http://www.amazon.co.uk/Introduction-Genetic-Algorithms-Complex-Adaptive/dp/0262631857/ref=sr_1_1?ie=UTF8&s=books&qid=1198240620&sr=1-1

this is a classic -
http://www.amazon.co.uk/Genetic-Algorithms-Optimization-Machine-Learning/dp/0201157675/ref=sr_1_2?ie=UTF8&s=books&qid=1198240662&sr=1-2

and so is this -
http://www.amazon.co.uk/Vision-Computational-Investigation-Representation-Information/dp/0716715678/ref=sr_1_2?ie=UTF8&s=books&qid=1198240877&sr=1-2

returning to the field, basically IMHO you're looking at something that treats architecture as a feedback system with inbuilt optimisation. call it adaptive.

An algorithm is nothing more than a recipe; a set of rules or steps that control the design process. Most designers have inherent algorithms they draw upon when they design, though they may not be stated.

genetics begin to refer to a particular organizational structure of these rule sets. For example, when a particular element can pass on it's traits, or rules, to another; a next generation inherits the 'gene' of the previous.

It's a very interesting part of the discipline. You can have a look at some of last years AD magazines...morphogenetic design strategies, Techniques and Technologies is morphogenetic design, Collective Intelligence, Programming Cultures...for a start.

Computation has definately empowered architects for the future. It's not about form, though it has to look good, too. It's more about technique and discovery.

It's controversial, as is, anything "great"

Yeah, you can either have an optimistic perspective on the issue or one of skepticism. Through all of it, you have to decide if the investigation of this design approach is leading us as designers, architects, etc... towards new ways of spatial expression or is just pointless.

At the Angewandte where I am at this year, the Henan, Hadid, and Lynn Studios are all employing this mode of design into their projects. All are in varying degrees. We has a cross studio review and I was able to listen in on the ideas. The intent of the professors was to use it only as a method, but not an "end all" solution. One of the main discussion topics was the redundancy of projects using this system as that the point should be to see if it leads towards new ways of perceiving design and planning (on both large and small scales) but shouldn't be , as I just stated, the sole method.

The projects were interesting and well thought out. I would say that the projects did a good job of using algorithmic methods on as tools while the individuals in the groups were able to remain the ones actively designing instead of a computer script.

You can check out project on the school's website. In addition, here is Henan's website-you may have already found it

Henan

Hernan* my apologies...

haha
sewer genomes, I love it... Well this topic has interested me so much in the last few years, because it seems that scripting is quickly becoming a missing link in design and production with its ability to intensify the production and create "emergent" typologies. I like to think about how Archi- Genetics is a form of Biomimicry , and the growth of the environment could respond and merge with user data, or environmental data, and the managing the data for more powerful design solutions,

It's a good idea to learn the scripting languages, maybe some java/javascript and actionscript. That way, when you have no job upon graduation you can get a job at a graphics studio.

I honestly think there are more opportunities to employ these methods as art, possibly furniture, than try to think about architecture.

Call me a skeptic (a cynical one at that), but blobby programming is pushing 10 years now with only a handful of really bad examples built (and those ended up using typical construction).


Fascinating and beautiful stuff, for sure, but realistic as architecture?

Trace, the concept of genetic architecture (or even the much more generalized "scripted logic") does not indicate form, application, or construction type. Those conversations are parallel to what Jayness is asking about.

That said, a number of the things posted in this thread have nothing to do with genetic techniques. Hernan's work, for example, uses the terminology extensively, but the resemblance to "genetic architecture" ends there. Nothing against the work, but it falls under another category. No clue why he was even brought up.

Maybe the original poster could clarify what he wants to know about: organic form, biomimetic behavior, or genetic design? These are all very different, and right now they're being thrown about pretty loosely.

.mm

I realize that I am over generalizing, but you have to admit that there has been terminology thrown around for decades and in the end they are simply complex ways of exploring form.

I actually think that creating sculptural form, whether it be through clay or some algorithm, is just fine. Turning it into architecture is fine, too.

But making pretty abstractions and never testing how it would be inhabited, built, etc., etc., and still calling it architecture is where I have the problem.

Maybe I just need to see examples, but it all sounds exactly the same as ramblings spewed for a very long time - over intellectualized abstractions with no beef. I've heard the 'its not about form' so many times too, but it has to be about form doesn't it? At least on some level.
Nothing is wrong with form if it works.


As always, I'd welcome someone proving me wrong.

just to chip in, i'm yet to come across anyone who uses digital techniques and call their work a completed piece of architecture. could it be assumed to be an experiment with architectural behaviors embedded?

perhaps there are no completed works of architecture to show examples, but don't you see it coming? could we say these concepts are still in developing stages, slowly coming to life as there is a growth in technology.

check out some of the ps1 competition winners' websites to know about the works they've done.

This issue remind much about the mid 90' recursive programming. Where some of us acturly looked into these matters and found no end to it. And how could it be any different, while programming and scripting hasn't changed that much, the oppotunities in recursive patterns by 2D isn't that different in 3D. But true some develobment is prone. yet what about buildability , manufactory and assembly are those problems solved ; I don't think so.
To have the items manufactoed , get it into an assemble system described in the core functions do this 3D ,and not only as smoothened undulated skins, but maneaging an uniform material deliver horisental and vertical also square, and maybe in particular square, What I find is failed , is how what's computed must, simply must serve all porpus ; lt's like forcing the brick act as roof, like the tapestry act as both wall and floor , or the roof to be the intire structure and allway's only ground floor newer more ,the roof become the building --- and that is realy a limitation, nomatter how variations in face size and shape force the form.
Still only few systems today deliver more, and the reson this present better on a screen is deep in architecture as how most architects engage forms as skins and buildings as attitude rather than as structure.
Anyway I reconise a whole lot of recursive programming there, one day there wll come a breakthru, but not before architects stop thinking in surfaces.

sure, it can all be one big experiment and/or process, however, that means it is not yet architecture.

My problem has always been that these types of 'experiments' are rarely ever tested against reality (see discussions from the blob era, over 10 years old now with nothing to show but pretty pictures).


But no, I don't see there being something inevitable emerging. This is all still form making in my eyes. If the experiments were truly stimulated by something that was relevant, not just 'cool', such as economy (Gehry is still one of the most sophisticated when building unique forms, and he never tried to sell the process as a single idea), environmental issues, etc.
Putting PV cells on the side is not gonna cut it (Lynn tried that bs 10+ years ago - "Curved PV walls goes here..." uh huh, PV cells aren't curved Greg).


The PS 1 winners that come to mind, such as Hernan's blob or Wiscomb's folds are very interesting, but not architecture (at least their winning entries). Having been a classmate of Wiscomb's, I can attest to his thoroughness of design. He did create architecture, some very beautiful stuff. His designs did push into reality and addressed the things that are rarely even thought of in other blobs (including Lynn's).


All I am saying is the rhetoric is identical, or close enough, to the bs of the last 15 years. Be very careful about being seduced by the process and pretty images. In the end, if you do not test it against reality (building a one-off with the school's cnc machine is only the beginning) then you are just making sculpture.

Which is fine, but then we should be discussing this in another forum.



Merry Christmas all

trace
No one can argue with your rationale. A shining example of successful generative forms aren't abundant in the scale of new buildings yet.

I do think some are anticipating increased automation, fabrication, and materiality improvements in prefab architecture that to lend themselves to the utility of generative computation .

Further it seems many schools continue to embrace computation as a cornerstone of the curriculum. I think Wigley recently said something about GSAPP being geared toward a world 10 yrs in the future.

Predictions notwithstanding mass production and algorhythm continue to 'affect' architectural realms, and entertain designers with fresh methodology , even if we only currently have sculptures to show for it.

I tend to believe that although scripting or various other forms of modeling form are beautiful they fail to perform.

For me i am much more interested in space making not shape making.
Does that make sense..???
Even with CNC and large scale (building scale) printing that is developing i am not really interested unless thought is given to process, context performane and experience.

However, i do like the genetic as in biological/operative design. Meaning a building that functions like organic system. Even if it does not look like it.

This and the discussion on biomimicry are interesting

those are not sculptures. to call them sculptures is to undermine fine arts. this stuff would be spitted out in the art world in a second. please stop attributing these to sculpture.

Sorry, but I see nothing wrong attemting to progress a building structure as a sculpture, it's just a laugh how architects succes to do just that.
For a true piece of art , as a sculpture building must be, one can not just omit the structure or hide a tradisional structure under the thin skin. Doing such a sculpture as a building is more a painting than a sculpture.

------- This is far from The Recipee-Architecture the computer shuld evolve ; and nothing learned from the obvious foults allowing a promise of advanced structural thinking delivering a covered brick house.

OA
Yeah, I don't really get why the word "sculpture" bugs you so much.

When, when I watch those videos from the Columbia school blog, I often question teh effacay of robottically built buildings,...but I am happy to see this sort of attempt at pioneering. I think we all understand teh linear equation of "a person making a space for other people". Algorythmic studies imply that social and environmental complexities woud be handled by something greater than designers, with trace paper.

i apologize if that sounds trite, I do respect the tradition of architectural process, and I think computation will be more handy when we apply it to complex systems with performative results.. i the meantime, form and sculpture are just the precursors.

Da ,

If a sculpture is a piece of are , guess it must be, then remember it don't exist before someone compute it into building compoments so, it still will be recipee-architecture ,something that only exist virtural in a computer memmory untill someone decide to compute it into manufactorable pieces .

hey where did you get the conclusion that the word 'sculpture' bugs me?
misuse of it, i point out. that is all. you can't attribute something architects are doing but can't truly explain it, to 'sculpture.' shape and form alone does not make sculpture. real sculptors are spending their lifetimes to be a sculptor you know...
but anyway, say whatever you want to say.

I always saw the installations at PS1, whether digitally derived or now, as some sort of sculpture hybrid. Most would not get offended by my personal assumption, and many would agree...so a rose by any other name....

Anyway, I agree with trace to a degree , but I think the point I would like to explore , personally at least , is how algorhythmic device can help manage complexity of social and environmental conditions and buildings....

A thing can be nice and beautifull, even it don't deliver.
No one said those are non nice not lovely and fine. But the request for buildability, structural tools, floors and walls are still relevant .

No one say this trend is not prone, that it is not interesting in a compute wise way, but it is allway's surface thinking 2D kind of emagining the math. defined volumes, still there must be a way to bring it out of the virtural world, if this is not possible, it's not recipee architecture.

i am not offended by your personal asumption either. i am not a sculptor.
but the use of the word is getting confusingly rampant among the architecture students.
blobwall is now a sculpture? he should file bankrupcy and move on.

OA, i agree with you. i think there are few that have approached "sculptural" but then again i have a personal bias towards Hejduk...

john was a real thing.

well, from now on maybe I'll risk using the word "Architectonic" vs Sculptural...of course I might get shit for using Archibabble...

I am just looking for a way to get to the next level with teh actual thread topic and in less time and without bothering about semantics with critics offering little perspective other than the occasional vocabulary omission or correction.

Everything can be modeled 3D, even with gravity and feedback compute of the nessery material dimensions. There are plenty skills defining the defining sections, plenty soft math. avaible by fingertip, and it don't has to be restricted to the skin only ,or be trapped in wonder materials not yet invented. Today's modeling CADCAM could look poor being best in fact, manufactoring 3D volumes from simpler 2D cut things, everyone would expect highly advanced 3D laser routering holographics and unbelivable different understood technologies NO ; delivery come by making it simple, widespread by P.C.'s and trainie CAD software , not the heavy invensment assembly line robots , but by plain 2D N.C. cutting.

Interesting points PER....

I wish someone form a recent Columbia studio would chime in here.

c'mon. that is not a vocabulary correction. i don't give damn about your vocabulary. i made a valid point of your wreckless use of a defination.
how can you go on the next level when you can't decipher the current..?

Who ??

blob wall? thats new to me.

Guess that describe a lose instalation , melt into the surroundings or free standing. Nothing that add to the structural capasity, an interiour gadged providing somthing that meet the eye not nessery usefull.

Wow, nice sculptural wall. LOL

I borrowed "sculpture ' from trace's train of thought,..I guess I should have used quotations, maybe footnotes. Anyway, misused words aside ,and not that any one cares...but my train of thought with this thread demonstrates a mere curiousity about getting past notions that generatively coded built forms , (whatever you call them forms, installation, architecture, or fodder for bankruptcy) as being artful arrangements. I too am anxious to performative properties, and significant architectural purposes produced from computation. I am the first to admit I don't know how this will or can happen, but it remains a intriguing question ...for me.

the blob wall is great... although is generic than genetic.

i do not see it as sculpture because it tackle the fundamentals of architecture.

i think sculpture is for viewing only, u cannot live inside a sculpture... that way, i think is unfair to say a building is sculpture whenever u see some funky form.

I know that this conversation has gone some completely other place, but it really is going to bother me if we don't clarify one thing (again)... there is a huge difference between biomorphic form and genetic algorithms.

Here's an example (in addition to Meta's example of the Bentley software's use of a genetic algorithm to optimize generic building systems)....



This is a 3D model of a truss that was optimized using genetic algorithms. It's not the brainchild of a wacky west coast designer or an AA student with too much time. It's the work of NASA engineers, who began at a typical truss configuration. 4500 automated design evaluations over the course of 18 generations (each sharing mathematical lineage to the original "Adam & Eve" 2D and 3D trusses) supplied this outcome, a truss optimized to the specific loads and vibrations.

I also find it ironic that the term "genetic algorithm" is responded to with knee-jerk negativity, while a thread on biomimicry in urban design goes on living strong. The two are natural extensions of the same line of thought when you start referring to "behaviors" and "systems" rather than "concepts" and "parti."

Now, to be fair, this thread wasn't explicitly about genetic algorithms or biomimicry. So I'll get off my soapbox. Greg Lynn doesn't do any more calculus-based design than any 2nd year undergrad at SCI-Arc: throw a lattice on it. Copy rotate. Whatever. Let alone genetic algorithms.

Instead, it's about the blatant misuse of the terminology. Just wanted to try and shed some light on one of the terms being tossed about here.

.mm

Mmatt,

Thanks for clarifying..

that is actually what i was getting at in my earlier remarks. i think the questions/topics are similarly originated. At least for me. I am interested in both when it comes to how they can simulate living, fluid solutions.../algorithms for the design...which solve or respond to actual context..


Not just random generated shape.

optimization formular for human behavour is often a lie. take the stock market for example...

architect believe it becos is easy to sell to their layman client.

thanks for helping the thread..good stuff