Connecting FormIt and Dynamo for Predesign Formal Thinking:

DAVID BEACH: So this is "Connecting FormIt and Dynamo for Predesign Formal Thinking." I'm David Beach. And we'll talk a little bit about that here in just a second-- a little bit about the topic and the presenters today. A couple of key things today-- site, design, virtual, and physical, all connected by data, is sort of the core of where this is going, in terms of your predesign thinking. And really begin to consider data as a key part of your predesign strategy.

So I'm Dave. I teach at Drury University, Hammons School of Architecture. I have a longer version of this. But the short version is that I practiced for about 10 years, 12 years. But I started teaching adjunct. And it kind of sucked me back in. And then just kind of reflecting on my own DNA, my great-grandparents were one-room schoolhouse teachers. My grandparents were schoolteachers. My parents were teachers. So the family business, the teaching mafia, kind of pulled me back in.

Ethan Kaplan is a fifth year architecture student at our program and a modern day Bucky pirate. That is, being more specific, Buckminster Fuller and the notion of pirates learn everything that you need to do-- what you need to do to thrive. And Ethan does a very good job of that.

Nia is third year architecture student, speaks five languages, is our university's principal cellist, and does not breathe without holding her sketchbook. Yes, although, you don't have it today. But that's OK. She'll hold her breath while she is presenting. So that's a little bit about us.

Drury is a very small liberal arts university in the middle of the United States. We are landlocked. We are isolated from those tech sectors. But we're doing something really well. And I'm really proud of this. We have somewhere between 25 to 30 AU presentations from our alumni in the last five years. And I'm really excited about that. Our application of technology is really starting to take off.

This class is very much about breadth, more so than depth. This is what we're going to cover today. OK, so this gets kind of crazy. In particular, site data in Flux, site topography, and existing buildings, we're going to be importing that into Revit. We're going to be pushing those files into FormIt.

We're going to be using Street View and Photoshop to match existing context. Insight and Flow for simulation. Dynamo, we'll start doing parametric data and design evolution, in particular. We'll be looking at Fractal for design variables and sort of the entry-level thinking of AI and generative design. Revit Live for HMD and fabrication and physical data.

And this is sort of the new part of this, all of that thinking, usually when I would give this presentation, Revit would be my hub. How I've done this type of talk in the past. But for predesign thinking and for what I'm teaching at university, I'm kind of shifting this kind of talk to think of FormIt as a hub for these things.

And when we're talking about predesign, in particular, I'm not talking about getting much past a simple box, OK. So I'm not getting into doors and windows and schedules. I am talking about early predesign thinking and what that means.

So let's start with the notion of site data. How many people are familiar with Flux? OK, cool, It is taking off at our university really, really quickly. Very, very popular program-- and so what Flux is doing, in particular-- this is online. We are selecting a location.

OK, so this is 19th and Main in downtown Kansas City. That pre-automatically generates for us topographical data, streets, roads, and if there are some specific site features as well. From there, we can download that directly. Or I can go directly into Revit, and there is a Flux plugin.

OK, so that Flux plugin allows me to find my base model that I built. And import that directly as-- if I select the topography layer, will bring in the topography directly into Revit as an editable topographic surface, OK. No monkeying around like what I used to teach with SketchUp. And then pulling that SketchUp file into Revit or anything like that. It comes in directly, and it's ready to edit, really, really cool.

Now, accuracy of this isn't something I'd want to build off of. You're going to hire a survey for that. But, again, for predesign thinking, this is set, and ready to go, and gives me some of the data that I need that fast, OK? So it's really cool. And obviously, it's taking off really quickly with the students.

Next, I go back into my Flux plugin into Revit. And I'm going to bring in the roads and building footprints. In some location, you actually get building heights as well. So you actually can bring in volumetric buildings. In this particular location in Kansas City, I don't have heights. So I'm simply worried about the building footprints.

Next, I need to group all of this stuff together because it is not at the origin. And if you've worked at all with Revit, you know Revit really likes the building at the center of the world. So I'm going to group all this stuff together and give it the name site. And then I'm going to start in my east elevation. And so this is a lot of data. So it takes a little bit of time to regenerate that view.

And I think it's about 1,000 feet above the plane right now. So I'm going to turn off my constraints, grab the middle of that topography, and move it down. So this is a two or three step process to move this topo file closer to the origin, is what I'm working on right here. OK, so, again, we're going to grab that and move it. So I know my site is someplace right around here. And I want to move that over to here.

So, again, it's a simple move tool. And, again, with this Flux data, it wants to be constrained. And sometimes there are some conflicts in there. So just make sure that you turn off the constraints. Also, it helps to turn off hidden line so that I can see through because at times there's a little bit of issue with lines overwriting each other. So we're simply going to slide that over into place.

Now to fine tune this, to make sure that everything is lined up exactly where I need it to be, the last step of this is, I cut a section all the way through the site. Double-clicking on that section, lets me really see exactly where I'm at on the site, what area that I want to bring to zero. So that I can start working with this data as something that I can put the building on.

Now, what I'm going to be doing with this next, is pushing this into FormIt. But if you notice, this is all ready to go. As I start building my FormIt file, this is ready to start receiving FormIt data back. So those two things are going to match up. Whatever I put it in FormIt in this location, is going to match up exactly with whatever I've built in Revit.

So those two things-- I don't like repeating data. That's going to be a theme. I don't like building things multiple times for multiple uses. The more mileage I can get out of a single object, the better I'm doing.

So the last step in this is the section box tool. The section box tool does something really nice for your files. First of all, it cleans it up, looks a little bit better. I don't have the dangling edges anymore. But there's something for fabrication, in particular, that is really nice that the section box tool does. It was sort of an unintended consequence.

So the point of the section box is actually to get-- in terms of what Revit does-- to get this little edge right here to display that I've cut through the earth. I've cut through the ground plane. But that's an actual piece of geometry. And what that does as I start to move forward is, I can turn this topography then into a watertight model, really easily, once I export it. Which is really important, in terms of doing the 3D printing, which we'll get to here in just a little bit.

So the last step for this, is to go ahead and export this as a DWG file. And we'll import that DWG file into FormIt. So Flux doesn't only handle data coming down from the cloud into Revit. You can also push things back up. And Ethan is going to talk just a little bit about his project, he's working on in studio in that.

ETHAN KAPLAN: So as Dave has kind of already began to hint at, data is starting to become the basis of where the predesign process begins. And what Flux is starting to become an amazing tool at, is coordinating this data between how you work. So Flux allows you to coordinate between programs that before would involve complex export processes and import processes.

So for my particular project, I work between Rhinoceros and Revit. Revit is very good at technical drawings and producing those sorts of things, whereas Rhino allows you to leverage model building capabilities. So what this means is, I can coordinate a model between what were formally two very separate programs, and now be able to iterate with them, and reintroduce making into the design process.

So what this results in is, you're able to work with the flexibility and the technicality that Revit provides you, with the freedom of making that Rhinoceros allows you. The heart of reintroducing the designer into the software process is being able to work how you work best, with the tools you need to work with.

So this is my desk. This was actually a week ago. And the population of models has actually doubled in that time because you're simply able to use tools like Flux to be able to produce at a capability that wasn't previously possible.

DAVID BEACH: Cool, so next step in this process is importing that geometry into FormIt 360. So, again, process is pretty simple. We're going to start by doing an import. We'll select the DWG file. And that is going to come directly in.

And one of the things you'll notice, inside of Revit, I did take the topography lines and set those to every 10 feet, a little less visual clutter, a little less visual data. Those sides come directly in. And what I'm going to use is a surface tool. So I've double-clicked this to modify the geometry. I'm going to select my four edges. Bam, now I have a watertight model with my topography.

That is a really easy step that isn't supposed to be really easy. OK, that's a really relevant thing in terms of getting more models. It does build a few little extra pieces. So hopefully, we can work to refine that over time, in terms of that tool not making those extra pieces around the edges. But it's not a terrible thing to go back in and delete those. You just need to be aware that those lines are there.

So the next thing that we're going to do is come in and create a new material for this that has a little bit of transparency on it. And that allows us, once I have some transparency, I can see through that to a satellite image as well. So I can get kind of the best of both. I have this line data. And then I can marry that with the satellite image as well.

So I'm just going to go with the sort of default warm, brown dirt tone and give that a 50% transparency. So that at any time, I can edit that if I want to see that as a solid surface. Or I want to see that as a transparent surface. So I can see through to my satellite. I can do both of those at the same time.

Now one of the solutions could be to drape a satellite onto this, but FormIt doesn't texture map that way yet. OK, it just doesn't have that sort of draping feature in it yet. So this is kind of leveraging those two things together, the ability to do a satellite per location, and then just leveraging transparency so we can see through.

So setting the location up, there is a button for that in FormIt, pretty easy. 19th and Main, Kansas City, Missouri, it's going to drop a pin location. And you simply navigate through how much you want to get. I always encourage people don't get way far out. I don't need all of that site. I need the data close to my building site, OK.

So I'm looking at a little bit less than a block area. Make sure that the ground plane is turned on. That satellite image is mapped to the ground plane. So if your ground plane is turned off you don't have a satellite image. And then simply nudging that DWG file that I imported over, so that my linework, my roads, all of that stuff start to match up. Really, really easy to do, not a technical issue at all, in terms of doing that. So you'll see that those things start to match up.

And now, I have a few additional things. I've got trees. I've got the width of the road. I've got a few additional pieces of information there, that I can start to work around on that file. So the next piece is starting to build context models inside of FormIt 360. And there's a couple of ways to go about doing this.

One of the first things to think about is an automated process. Can I just grab that building file from someplace, a database somewhere? And at times that works. At this location in Kansas City at a Crossroad districts, none of those models were available. So we knew we were going to have to build them from scratch.

The other thing that I will say to that, I want models that I can use in a variety of things. I want to be able to 3D print them. I want to be able to see them in FormIt. I want to be able to see them in Revit. And I want to be able to see them in VR.

That said, more often than not, I found myself kind of finagling downloaded models to get them to do what I want them to do. And I've spent more time doing that than just building the geometry from scratch. So I've really looked at FormIt. What's the best method to quickly build one of these custom models inside of FormIt?

NIA DAMGOVA: Yeah, so to start building this model, we'll just use the satellite image. So we can go ahead and trace the footprint. And we'll start by aligning the axes in FormIt to the building that we're modeling. Just so we can snap the geometries really easily and keep everything orthogonal.

So here we just start tracing out the footprint to establish that surface. And, again, it's all in the satellite image. So we can just trace. And we don't have to worry about dimensions. Now to get the height of the building, we'll actually go into Google Street View. And we can use this rectangle tool which tracks surfaces in Google Street View.

And we know that it's approximately seven feet tall. So especially for shorter buildings, It's a great way just to quickly approximate height. Here we see that it fits five times on the building. So it's 35 feet tall. And using this information, we just go back into FormIt. And we can drag our surface to create a 3D volume.

Next, we go back into Google Street View and take a screenshot of the facade. The problem is that it is in perspective. So we just use Photoshop, dragging some guidelines, and then using the distort tool to pull the edges of the image around and get it out of perspective. Since we'll be photomapping this back onto the surface in FormIt to help us start modeling all of those details.

And so once we're satisfied with how it looks in perspective, we can go ahead and crop the image just to the facade. And we'll save this image as a JPEG. A good practice is to save all of the information according to the building address, just for easy organization and communication.

Next, we'll create a new texture using the image that we just created. We want this texture map to fit the scale of the facade. So here in the horizontal and vertical scale, we'll enter the dimensions of the front facade, which we are modeling. And then we'll import the image. And this will allow us to map it onto the facade once.

So here in FormIt, it's really easy just to measure all of the dimensions that we've already added to the building. And using the paintbrush, we add the image to the facade. From here we can start locating the details, like the parapet wall and windows. So we'll use the line tool to start breaking up the surface.

In this case, we're starting with the parapet wall, creating multiple surfaces on what will be the roof, and just dragging up and down to match our image. Here, we can see that the parapet has kind of a slanted surface. So we can drag surfaces up and down. We can also drag individual lines and edges to start matching it up.

The next thing we'll be modeling are the windows. And we kind of run into an issue, because FormIt will remap that photo on the surface if we close geometry. So to get around this you can just set guides without closing the geometry. And once you have all of your details located, you can just go back to kind of the more generic texture that we'll be using, since we don't need the photomap any more.

Go ahead and close that geometry. And again start dragging the windows and details to add the depth. And we can do this for all of the buildings, just to get an idea of the character of the site that we're working with.

DAVID BEACH: Cool, and it's kind of that fast. So we start this with teams of students working on sites together, a group of four or five. And then, of course, we go right into fabrication because we want that model to begin testing our pieces, testing our ideas.

So in terms of fabrication, one thing to notice, sort of peeking underground, once I start moving these things up and down on the topography, I have to extend the base of that building down. OK, remember that, it's going to come up here in just a minute.

So I'm going to isolate the building and select it, make sure all the geometry is selected. I'm going to export Selected Only as an STL file, the address of that building. Again, because I've got 40, 50, you're going to see in a few minutes, 250 buildings. And I'm going to bring that into Cura. Cura is just sort of the default slicer. Everybody should know Cura, that's doing this.

So if you notice, it's sitting a little bit high. That's that additional 10 feet. Because we're always going back and forth, we developed a spreadsheet to help us understand scale. So I know I wanted to build this at 1 to 40, that gives me a scale factor to reduce to. OK, I'll come back to this Excel spreadsheet in a little bit. That went really fast.

So I know to adjust my scale here appropriately, in terms of going FormIt to Cura. And then I also know I need to take 10 feet off of the bottom of the building. So 10 feet 6.35 millimeters. Cura allows for that. Sink it into the base 6.35 millimeters. That's now ready to set in my model and print.

The next step after this, would be go in, bring another building, go in, bring another building. And we might do these one at a time. Or we might put 10 or 20 of them on the build plate at the same time. It just depends on your comfort level with how the 3D printer is working that day, to be perfectly honest about 3D printing. That starts to build this.

Sometimes we put the topography underneath, depends on how much topographic change there is. Sometimes we just two-dimensional print out the geometry, or not the geometry, but the satellite photograph, and start placing the models on it. So again, that's how we start doing these. But the students, as they progress through the program, get more and more ambitious, in terms of handling these things on their own, dealing with the sites that they want to deal with.

So Yung Su doesn't work quite this fast in real time. But this was about 40 hours worth of printing, and about an hour and a half of placing everything organized in the site. So you'll see these projects, what starts as a team, the workflow gets so quick, that students are generating these things on their own really, really quickly.

So that's some of the physical data. But we want to start getting into both micro and macro climate environmental data as well because that should absolutely play a role in the design process. So going back into FormIt, hiding everything, except for basically putting a basic box down. I don't know what I'm designing yet. But I know I want the data to help me with the design process.

So the first thing I'm going to do is look at solar radiation. So, again, selecting all the surfaces and starting the solar radiation file, allows me to see BTUs on a surface. In some locations, that's just data. You're kind of done. I know what that data means.

In a location like Kansas City, that's a temperate climate where in the winter I need to harness that solar gain. And then at times in the summer I need to shield that solar gain. That becomes really, really important.

The next thing I want to do is go in and run an energy simulation on this box. So the box represents my square footage for the building. I'm assigning levels to it. Once I put levels to it, FormIt knows, OK, this is an important piece. This isn't a standard piece of geometry. I'm going to run a simulation on it. You generate the insight. And that gives you this data back.

So, in particular, for students, this becomes really important. They can start to see the modifications, some major, some minor, that their design decisions are going to make. Because these early design decisions, where I place windows, how much glass, all of those things are incredibly relevant and great to teach.

For practitioners, sharing this data with your client early in the process, helps you establish a sense of pattern and compromise for what the building opportunity is. It raises your level of expertise, in terms of perception. Client perception often is devaluing.

While architects are building a monument, they're not looking technically at things. That is a documented piece of information. We need to improve on that. Sharing this data will be really helpful.

As you get into some of this pieces of data, you can see the dramatic change that something like an occupancy schedule has. Why that's relevant to predesign thinking is pretty simple. How you begin putting programmatically different building elements together.

Are there ways that we can start closing down the building at different times? Because when you look at that kind of swing and operating schedule, people have an incredibly large dramatic impact, in terms of how the building is going to perform.

Moving from that, so this is a little bit of an aside. Because typically wind is something, when I work with wind, it's a little bit after predesign. But it's so cool. I wanted to throw this in. The online version of FormIt gives us these little wind charts, which are really great.

And, again, choosing the close weather station, I can see these monthly wind roses, which are incredibly-- again, for a temperate climate like Kansas City, there are times I want to block, times I want to get wind. So this is going a little bit further than we typically would be for predesign thinking.

But subtle changes, in terms of how the geometry works, can make massive changes, in terms of how wind can start moving through buildings. And that's something again, we're going to have to start figuring that out. I don't think there's any question that we're going to have to start working with more natural ventilation in building, in order to reduce energy, to come anywhere close to the 2030 challenge goals.

Oh, one more thing, I do want to back up on this. So this is Flow. If you haven't worked with Flow, it's a really simplified version of CFD, Autodesk CFD. It is not really designed for internal airflow on buildings at all. But you can make it work. What you have to do is first set your scale up correctly, make sure your building is the right size.

And next, make sure your voxel size is small enough, so that the voxels-- that is what CFD is essentially looking at moving through. Those are your particles moving through space. That they are small enough to fit through a window.

We were having all kinds of problems early on. Because we brought building in, say, oh, this is a great big huge thing. It was making four or five foot voxels, wouldn't fit through a window. We were getting no results. So those two things are really important inside of setting up Flow.

So as we move forward then, I want to start building some parametric data. And so this is a little bit talky. This is a little bit academic. But it's sort of an important thing to go through.

One of the reasons why AU is an important conference to me, is I'm always trying to figure out, not what I'm teaching next semester, but what am I going to teach 3 years, 5 years, 10 years from now, OK. Because it takes us a long time to get those curriculums in place, to know that whatever we're doing now is going to have the right ramifications for later.

And so one of the things that we are adding on as a core component of our curriculum is visual programming. All of our architecture students are going to start having visual programming as a compulsory part of learning. Now the interesting thing about that, technically speaking, is that that's going to marry up with the same semester that we do architectural programming.

So we're going to do visual programming and programming programming in the same class that I get to teach. And so I'm kind of struggling with that. We need new terms. We need more words to get these things done. But what I can tell you is, there's a really interesting opportunity to start pairing those two things together.

And so in terms of learning Dynamo, it's one of the first scripts that I dove in to start writing. And as I got into doing that, as I've gotten into working with Dynamo, the other thing that I started to work with is, why is this important? Why do I need to start teaching visual programming?

And the thing that I've really begun to come to grips with is, it's not about complexity and iteration to me. Because complexity is great, but I've seen the amazing objects, and I still feel like, yeah, but I can model that. I think I can model that. It might take me two or three different programs. It might take me four or five tries. But I can model that. Iteration, if I do it in Dynamo, I get more iteration, which is incredibly important. But it might not be the end all and be all.

But I think the next thing down on this list is artificial intelligence. There's no question that it's coming. You see it all over the conference. Everybody is talking about it. But as an educator, I'm trying to resolve, how am I going to start teaching that?

And I think the first key component to teaching how to integrate AI into a process is probably going to be through visual programming. That is we're receiving data. We're manipulating data. And we're outputting data. Which is exactly how visual programming works.

So to me, visual programming is kind of that gateway drug for the students to get into artificial intelligence and design. And so that's really kind of what sealed the decision for us. So, again, what we're looking at is, how can I do programming, and programming, and actually push that notion of the bubble diagram?

So that's what I learned when I was in school. How do we create a bubble diagram to represent space to help us resolve spatial programming in a design? How can we push that further? And how can we implement visual programming to do it?

So we're going to jump into Dynamo. And I apologize, this is a little bit tricky to read on this. But all of these videos are in the class handouts, and all of them are on YouTube. So go back and watch them. And I've expanded on them as well. This 50 minute talk unpacks to two and a half hours online for you to come back and watch later if you have any questions.

So step one, I just want to drop a rectangle down. So I brought in a node, rectangle, and place that rectangle by location, which is the CS part. So essentially, I have a length and a width. And the first thing that I want to do with this length and width is set up the parameters of my site, which for this one, I could do a rectangle. It's a rectangular site. I can keep it pretty simple.

So I'm going to set up two numbers. Those are not going to be variables. They're sort of locked in place. Those are my site constraints. 50 by 120. And I'm simply going to plug those into my width and my length. And Dynamo, I've got a site, super simple, OK.

The next step, when we're working with building programming, we know we have a set of spaces. And we know that we have target square footage for those spaces. So each space, essentially, is a variable. Length times width, and then when I think of it volumetrically, times height. So I can build a really simple math script to help me build that.

I'm going to establish 300 square feet, lead designer office. I can start using different nodes, then to build variables upon that, OK. What times what equals 300. So I'm going to use a math node. Basically, 300 divided by something, a variable, to output a new shape, OK.

And I want to be able to do that. I want to be able to vary that. So I'm going to use a slider. So a slider allows me to scrub through any different number variable to generate that. So I know I have my ultimate target size, 300 square feet.

I've got a slider that's going to be my variable. And then I'm going to be able to lock in a height. I don't always want to move the height up and down. It's an office it's going to be 10 feet. I'm set and ready to go.

So the next thing, I'm just going to bring in a division node. So I'm going to take my 300, and then divide it by the slider, OK. Pretty simple, simple math, elementary school math, my eight-year-old is starting to do this math. It's pretty exciting.

Except, it's the new math. They're putting it in packets. Which unsurprisingly, is visual programming. I hated it too. I heard the chuckles. I hated it too, until I made the connection. Oh, they're teaching math as visual programming.

So now, I'm going to copy that rectangle over again. And then I'm simply replacing my inputs, from the side inputs, to the length. And now, I have an office that's completely unusable, 1 foot by 300 feet. But I can use that slider and move that slider around. And what I do have is a surface that always maintains its integrity, its target value of 300 square feet. It's pretty simple math, OK.

I'm not going to spend this much time on walking through each Dynamo piece, by the way. But for people like me, that are still very new to Dynamo. I think spending a little bit of time on this first one is a really helpful thing.

So now, I wanted to do more than simple bubble diagramming with this. I want to actually see this as a more volumetric expression. So the next thing I want to do is start extruding that form up. So we're going to use a curve extrusion node and a rectangle.

Some of the things that you have to learn with Dynamo, is what a rectangle is defined as because the input and output values are really important. That helps you kind of understand how these things are strung together. OK, so I know I've got my height. How much I'm going to extrude.

And I know I've got my rectangle, which is technically defined as a curve, even though it's not called a curve. And that's going to give me that extrusion. So now I have a volume. I can adjust my slider. I'm always going to have a volume, that's always maintaining its integrity of 300 feet.

Now, as I start to build more of these, though, I'm just going have a lot of gray boxes, something that I can't really visually discern. So the next step to this script, is to go ahead and add color to it, OK. And you also notice in there, if I get a value that I can't divide by, you will start to see these errors pop up in Dynamo.

And I'll talk a little bit about this in a second. It's really important to understand Dynamo, at least far enough, so that you understand errors because it's going to be part of all of our workflows very soon, I think.

Oh, before color, I forgot about this part. Before color, I need to be able to move that box around. So that's simply translate geometry. OK, so I can take my geometry in. So my box becomes my geometry in. And then I want to build three more sliders for my xyz value. So that, essentially, I can move that office anywhere around my site. So I'm going to copy those sliders over, one, two, and three.

And then for your sanity later, for your sanity of the person you might be passing one of these scripts off to, label them, label them. In Revit, both these guys will tell you, label your 3D views. Please label your 3D views. At least, if nothing else, label your 3D views. And inevitably, I will sit down with them. And it will be, oh it's not this view. It's not this view. It's not this view. It's not this view. Because everybody forgets to do that.

But same thing in Dynamo, same practices, get these things labeled really quick. It's a simple double-click on the top. And then I can come back and label office x, y, and z. So that as I'm passing this on to somebody else, or as I'm doing edits a month later, all of these flyers are really quick and easy to find.

So then I'm going to drag my output node into the x, y, and z translation node. And now, I can start to move this around. Now if you're new to Dynamo, you'll notice, I've got this little extra box here. And what's happening, each one of these nodes generates a piece of geometry if that's that node's job.

So one of the next things that I do want to do is, I want to go back in and hide the preview of any portion that I don't want to see. I want only my final output here. So by right-clicking on any of those nodes, I can simply go in and turn off its preview. So that all that's left is that final output. So that's a right-click, turn off preview, select, right-click, turn off preview. And so I'm left ultimately, with just that final box in its location.

Now, as mentioned, color, we want to actually give that a little bit of tonality. So I can actually build a color palette that's based around the type of spaces that I'm trying to organize. And, again, when you think about something like Dynamo, when you think about working in code, or working and building these scripts, there are several steps, a lot of times, that you need to go through.

So this is display by. I want to display by color. But to get that color, I need an RGB value, and then something that turns that RGB value into an actual color itself. So it's not simply a single node that's going to do that. I need a series of sliders to vary my RGB value. I need a second node to turn those RGB values into a color. And then I'm plugging color to color for the display, OK.

So working through that really quickly, RGB values. And then I'm establishing limits on the slider as well. Saying, my minimum is zero. My maximum is 255. So that I stay within the range of what an RGB color value can be. And then I'm going to build, simply, the node that's going to translate those RGB values into a color value.

So you'll notice, on this node I have A, which is my alpha channel or my transparency, and an RGB value. So I'm going to plug my R, G, and B in to my color. And then I'm going to make another numeric node that is going to be my transparency or my alpha channel.

Because typically, with something like that, I'm not too concerned about a little bit more transparency or a little bit less transparency. I just give that a value. And that's usually something I'm pretty happy with. So we'll set that as 200. I connect 200 as a value in. And now, I have a semi-transparent purple box.

So the bigger question that I had in my pre-Dynamo days, is that was a whole freaking lot of work to get a purple transparent box on screen, OK. But the power of this is simply what I can do with this afterwards. All of these, copy, paste, I can come back into this, change that. This is my showroom.

I can change one parameter, 2000 square feet. Immediately, I've got a brand new box. So I can do that. If I've got 20 spaces that I'm trying to consolidate on here, boom, boom, boom, boom. I can set each one of those up and quickly iterate through multiple ideas. That's the power of this to me, especially in terms of teaching.

I've taught two things here that I'm really happy with for my students. I've introduced them to visual programming. And I've introduced them to thinking about programming more than a set of bubbles on trash paper. Although I love my bubbles on trash paper. But they're really getting into spatially, how these things start to come together. Ideally, they're thinking back to the data that we've worked with, the context that we've built, as well, thinking about how all these things come together.

So the next thing that I want to do, is I actually want to take all of this stuff, and then drop that into my FormIt 360 model. And that's why I'm doing this in Dynamo Studio and not the Revit-flavor Dynamo or something else. Because what Dynamo Studio will allow me to do is essentially go save to web, OK.

So the next thing I'm going to do is actually open up the final programming file that I put together for this demo. So let's just save this out as the WIP file, the Work In Progress file.

And then we'll open up the final three-dimensional Dynamo bubble programming file, and get something like that, looking at multiple spaces, multistory building. So my vertical circulation becomes an opaque box. But I want to be able to see those volumes to stairs and elevator, and how they're starting to tie all the spaces together.

So from here, saving it to the web. So once I push this to the web, I can open this exact same script. Not a three-dimensional, locked in place version of this script, but an actual script that I can continue to work with inside of FormIt. And we'll do that in a couple of different ways here. So we're just going to give this a name, Building Programming Test 5, publish it to the web, and then we'll switch over to FormIt.

So inside of FormIt, there is a Dynamo button. I'm refreshing that. There is Building Programming 5 that I updated 12 seconds ago. And I can bring that visual script in, and place it exactly on my site, and see now how my programmatic thinking, my building programmatic thinking, fits and works within the context of the site.

Am I making right decisions? Can I talk about those decisions with other people? Can I sit down at the studio? Can I turn this then into a graphic that's going to help explain how the programming is work, and how it relates to the site context? All really, really important things. All things that I want to teach. All things that when I was in practice, were incredibly important for me to communicate early in the design thinking to a client.

So I always am interested in working with my students on seeing three parts of building, space, structure, and enclosure. And I want them to see those things as independent working. So even though this is about predesign thinking, I like integrating structure into the predesign thinking. Because for architecture students if you can think back to when you were in second year, structure is a chipboard or mat board, something that the model is going to hold up.

The structure should be a part of the expression of the design. And if you can integrate that in early on, some really good things happen. So this is a script by Josh Goldstein, who is a Dynamo genius, OK. And I'm simply bringing that script in. So I downloaded that off the web, opened it up in Dynamo Studio, published it for myself on a Dynamo Studio. That sounded illegal. It's not. He shared it, OK.

So I can come in, and double-clicking on this, opens up all of the variables that I set up in Dynamo. OK, so all of these variables over here, on the side, are all elements now that I can see directly inside of FormIt, that I can manipulate, no different than I would be manipulating them inside of Dynamo, itself.

If there is something I don't want to see, like I love the extension lines on this, I think they're awesome. I want to keep them. I'm going to turn off as input. So if I turn that off as input, if I were to update this, publish this to the web again, bring that back into Dynamo, or bring that back into FormIt, I would not see extend grid lines anymore, as a variable, on there.