Dynamo from BIM Automation to Generative Design - Part 1 of 4

LILLI SMITH: All right. Good morning, everyone. Welcome to Dynamo Workshop, BIM Automation to Generative Design. We are really excited to have you all here today. Thanks for coming all the way to Vegas.

My name is Lilli Smith. I'm a senior product manager at Autodesk. I work in Boston. I have a background as an architect. And I get to work with this really passionate team of people who are really interested in improving technology in the AEC industry. So it's a lot of fun.

We thought we would start out today with a bunch of legalese. So please read this all and take it to heart. This also boils down into haiku form which is, you know, we don't make promises. We are going to show you-- seriously, we are going to show you some beta software today. And it's just important to remember that it's in beta and to not make purchasing decisions because things do change.

All right. So I am thrilled to be joined by this rock star cast of instructors that we have today. They're coming to you all from Europe. We have Sylvester Knudsen, who is from London. And Sylvester is a computational specialist at Metaspace. And before that, he worked as a-- in Denmark for MT Hojgaard as a virtual design construction specialist. So Sylvester. Sylvester is passionate about trying to make the AC industry better through automation and optimization.

Jacqueline Rohrmann comes to us from Frankfurt, Germany. She just finished her master's degree in civil engineering with a heavy focus in computation, and also had a smattering of architecture classes thrown in as well. She is, in December, she's moving to Stockholm, and she's going to work for Rumble. And she's excited about that. So you also might recognize her because she has a YouTube channel with over 10,000 subscribers. You can find her at That BIM Girl where she has tech insights, demos, and also has a lot of fun.

So Alvaro Ortega Pickmans is joining us from Barcelona, Spain. He holds a master's degree in building engineering from the Polytechnic University of Madrid. He is also with Sylvestre a computational designer at Metaspace. And he also spent several years working at Fosters & Partners. So he's known for having an open and curious mind and for pushing himself at every chance. So thank you for joining us, Alvaro. Oops.

OK. So the schedule for today. So I'm going to give a brief introduction. And then Sylvester is going to kick off the first block. So we've broken the day into four different blocks. And the first block is going to be about automating BIM workflows. So Sylvester is going to lead us through three examples there. Teach us some stuff.

We're going to take a brief break. And then we're going to start to get into generative design. And Jacqueline is going to give us some pretty basic starter exercises and move towards more complexity. After lunch, Sylvester and Jacqueline are going to give another block where we'll look at a specific office layout work flow and go through how to make that, how to add metrics to it, and how to use it with Project Refinery. And then the last session of the day, we have a really cool workflow that Alvaro is going to teach. It's a lighting layout workflow. And then we'll have beers.

There are two other Dynamo workshops going on concurrently. The second one is on Dynamo for BIM management and deployment. So it's all about what's going on behind the scenes of Dynamo, how to make scripts in successful ways to give them to other people. How you can teach Dynamo effectively. And then the journey of a project involves different steps along a project that ideas of how you could use Dynamo. These sessions will all be recorded and posted afterwards.

So if you're interested in any of these topics, you can check those out later. The third workshop is targeted more towards software developers. So it's going to look at things like Python, C# for beginners, how to make zero touch nodes in Dynamo, and how to make extensions for Dynamo.

So in this class we have 79 registrants representing 78 different companies. And I think architects are the largest segment. But there's also a bunch of building engineers, civil engineers, and construction professionals. So we're excited to welcome you all. We hope you'll find this useful. And then just in the other workshops, the total number, 231 total people representing 200 companies.

So why are we all here? So we all know that by 2050, there's probably going to be or there is going to be projected 10 billion people on the Earth. We're going to have to do a lot of work to build 13,000 buildings a day and wrap the earth 30 times every year with rail and roads. So I'm going to assert that there is a problem. We need to do more work at a higher quality with less resources.

This has been a traditional way of delivering buildings-- drawing it on paper and giving it to other people to build the building. We've evolved into digitizing these methods, making it easier to change, to deliver drawings to other people. Autodesk has been in this business for more than 30 years.

But what we really want to do now is to capture the intelligence in these professionals heads and combine it with the algorithmic opportunity in the computer so that we can create better designs and have the computer help us.

Autodesk has been experimenting with some of these technologies. We, a couple of years ago, actually bought in architecture firm that works inside of Autodesk research. And they did a project on our Toronto office which was moving to a new space, and they were really interested in using generative workflows to design that space.

So they began by surveying the existing employees and trying to come up with some metrics about what would make a good office. How could they measure the success of a good office? So they develop these ways of asking people, how much distraction can you take in your day? Who do you want to sit by? How much daylighting? Is daylighting good for your work, or is it distracting?

And then they also, you may ask, well, how do you measure low visual distraction? So they actually came up with some equations to measure this. So they defined it by the number of people that you could see from a certain point in the office divided by the total number of work spaces. And they use these methods on their existing space and tested them out so that they could then use them to design the new space.

Similar for adjacency preference. All boiling down into the scoring equation. Circulation. So then when they had these metrics in place, they could develop a kind of parametric parti-- an idea of how this building could flex and change, and ways to measure it. So here you see the kind of big blue bars are conference rooms, and in the other space, there's kind of open office. And so they could test out all of these different designs automatically and have each one be measured.

Once they had that working, they could go even a step further and they could declare their goals upfront. So if they knew that they wanted to have maximum daylight and best plan for adjacency, they could declare that ahead of time, and then use the computer to help them to narrow down the options that the computer was making to get to the best options.

The Living also worked with a construction company in the Netherlands called Van Wijnen. And they were interested in using some similar methods to do a neighborhood design. So in this case, they were looking at very different goals around backyard size, views, solar gain, cost, and profit, but using some of the same techniques of making this flexible model and being able to measure it.

But how can we do this? So these tools have been kind of closed. They haven't been very accessible to a lot of people to be able to do this. So Autodesk is really interested in trying to make these ideas more available to more people. So one way to start doing that-- so we need a way to kind of program and declare goals. So one way to do that is with Dynamo visual scripting.

So how many people here, I think we said-- how many people here have some Dynamo experience? Oh, wow. So a lot of you. Great. So you know that all Dynamo graphs kind of boil down to this very simple three parts, right? So in this case, we're reading some information in. An address, in this case. We're doing some compute in the middle. In this case, turning it into uppercase. And then we're writing it back out. In this case, turning it to uppercase and writing it back out to Revit.

So Dynamo is saving people a lot of tedium and time, and it's really gained a lot of popularity by being a tool in Revit and saving people a lot of time. So Sylvester is going to start out the Dynamo kickstart block of workflow automation examples by teaching some-- this is the first example he's going to go over, which is a room occupancy calculation. And here we're showing it. This is the-- let's see if this thing works. No.

The dialogue in the right is the Dynamo player. We're going to show you actually how to write this in Dynamo. But if you want to give it to somebody else who maybe doesn't want to see Dynamo and just wants to kind of play it as a macro, they can use Dynamo player to do this.

The second example that Sylvester is going to show you is about using information in a Revit model to do things like visualize it in terms of colors. And then the third example he's going to look at how you can interact with Excel as a checking mechanism or way of transferring data back and forth to use in your work.

So that's our kickstart block. Our first block. Our first 90 minutes. And after that, we're going to start to get into generative design workflows. Because we think that this knowledge that you're gaining in terms of being able to setup logic and use visual scripting can also be used in generative design workflows.

And our goals are really that we want to be able to learn from more designs than is physically possible to generate or evaluate by hand. So traditionally, in your work, you might come up with three different options. You might come up with 10 different options. But what if you could come up with hundreds or thousands of options?

So we have created Project Refinery. So how many people are familiar with Project Refinery or have tried it out? OK. So some. A good number of people, great. So we created Project Refinery, actually, for the Van Wijnen project that was doing the neighborhood designs. They came on stage at AU main stage on AU 2017 and told their story.

But they wanted us to provide them with some software that they could use for the next neighborhood design. So Project Refinery was actually born out of that. But once we had it for this one CUSTOMER we could make it into a public beta for a bunch of different customers, which is really our goal is to make it available to more and more people. So we put it out as a public beta last year at AU at AU 2018.

This is where you can go to download it yourself if you haven't already. And so we've spent the past year doing a lot of teaching and experimenting. This was a hackathon in London where people were looking-- some architects from Arup, 3D Repo, and Hopkins were looking at what shape of building addition can we make on a rooftop? And this was in London where real estate is super expensive. So what shape can we make up there? How much can we add, how much mass can we add so that you can't see very much of it from the ground?

We also had a summer residency program in Toronto where people came for two weeks. This was a project by Parsons in Dubai. And they were interested in looking at this massing study in this urban environment, and looking at trying to optimize each plot for minimizing sun exposure because it's in Dubai, and maximizing the view quarters.

This is one of my favorites. It's a liquor store optimization. And they're actually looking at the size of the shop, the depth and the width. Where the storage door could be located. Where the cash register should be located. All these different things that they were moving around and testing in their automated design exploration. And they say they were able to make better decisions because they were backed by data. They could do this faster, and look at a greater variety of options.

So in the second block, in Jacqueline's introduction, she's going to take us through-- this is kind of our hello, Project Refinery exercise. If you've tried out Refinery, you may be familiar with it. But it is just a really simple look at a basic massing study and how you can look at the maximum volume and minimum surface area.

And then she'll teach us how to get maybe a little bit more sophisticated in a slightly more sophisticated parametric parti of this massing study in an urban environment where we're going to look at some of the-- some similar goals of minimizing facade area and maximizing floor area, but also things like solar exposure and window-to-wall ratio.

The third session in our generative design exploration is going to be looking at this office layout. Jacqueline and Sylvester are going to teach this together. And they're going to look at how you can layout desks in an open office area, maximize the number of desks, how the work spaces are utilized, and the area per person.

And then the final session of the day is going to be a look at laying out lighting in an office space, and looking at how you can minimize the number of light fixtures but still get a really good quality of light over the space. Alvaro's going to teach us this one.

OK. So where are we headed with the Refinery? So right now, Refinery is really targeted towards the design technologists. And this class is targeted towards teaching you all how to make these scripts that you can run in Refinery. Maybe give to other people in your office to run. But it's really targeted towards people who are writing the scripts and running them themselves.

But we think that there is also a good chance that there are probably people in your office who are not interested in Dynamo and might want to use some of these techniques but are not, maybe not comfortable creating their own scripts. Want to reuse some things that have already been done.

So similar to Dynamo Player, we think there is a good opportunity to have a generative player where you can use these scripts directly from Revit and run them in the context of your Revit model. And then we also think that there could be some good opportunities for extending this functionality to some of our other Autodesk products, Civil 3D and Formit, and probably more, too.

So as well as extending kind of to this project designer persona, we also think that there is a good opportunity to extend to the project manager persona. So someone who is not even working in Revit or doesn't have these tools on their machine, doesn't really know how to use them, but still wants to use the benefits of looking at the results of these automated studies. So we think that it could be super interesting to be able to publish the results of the study in a way that your stakeholders can review the results. And then looking at ways to continue with cloud storage and compute.

So in our work to make some of these ideas more accessible to people, we are also developing the Refinery Primer, which is a good way to learn to read more and get started with this. We've been working on this the past several months and getting a lot of rounds of feedback. And we feel like we're finally ready to kind of talk about it, advertise it. And this is an open source project. You can leave your comments or your contributions to it if you are so inclined.

We are also-- OK, this is playing. So this is showing some of our latest version of Refinery that we're working on that works directly from Revit. And we're working on making the UI more friendly to people who would be using it in a Revit context.

We are going to be testing this at the ideas exchange here at AU. If you would like to do some user testing, we'd love to get your feedback on the latest version. We're going to be using a slightly-- our previous version today because we wanted to give you a stable experience and one that we could prepare our class for. But come see us an Idea Exchange and you can check out this new, new version.

All right. So this is our time-- we're also having some idea exchange sessions on Dynamo, and also some AI and machine learning experiments that we have going on in our group as well. On Wednesday night, we are having a party at Senor Frogs. So the Dynamo team really knows how to party. We are going to combine seven-minute talks and a party. So it's experimental. We'll see if it's fun.

But it's going to, they're going to be-- if you've ever been to a Pe-cha Ku-cha or Pecha Kucha, they're timed slide. So 20 slides, 20 seconds each. And it can be a lot of fun. We have seven presenters. If you are pre-registered, you can come in right at 8:00. If you're not pre-registered, still come. We're going to have a line for entry at 8:30 depending on space.

A couple classes of note. Dieter Vermeulen is one of our great refinery users. And he's a technical sales specialist in Europe. And he is going to teach a class that's being repeated on both Tuesday and Wednesday. I think he's also having a live stream of it.

And if you can't check it out here, if you're busy at this time or these times, you can check it out afterwards. But I think this is going to be a really good Class he gave me a preview of it, and it has a lot of really interesting examples in it. And he's going to post all the data sets as well.

Kean Wamsley is going to talk on Thursday at 10:30 about the Toronto office design. So they have-- Kean has put out on his blog the Dynamo script that shows how the Toronto office really works behind the scene. Oh, it's kind of complicated, but it's super interesting as well to learn from. And we're thrilled that there are 13 classes on Refinery this year here. So you can search. If you just search keyword refinery, you can find a lot of classes this year.

So that's it. I'm to hand it over to Sylvester for block 1.

SYLVESTER KNUDSEN: All right. Hey, guys. Can you hear me all right? Good, good. So first block of the day. Just a brief repeat of what Lilli just said. She just did the introduction, and we're now going through the first block, which is the Dynamo Kick Start block. Feels weird standing up here.

And what we're going to look at is just BIM workflow. So how can we automate things in Revit with Dynamo? How can we use Dynamo as a middle man to communicate between different applications? So for this exam, we're going to do Revit and Excel, and how can we merge those two things. And after that, we're going to have a small break, and then a very important thing. After that, 6 and 1/2 hours to free beers. So hope you're excited about that, because I am.

So is the screen jumping a lot for you guys, or it's just me? It's kind of annoying. So just quickly, the learning objectives of this class. So a lot of you guys had experience with Dynamo already, I could see. So hopefully it won't be too basic for you. But this is mostly just to get into the Dynamo mood and just get everyone on the same page on how we use Dynamo and how we create logic in Dynamo. So after the issue, you'll have a better understanding of that. And you'll be able to take maybe not the exact workflows that we did, but the concepts of the workflows, and take them back home and create some cool stuff with that yourself.

So just, Lilli already did it briefly-- we just want to know a bit more how many have experience with Dynamo or more like your experience level. So none of you are completely new to Dynamo, right?

AUDIENCE: I am.

SYLVESTER KNUDSEN: Awesome. So happy about that. Also I'm going to-- perfect. I already like you guys. What about intermediate level? Who uses Dynamo on a regular basis but not consider yourself expert yet? Good. And the rest of you are experts. [INAUDIBLE] Yeah, not you. I know you. Cool, cool.

So before we're going into this, I'm just going to go through-- and I'm going to have to sit up here, which is like, I'm kind of hidden away. But you don't need to look at me, so that's fine, I think. Did everyone get the frame working? Like, your machine is working. Are you inside of something that shouldn't look like this? If you're in this, you're doing something wrong.

Do we have a frame in here? Oh, something's happening. Yours should maybe like look a bit more like this. Does it look like this or did you already open the desktop? Good. So if you're on this, just go to the desktop. And you should see something that looks like a regular machine. This is taking a bit long. Good. We're back.

So yeah, you should be looking at something that looks like this. And if you go over here on all of your icons, there's this zero data sets. So if you open that-- let me just close this-- you will see a lot of different data sets. And it can be a bit hard to find because the numbering is super weird. But if you go to the DW 33 something, something, Dynamo for BIM Automation to Generative Design, you'll have the entire data set that we're going to use today.

So in this folder, you have the AU 2019 workshop content. And we have all of the different blocks. So from block 1 to block 4. And what we're going to do now is start the Dynamo Kickstart, obviously. And so if you open this, you'll see we have three different folders, and we're going to use these folders today.

So we have the Revit files which is just containing the Revit files that we actually going to use. And we have the BIM files, which is again split up to something else. So we're going to do some challenges. So we're going to build something from scratch. And some of the times, you're just going to look at me doing it. And if you want to do it yourself as well.

And we also have some really cool challenges where you guys are going to do it. And one of them has prizes. Like, look at this. So one of you are going to win this. So free beers and free bags. This is going to be good.

And so in the challenges, you'll see we have all of the things we're going to go through. The challenges are mainly scripts where-- some of it are put together, but most of the script is not put together. So you can see all of the nodes that you actually have to use. So you can use that one as a template if you want to. If you want to just start from complete scratch, do that. However you like it the best.

And the completed folder has all of the scripts, completely done. So if you fall behind or something didn't make sense, you can always go back and check these scripts. There's also images of all of the scripts completed. So you can also have a look there. And during the challenges, if you get stuck, obviously look. But if not, maybe don't because it's going to be too easy then.

Yeah. I think that's it about the folder structure. Is there any questions about that, or does that make sense? Cool. Also, there's like, what, 1,000 lab assistants in here today. So if anything goes wrong, just throw up your hands. I don't know where any of them is right now, but I think they're in here. So just put up your hands. And if you have questions, put up your hands. We'll take it from them.

Cool. So first thing we're going to look at is interacting with Revit. So let's go ahead and start up Revit. So if you are on your frame-- almost did that again. So don't do what I'm doing right now. There we go.

So go ahead and open up Revit. And inside of your Revit files, let's open up the-- let's see. Let's open up the-- forgot what file it was. Just give me a second. The advance sample project. Let's open up that one. And sometimes it is a bit slow of opening up things. But when we get into it, it picks it up again.

Cool. So let's just go to the entry level view. You should see something like this. And let's start up Dynamo. Which is a bit slow. Is everyone with me at this point? You're just trying to start up Dynamo? Cool. So you can now either just open a new one and follow along with what I'm doing. Or you open the challenge and have a look at it or open the completed one.

But we're going to go through it step by step up here for this example. So I'm going to go ahead and open a new workspace in Dynamo. And obviously, you are familiar with what's going on here. The library and all of those things. So won't go too much into detail about that.

But a really good way to start creating your graphs is kind of thinking from the end first. So what's actually the objective that we want to do? And in this case, in this example here, what we want to do is obviously and calculate room occupancy based on the area.

So does anybody have an idea what the last thing we want to do? So let me just open this. So what we want to do exactly is we want to have these rooms. And you know the rooms has a occupancy parameter. So what we want to do is we want to make sure that we know how many people can we have in this room?

So what would be the end goal of our script? Does anybody have an idea of that?

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: I can't really hear.

AUDIENCE: [INAUDIBLE] parameter in the room.

SYLVESTER KNUDSEN: Yes. Exactly. Awesome. So first thing we want to do is write something to the occupancy parameters. So let's go ahead and make some groups just because we want to split up our graph so we don't know like what steps we need to take.

So a good way to create empty groups is you can go up to-- I can't remember. Just Control-W, and you can create new, small nodes. And let's do two of them and group those together. So this is going to be our output. And we'll call that set parameter in model. And let's call that pink or something like that so we know that's the output. That's the goal we want to do.

So again, starting from the last part of the script. so we might want to find the node that we actually want to end up with. And does anyone know what that node might be when we're trying to set something in the model?

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: What? Yes. So again, we want to set parameters. And we want to set an element parameters. So the way Dynamo names their nodes is it's like the dot annotation. So if we put in elements-- element, dot, set parameters, as was just said. And just make sure they're spelled correctly. Parameter.

So we have a custom node here which also does something similar, but today we're just going to use the building node which has the set parameter by name node. So let's put that in. Let's just put it in our group right away. So this makes it a bit easier for us, because now we actually know what we need to do for the rest of the script.

So we know we need to find some elements. We need to find a parameter name, and we need to find a value. So we can start by making some group. So first of all, we need to find some elements. So let's make a new group. And let's not put anything in it. Just create the group. And let's call that get element. And in this case we know we want rooms, right? So let's just put that in as well. Rooms. Cool.

The next thing we want to do is we want to find a value. So we need to calculate the occupancy value in some way. So let's create another group as well-- we could just copy this one-- and create or call this occupancy calculation. Calculation. Is everyone with me at this point? Yeah, cool.

So let's try and also filter some rooms just to make this and the example a bit more interesting. So we only want to actually set the value in rooms that are called, let's say, instruction. So we need to only find the rooms-- the instruction rooms. So that means we have to add another step to our little script. So let's create another group. And let's call this one filter rooms. By room name.

Cool. So now we basically have our script. Like, the blueprint for our script. So let's put this in here. Let's try and arrange it a bit better. We know that get elements is probably the first thing we need to do. Then we need to filter some rooms. And then we have to create the occupancy calculation. So now we can start from the beginning.

And to select rooms in Revit, let's go ahead and find a category node. Categories. And let's pick the rooms. And for starters, we want to just find all of the rooms, and then we'll filter them after that. So when we have the categories, we can go ahead and say all elements of category.

And we can connect these two. And we can now see that we have a list of 91 rooms. So these are all of the rooms that are in our model right now. So let's add a watch node just so we can see what's going on. Let's add all of this to this group just so we have an organized. It's nice to organize it as you go just so don't get too messy.

So we can clean up this. And maybe get rid of these nodes because we don't need them anymore. Are we all on this step?

AUDIENCE: No.

SYLVESTER KNUDSEN: No. Oh. What's wrong?

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: Rooms. Oh, yeah, sorry. Can you actually see what I'm doing up there?

[SIDE CONVERSATION]

OK. If you're having trouble, just raise your hand and someone-- oh, shit. Everyone. Who else had their hand up? You did, awesome.

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: So if you-- the first node you want is the category node, right? So if use search for category. Oops. That one.

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: Yeah. So you can see this one is like a selection node for Revit. So if you just click that one, it should put it on the canvas. Yep. And then you can, in this drop down menu, you can select rooms.

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: Yes.

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: Exactly. And then this is just getting us a category, right? So we can't really do anything with this. But what we want is get all of the elements from that category. So the way we would do that is to search all elements of category. Yeah. So it's actually right there.

AUDIENCE: And then just double-click it?

SYLVESTER KNUDSEN: You can just click it, actually. Yeah. And then it'll just-- yeah. And then you can see this takes in the category, right? That's it's input. And we have a category here, so let's try and select it and put those two together. If you press-- yeah, just press there one time. Yeah. And you don't need to-- yeah, you don't need to have it pressed all the time. You can just click there. Yep. Exactly.

And then you can see, if you hover over it, you can see the output of this over the gray area. And then just go slightly down to the white one and pin it. Yeah. So you can see now we have a list of room elements, right? So these are actually the actual Revit elements you have there.

AUDIENCE: OK.

SYLVESTER KNUDSEN: Yeah. Awesome. That's the first step. Sure. OK, guys. Should we move on? Or is someone still having trouble? Oh, OK. Yeah. So is everyone-- everyone needs to start Dynamo from Revit. So there is also a Dynamo sandbox on the desktop here.

So we don't want to use that right now because you won't be able to do anything with Revit if you're in that. So just make sure that you have Revit open and-- sorry, that you have Dynamo open in Revit. And you can check that by looking up here in the left corner. You'll see a Revit icon if you have it open in the right place.

OK. Ready to move on then. Yeah? Well, we have to. So next thing we want to do is we want to filter all of the rooms. Let's just put this down here. So next thing we want to do is filter out the rooms because we only wanted the rooms called instruction. And there's a couple of different ways that we could do that. Anybody have a suggestion?

AUDIENCE: Filter by [INAUDIBLE]?

SYLVESTER KNUDSEN: Yeah, for sure. That's one of the steps, yeah. So how would you make that the mask?

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: Yeah.

AUDIENCE: Do a little text node [INAUDIBLE].

SYLVESTER KNUDSEN: Yeah, OK. So let's do-- you said filter. Filter by Boolean mask. So let's just put that on the cameras right now. And we want an equal equal, because we want to check if something is equal to instruction. And just another note. If you don't like going up here searching, if you're too lazy as me, you can just right click on the cameras. And you get this little thing up here. And you can actually search the same way here.

So we want the equal equal node as well to make sure that something is equal to instruction. So actually, let's just put instruction in here as well. So if you do a code block by double clicking on the canvas, you can make a string in that so you can put some text in it. Sorry. Yeah. Tell me if I'm not zooming in well enough.

So the value that we want to check for is instruction. Instruction. And let's just make sure that was right. Yes, it looks right. So if you do a code block like this in quotations, you will have a string, which is just a word, or more words. In this case, it's just a word. So we want that to-- we want to check something if it's equal to instruction.

So what do we actually want to check? What should be the other input in this one? Does anybody have an idea of that?

AUDIENCE: Parameter name.

SYLVESTER KNUDSEN: Parameter name. Or-- yeah. So-- sorry?

AUDIENCE: Room name.

SYLVESTER KNUDSEN: Room name, yeah. So how do we get the room name?

AUDIENCE: Get parameter value [INAUDIBLE].

SYLVESTER KNUDSEN: Both of them, right? Yeah. So we can do it in two ways. I think this one is new. So you said room name. Room name is a node. I think it's fairly new. I don't think it's been here all the time. I can't really remember. But this is just a easy way to get a room name, right?

So if we look at this, though, we see, OK. It takes in Revit elements of the category room. And luckily, we just did that. We have a bunch of rooms here. So we can just try this to see if that works. Yep. We get a bunch of room names.

The other way we could do this is by and by getting a parameter of value. So if we do element. Add Get Parameter. Value By Name. Right? So this is kind of like the opposite of our ending node which is setting a parameter. So this one is getting a parameter.

So we could get the same result by just doing the elements in here. And then we have a parameter name. So the parameter name would be whatever the parameter is called inside of Revit, right? And when we're using this get parameter value by name or set parameter value by name, it's very important that we're specific and they have exactly the same name as the parameters. We need to spell it the right way and all of these things.

So one thing you can do to make this easier is you can get a list of all of the parameters that an element has. So if we search for elements.parameters, we can get this little note here, element.parameters. And this takes an element and outputs a list of parameters.

So what we see here is-- this is actually all of the parameters that are available on this object. So we can check this in here if we don't want to go back and forth in Revit. For example, now, when we don't have two screens, it's super annoying having to go back and forth. So we can just do like this and we can check.

So someone said that we needed that the room name, right? So let's see if we can find-- there's something called name. So probably that's the right parameter. So we can go up back up here and make another string. And just say name.

And if we plot that into the parameter name, we'll see that we get, actually, the exact same output as the room name. So the room name is super easy because there's a dedicated node to get that parameter in Revit, right? But there's not a dedicated node to each parameter for each object. So most of the time, you'll have to use this one to get the value.

But for now, let's leave this here just so we know we can do it, and let's clean this up a bit because it's getting messy. So let's add all of these to the group. And we do that by selecting the node, all of the nodes, and the group that we want to add to. And then we can right click one of the nodes and just say Add To Group. If we clean it up, if you right click on the group and clean up the node layout, it looks not very much better, actually, because we need to do some more things.

And so we have all of the names, right? So these are the ones that we want to check. So we can make a check. We can say we only want the elements that are equal to instruction. So if we take these values down to the equal equal node, you'll see we get a list of Boolean values.

And we can now use these values to actually filter out the rooms. Because we can't set or do anything with these values. We can only use them to filter some other things that we can use to something. So if we have this filter Boolean, list filter by Boolean mask, and we hover over the mask, it says that this is a list of unions representing the mask. So obviously, that's our Boolean list. Let's put that up there.

And what do you think we need to put into the list? So this is the list we want to filter.

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: My what?

AUDIENCE: Rooms.

SYLVESTER KNUDSEN: My rooms, yes. So if we go back here, we have our rooms right here. So we can put these into our mask or our filter. And what you'll see, we get a dictionary, actually. So we have a in key and we have an out key. And the in is all of the true values. So you see we have 18 rooms here, meaning we have 18 rooms called instruction.

The out key is all of the false values. So these are all of the elements that are false here. The ones that are not called instruction. So this is-- filtering things like this is something that, when you use Dynamo forever, you do this all the time because you need to filter out in a list. Does this make sense, and is people with us at this stage? Any questions about this? Cool. No? Good.

So a quick thing that you can also do instead of using lists. Dynamo, in 2.0, we started getting more nodes for creating dictionaries, which is another way of managing your data, which is, in a lot of cases, better.

[SIDE CONVERSATION]

Is there--

AUDIENCE: [INAUDIBLE]

SYLVESTER KNUDSEN: Yes, sorry. Dictionaries. Back to dictionaries. So how many of you have worked with dictionaries in Dynamo? Some of you. Good. Not a lot. That's good. So does anyone know what a dictionary is when we're speaking, like, Dynamo and stuff like that? Yeah? No? Some of you. So a dictionary is basically just what it sounds like. It's like something where we can look up something and we'll get a value out of that.

So for example, in this-- can you guys see this OK, or is it-- let me do it like this. So in this example, we have a bunch of names over here, and we're just doing some splitting just to get a list of their first and last names, stuff like that. But what we actually have now is the dictionary where the key is a name, a first name. And the value is the last name. So a dictionary works with keys and values.

And the key is the identifier of the value, right? So if you want to search something, you can search by that value instead of using indexes. So in Dynamo, we have a bunch of cool nodes that can help us manage these things in a good way. So let's try and jump back and just look at how this would look if we did it with a dictionary.