Get into the Flow

TETSUYA HISHIDA: Hi, everyone. I'm Tetsuya Hishida, Senior Product Owner of MEP Design, and I'm really excited to present at AU. Last time I presented at AU, I was wearing my hat as a mechanical engineer all the way from Japan. And that was back in 2015.

So fast forward to the end of 2021, I made a big move, packed my bags and came to the States. So I'm super thrilled to reconnect with you guys. And I would like to thank all of those who gave me insight for my journey.

And our presentation is Get into the Flow, and I have a co-speaker. Martin, could you introduce yourself?

MARTIN SCHMID: Yes. Hello, everybody. My name is Martin Schmid. I'm one of the Product Managers here at Autodesk. I focus on MEP functionality within Revit, as well as our fabrication portfolio.

TETSUYA HISHIDA: Hey, I believe this is the least exciting slide. We do have contents referring to future development, but we cannot guarantee it so please be aware of it.

OK. So the learning objective-- first, you want to-- we introduced a new solver. We'll touch the details in a second. Next, this new solver would allow an easier way to configure connectors and MEP families. So this connector configuration used to be a headache for lots of content creator. We can simplify settings so I think we can make life easier.

Third, I'm going to show what we're working on. And as overall, I think this will help you establish your workflow.

MARTIN SCHMID: All right. So I will start with going over a high level strategy of the MEP space here at Autodesk, and where our various classes fit and, in particular, this class. So Autodesk's overall mission, so this is off of our corporate website, is to empower innovators with design and make technology to achieve the new possible. We all know construction continues to transform at a very rapid pace and we're witnessing a shift towards more cloud connected data-driven approaches to address demands and challenges on the industry as a whole.

And so our overall MEP strategy is to address some of those needs. So our overall mission and vision for MEP is to deliver a higher performing, more sustainable MEP systems with integrated design analysis, coordination, documentation, detailing, and fabrication. So the core concepts here is really that piece of integrated and bringing all the various stakeholders of a project together working on a common repository of data, ultimately, in order to achieve very subjectives throughout the life cycle of the project where the outputs of one stakeholder throughout the chain become the inputs to the next stakeholder on through the process of any given project.

And as a result of that, we've also been talking for a number of years now of these three rings that we share and are part of those overall workflows, the three major stages of the design process. The first being design and analysis where engineers are involved at early stages and coming up with the core concepts, doing the initial preliminary analysis, and driving initial decisions around the structure and how the overall project will-- and building systems will function.

The next is coordination and documentation, and that's kind of historically where Revit started being used to create 3D models and produce documentation from that, and is a significant part of how it's used today. And then the end of the spectrum there, detailing and fabrication, this is kind of the next stage where once the design process is completed and that goes into that next stage where contractors are leveraging that information to inform estimating, detailing and, ultimately, outputs to fabrication out to the field for construction and installation.

So recognizing all those various stages of the process. And not only that, but also thinking about the needs of electrical, as well as mechanical-- mechanical including piping and plumbing and so forth. We have a number of classes this year at AU that kind of touch and connect all these various themes together. So those are certainly not all the MEP classes at AU this year, but a certain key set of those are really kind of playing together on this theme of where the fit on this overall diagram of needs for electrical, to mechanical, and across the various stages of the design process.

So hopefully, you have a chance to check out these other classes, both online and in person down the road. So with that, I'll turn it back over to Tetsuya.

TETSUYA HISHIDA: Thank you. OK. So new flow solver-- learn how the flow solver works in duct and pipe. We introduced the new flow solver, and we borrowed the technology from our manufacturing software, called Moldflow. Some of you might know it.

But technically, it is a solver that can handle very complicated cases. And don't be scared, we tuned it up specifically for building engineers. And today, we will demonstrate how it provides value to your workflow.

We heard a lot of voices around flow and mechanical design, like, why can't you make the flow work in a smarter way? Why can't you make it work in fabrication elements? And as an ex-engineer, I know the fact that fluids used for various things, and also it is literally the key information to hand over from design to construction, and also to operation.

So we did it. In Revit 2024, we released two major capability in terms of flow. And both are derived from this new solver. There are tons of value, we believe. But to make it simple, we grouped it into three buckets.

First, background processing. Second, network complexity. Third, supporting fabrication. I'm going to, obviously, touch those in detail.

But the first one is to do those calculations in the background quite-- just it's obvious. The second one, we are going to support much more complicated design conditions. The third one, we will support flow and fabrication elements. Maybe fabrication is something named a bit misleading. We believe fabrication can be used also in detailing design, or maybe some design phases. And that's one step forward to realize those workflows.

And I mentioned it's a new solver but, actually, it exists on the piping side for more than five years. First, it was more like a small new option for closed loop. But actually, this is not only for closed loop anymore.

For example, ducting tends to be open loop, and the air is released into a room via terminal, and coming back via plenum or so. So this option on the mechanical settings, we put this small button on, and the new solver would kick in. You will love to know why you would want to stick to the legacy mode, so maybe your question was, why do we keep it? Please let us know.

We can't technically remove it and automatically upgrade it to the new solver. But we did consider throughout, but we don't have access to every single model throughout the globe. So maybe you can see our urge, but please let us know if there's anything you're worrying about.

OK. The first topic, background process, compute data in parallel. So the first one, left-hand side I'm showing what was there in the legacy. And right-hand side is how it is in the new solver.

In the legacy solver, we did have some sort of mechanism to make calculation faster. But the key thing is that every modification to the network results in recomputation, which can be time-consuming. In the new solver, we are now doing in background, and the new method is using this paradigm of background processing.

There's this circle on the right lower part of Revit UI, if you click it, this panel would appear. This paradigm is not only for MEP, we use it also for something like color-coded color scheme, or like PDF exports. So we decided to utilize those.

And one thing to note is that it's a bit technical. Maybe you can cover it on the handout. But there is a process called revit.exit. And also, there's a process called revitworker.exit. And this network calculation works on the workers exit. So it means that it's a kind of a parallel process. It doesn't hinder the other things they're doing on the molding side.

And processing is done in the background. The new method will still compute flow and pressure drop without preventing you from making further edits. And maybe when you're dealing with a gigantic model, you will see this calculating on the property palette.

But it's going to disappear very, very smoothly. And I think it's almost stealth. You wouldn't even notice these things happening. So don't worry about those.

And the next one, network complexity. So I'm going to deep dive into seven examples today. And the first one is diverging and re-converging. I think there are various cases this might happen.

I heard that you need to divide the ducting into size of the filter, especially, this is common in pharma industry. The case shown in the image below is a case very common in Japan where I came from. So it's a duct splitting into two smaller parallel branches at beam pockets.

In Japan, the plenum is insanely tight. And to make things worse, we have worse, we have earthquakes. So the beam is huge. And with the original solver, note that the flow stops at the first T upstream of the terminal. This is because the original server doesn't expect such a condition.

The new solver on the right-hand side, you see the flow continues beyond the parallel routed elements. Now, some with a good eye might catch that the flow value is kind of odd at this juncture if it doesn't send the coefficient of fittings to calculate flow. So although we are sending the pressure loss at the straight length to the solver, it might not be the best result for this case. But the important part is that the calculation doesn't die here and it will go through. So if the flow image on the right is like a super critical thing for you guys, please let me know.

And example two, system mismatch. There are cases you want the flow to propagate throughout different systems, which might be caused intentionally or unintentionally. The below is a slightly artificial example of this. So you see hydronic return and supply connected.

And on the left-hand side, it's how it was. For anyone who has ever used flow before in Revit, I think you are-- I was also one of those so we might share that. Frustration seemed to flow value in these cases.

But on the [? right-hand ?] side, you see that flow is actually propagated through. And there is a small tweak in the family side to make this happen. So I will touch those in the following chapter.

Example three, primary secondary loop. Now, things get more in thermodynamics. There's always this simple equation showing the relationship of flow and Delta T, which is temperature change. So uppercase Q the left is flow volume. It means that if you deal with the same heat, less flow means larger Delta T. Because Delta T is on the right bottom there.

So for boiler, which literally boils things, if you send the same temperature, people would get burned. So in this case, suppose the necessary heat transfer was 165 kilowatts if the Delta T is as shown.

43 degrees in, 90 degrees out in boiler. 50 degrees in, 43 degrees out in terminal. The flow volume would be different. Flow volume would be 3 cubic meter per hour on the boiler side, and 21 on the terminal side.

So there are several ways to make this happen. Maybe there are valves or other things that you would want to identify in each cases. And the pump position might be different, slightly, depending on what you want to achieve.

But the important thing is in the Legacy solver, which you see on the upper part, flow is color-coded as 0 at most of the part. So it kind of make no sense. And if you see the lower example with the newer solver, it's calculating throughout. So this is because the new solver goes out through all the network and calculates every aspect to make things straightforward. So that's why we called it network-based solver.

Sample 4, primary/secondary branch loops. So having a primary and secondary branch like this and configuring a more complicated system is super common in hydronic systems. So the Legacy solver is more like a simple elementary school math so it cannot solve these cases. And when you go into this new solver, you'll see that in contrast, it provides more rational results.

Example 5, backup pumps. So this option is valid only for the new solver. But there is a concept called equipment set. Equipment set is when you want to set a backup pump, which is-- well, you might call it a standby pump. But there's 2.5 cubic meter hour per each. But the flow propagated from it. It's also a 2.5.

If you try to use simple math, it's going to be 2.5 plus 2.5 and 5. But one of it is going to be standby, and one is on duty. So this is how you can configure backup pumps. There's an image on the left-hand side, you see constraints on duty is marked as 1. On standby it's marked as 1.

And Example 6, parallel source equipment. This is the Legacy solver version. But there are cases where you might want to size different pumps in parallel. Like in this case, chiller is 6 cubic meter per hour versus 12 cubic meter per hour. There are two chillers, and there are ways to solve these in the new solver.

And it's using this ignore flow analysis property. In this example, it should be noted that the valve between two pumps are set to ignore flow analysis. This is kind of working as a shutter valve, or isolation valve. So the piping exists in the real world, but it's not intended to send out flows to each other in this case. So I put this ignore flow analysis check on, and it made all these network calculatable.

And example 7, this is adding a bit more complexity. Here's the same layout, similar layout, using some backup pumps. And small internal pumps are 100% redundant and while the larger chiller pumps are 50% redundant. And we can also tag those. It's a bit small, but you see duty 1, send by one on the upper side. Duty 2 stand by one on the lower side. And that's how the math works.

Because on the upper side, you see 6 plus 6 equals 6 because one is standby. You see 6 plus 6 is 12 and one standby. So you see 12 propagated. And this is possible only in the new solver. And just for reference, this is how it is on the old solver. So just again, Legacy solver doesn't support these pump sets.

OK. So the third topic, fabrication element supporting flow. And one important thing we use flow for is sizing duct or pipes. I hope some of you are familiar with the ductulator. Recalling my days as an engineer, I was handed this ductulator, on the far left, and it's quite expensive, surprisingly.

And I googled a bit and found out that these are quite commonly used among other countries. And for me, it used to be a kind of a magical tool. But after I learned a bit, I realized that there were a lot of assumptions there.

So there's a key-- there are key parameters, such as roughness and fluid, which would define the friction loss quite significantly. But in this ductulator, which is an example I brought from Japan, it's assuming that it's steel. But for example, if there's maybe sound-absorbing insulator inside of the duct, the roughness is totally different.

But this ductulator only works for those cases. I know that in other [? geo, ?] you have cool Excel spreadsheets to modify it. But to be honest, I've seen very few people actually going into that settings and changing it.

In Revit, you need to be kind of aware of what it is in the property. I'll show you in a minute. But roughness would be defined per duct type and also per segment type-- per pipe segment. And fluid, the ductulator assumes that it's 20 degrees, 50% humidity. In Excel, it's the same. You can modify it, but it's kind of manual. In Revit, you can associate those values in air and hydraulic.

So the roughness, where is this information? There are lots of properties, but this key information from roughness stays in this type property here. So if you open up a duct and go into type property, see this roughness, 0.09 millimeter.

And we also have this for fabrication elements. Fabrication elements have it as grayed out. You see in the middle, it's 0.09 millimeter. It's the same. But this is derived from the fabrication database. You might be able to refer to other classes and check how you can go into those database and modify things.

And fluid-- design element on the left-hand side, we have these in the mechanical settings. We'll be able to set dynamic viscosity and density, which is the directly used parameter for the calculation. And we introduced a new tab in the fabrication settings. If you see on the right-hand side, there's this third tab, fluids.

And we noticed that quite a few users actually define those. In fabrication data set, it was not used so much. So we noticed that maybe we can just allow users to map it to what they already have in the design environment, which is Revit. So you can go into each services name and service and associate which fluid you want to associate-- which you want to link it to.

And friction factor. Sorry, this is a bit technical. But as an ex-engineer, I was a bit confused, OK, there's this equation called Hazen-Williams and also, there's something called Colebrook. What's the difference? And I'm briefly touching about those, as well.

So friction factor is the major thing defining the loss at the straight segments. And Hazen-Williams is an empirical relationship, empirical equation, which means that it's not pure science. It's like a statistics. It gathers some values and made relatively true-ish equation. But there's a lot of assumption there, as well. So it's water for a certain temperature.

And what we use in Revit is more using the computational power to make it more work in various fluid types and various temperature. You can even choose which method you would use. And those are in fabrication from Revit 2024. If you choose one element, straight element in the fabrication part, you will see this friction 0.0965 pascal per meter, for example.

This is the key thing you would want to know when you size a duct. So if this is to the friction loss is-- if it's too big, you need to make the duct bigger so that that loss is going to be more negligible. So you can go into this parameter instead of open up your ductulator.

And also, I brought you one recording of the model. This is the Snowdon Tower, and we use this among those other classes, as well. So I'm just showing a sample SDK, which we sent with-- which we provided from 2024, Revit 2024. So you can even visualize the flow there.

And I notice that, OK, this branch doesn't have flow. And went in to investigate it and found out that, OK, there is a cap on top of the branch. So we calculate it and see the flow arrow again. This is one way to show how the flow propagates through fabrication at this moment.

We're working on more sophisticated non-SDK, or non-addon way to show these. But this is at this point where you can play around. And even you can maybe build your own tools to conform to your local regulations.

And as an example, you can export it as a CSV file, and it's totally configurable if you open up your Visual Studio and maybe change some of the parameter or order, or how you want to show it. And if you are not the guy who wants to use Visual Studio, maybe ask some of the member, or some of the third party, to build their tool, or just use this SDK, for example. And you can start maybe playing around with Pivot Table or those tools you're more used to to investigate through what's going on in the network.

OK. So I just mentioned SDK. Some of you might think, oh, what the heck is SDK? It's shipped from this web page. I've shown this QR code on the left-hand side. So there are lots of sample codes you can leverage.

And the particular example I've shown in the video is called network pressure loss report. So you can go in and use that as your reference and maybe start modeling your own work flow. And for the fabrication database, we already started to utilize this manage.autodesk.com. You might call it Autodesk access, so whichever is OK.

There is a small icon on this toolbar. And click it, it will bring you to this web page. But there's one small detail. You need to go into this view details and check these libraries, and you'll see imperial contents and metric contents.

So to be honest, I think for designer, it's a great idea to start working on these fabrication elements. And fabrication, as I mentioned, might be a wrongly named concept. It's more like a smart contents and smart way to model things which you can even send to Cam and cut through. So it's much more easier for maybe fabricator or detailers if engineers are capable with these tools. So please give it a look.

OK. So I've gone through the three major value we would provide through this new solver. But the secret recipe, or the secret sauce is, you need to configure the connector. And it's much easier-- it's been a while since Martin made this class in Revit Autodesk University 2008. It's a classic course. I also saw it and it's a foundation of what we're talking today.

So basically, the idea is that you first need to understand what kind of thing you're modeling, and go into each configuration for connector or properties. But it's been a while since that presentation, and we did make it a bit more easy here. So there are two key properties in blue rectangular here.

Flow configuration, which is preset or calculated. Or system, which is something I'm going to touch briefly later. And also flow direction in and out. And there's-- sorry, this one small rectangular I forgot to mention.

You can associate parameter to the flow value. So those three things are what you would need to be aware of. On contrast, the red rectangular is something you might be able to make obsolete from this contents.

You see that flow factor is something you use when you use system as a flow configuration. And also flow system classification, as I mentioned, is not used in terms of flow calculation. It might be used for color-coding or other documentation purpose, but just for the sake of flow, you don't have to mess with it.

And system, there's this preset calculating system. It's kind of a difficult concept to understand, so please refer to our handout. And it's a concept to enable parallel pumps, or maybe, potentially, somewhere around not connecting things together but calculating flows. But it's a bit heavy topic to touch today, so please, again, refer to our handout for this concept.

And going back to the important properties which you would need to take care of is this flow direction. And to ensure the flow value on the-- so this is where you want the flow to be propagated across this equipment. You need to do this to ensure the flow value is going to be the same from inlet to outlet, which is this flow direction in and out. You can do either of following two things.

So you can link the connector, or you can associate the flow property to a same parameter. It's important so I'll repeat it.

The basis is that the connector is in and out. And there's two ways to make the flow propagated across this equipment. You can link the connector, or you can associate the flow to the same parameter.

So it's a bit still complicated, right? And there are cases where flow configuration can be set to preset and you can link it to calculated. I'm not touching into the very, very detail, but there are lots of contents configured like this.

But take a step back and think a bit. So preset, it means that you already know the value. But within the same family, it says, OK, I don't know this value so I'm going to be affected by the other things.

But if you link it, it means that Revit is one-- on the one side, it says, OK, I'm going to grab this value from this family and propagate throughout the network. But on the other end, it says it's always calculated. So it gets confused, literally. And we're still working on some mechanism to warn, or guide people making contents. So maybe you can give us feedback how you want to get notified.

But there are cases where we do not recommend to link preset to calculate it. Or link a associated same parameter to the flow for these connectors when it's set to preset and calculated.

It was a bit technical, but that's the slide to wrap up my part. So I would pass it back to Martin.

MARTIN SCHMID: All right, thank you, Tetsuya. So now we will cover the next steps and connecting to that vision that we talked about at the beginning of connecting all the various stages from very early conceptual design, all the way out through fabrication. So as you saw, one of the new things in 2024 was starting to get that flow on that design data on the fabrication elements. But of course, there's more to do, so let's look at what some of those things might be moving forward.

So when we look at connecting the workflow, and we look at the pieces that are in place, some of the things to be aware of are some of the new capabilities that have been introduced over the last few years related to the energy analytical model through the systems analysis capabilities, and using that to conceptually define the system analysis model, using that for heating and cooling loads, and as a result of that, starting to get some of that fluid data.

So the green boxes are representative of functionality that's in place that has that connection of starting with data from the model and carrying it downstream. And then the things that are in blue are things that are kind of disconnected in the process, if you will. So if we refer back to the flow equation that Tetsuya shared earlier, we talk about that volumetric flow rate, and we were talking about connectors.

And one of the key things about those connectors when you look at air terminals, or hydronic coils is, it's up to you to make sure that you place that value on those components. But that information is coming from earlier stages in the process that now exist in parts of the Revit workflow. So kind of the vision moving forward here is to start taking the results of that computation and using that to populate, and to connect that data within the model.

But it's not just a matter of taking the flow information, because there's other inputs in that as the equation shows, right? So there's the output, which is ultimately the volumetric flow rate. But some of the information is still missing.

So when we look at, say, the different types of fluids, different types of glycol, whatever the other fluids that might be in a hydronic system, we would also need to be aware of the specific heat at various temperatures. And that's not a property that exists in Revit yet. But those are the types of information that as we continue to grow the data models, and so forth, that the software is aware of that that could be introduced into the connected workflow and become part of that critical equation.

And then, similarly, another piece that's in there but not connected to using that to compute flow is the information on the systems themselves. We have the supply side of the system, the return side of the system. And those have temperature associations [? with ?] them.

Today, those temperature associations are just used to compute the pressure drop associated with that. But it's not, of course, used to compute the flow. So again, these are things that we're looking at how we start to intertwine more of this information, connect these workflows so that there's less manual data entry and connecting these various aspects of the workflow.

So when we kind of zoom out a bit and look at a little bit of the bigger picture and we talk about what those pieces are, again, I referred just a moment ago to the systems analysis capabilities. And so again, if we look at this kind of a bit broader perspective, we're really thinking about how we connect that analytical data model, ultimately, out to the physical model to have that design model inform the downstream result of the design process and that deliverable outcome that's ultimately used for construction.

So again, part of those design inputs are this pieces from the systems analysis defining the system zones, and the spaces, and the surfaces, and the materials that go into that, all those parts that go into the energy analysis, and to the heating and cooling loads. And as alluded to, we have the analytical systems now where you can start defining the key analytical components, or the key system elements, the different source equipment, chillers, boilers, so on and so forth. And then the terminal components, such as the coils, such as other components within the system.

And then, of course, all those things get connected by way of the air loops and the water loops. So again, conceptually, the systems analysis defines the data structure around capturing all that conceptual design data. And then, ultimately, the vision there is then to use that information, use that for the inputs into the computation. The results of that computation then should be driving the output, taking the results of that, using that as the flow information that is then used to size the duct and pipe throughout the network so, obviously, more work to do there.

But these are the types of things that we're talking about when we start talking about connecting the analytical data that's being captured now with that kind of first ring of the circle, all the way through to the physical model. And the goals around that are to support more automation, more data validation, more iteration. Anytime today if there's a change to the design where some classrooms are merged into a bigger classroom, or where a large conference room is divided into smaller offices, right now, that's a completely manual process to propagate that information, redo your heating and cooling loads, and then propagate that through into the components within the model. Whereas, as we continue to capture more and more of this data at that early stage of the design process, the more easily we can use that information to support that model iteration and design requirement change.

And then, of course, along with that, there's still more to do on the flow and pressure drop side of things. So again, specifically, as Tetsuya had alluded to, we don't have the friction loss coefficient information on the fabrication fittings yet. That's kind of a next stage iteration that we're working towards.

But we're not looking at just taking the way that it works on the design elements and moving that forward, because we know that there's challenges with the way that that works on the design elements today. And so there's three bullets there on the input side of that process that are challenges. So the first one, there are cases where it's just necessary to manually input the loss coefficient. It might be a unique fitting, it might be something that the condition doesn't have a table entry for, whether that is information from, say, the ASHRAE tables, or for the piping tables that exist in Revit today.

But one of the challenges there is that manual input only works today when you have a single path, so something like an elbow, or just a straight through fitting. You could see-- on the bottom image there, you can see that you have the additional options of specifying a specific coefficient, or a specific loss. Whereas up above on that middle image, you see those options do not exist on the T.

And the reason for that is, there's no atomic place to enter a value for the two different paths that exist going through that T. So that's one of the challenges that we're looking to overcome as we continue to iterate and provide more flexible capabilities.

The next is user definable tabular data. So again, I alluded to ASHRAE tables. Presently, there is no way for you to define your own content tables. So if you don't-- say you're in a different Geo and you want to use reference data from somewhere else, there's no way Autodesk we can provide tables for every Geo around the world and then keep those all up to date. So ultimately, what we want to do is provide a mechanism by which those tables can be user definable insofar as that you set them up once, and then you set it up with your content, and then you use it, obviously, over and over and over again. But then also looking towards the opportunities by which then different user groups around the world might come together to work to define standards for their regions, and for different geos, and so forth. So ultimately, we want to provide the flexibility to allow that table to be maintained in a way that it's not hard coded within the software.

And then the last bullet there is fitting table and association-- fitting/table association. And what's meant by that is, today, the way that pipe fittings work is slightly different from duct fitting. So when you set up pipe fittings within Revit, you can go into the family definition of that pipe fitting, and you could preset which fitting table, essentially, it's associated with. Whereas on duct fittings because of the much more variety of tabular information, and the same component might actually refer to different tables in different states, it's an instant driven setup process.

And as a result of that, we don't always get it right. There's cases where you go in and you say, well, it's selected the table for the mitered elbow instead of a radius elbow, or whatever the case may be. And so as we move forward, we're looking at some of those challenges and looking at how we can embed more flexibility, again, for you to, again, control on the content which tables you expect to use in certain contexts. So that's kind of a big lift to get to the point where we provide far more flexibility to make this much easier to use moving forward.

And then as Tetsuya also alluded to on the output side, he showed you a variety of different diagrams that were color-coded. He showed you the SDK example of extracting that data. And these are all different ways of viewing the results of these different computations. Some of the built-in functionalities are, of course, the system inspector.

Some of the data is in the system browser. There are the pressure loss reports. There's the color fill diagrams. And not all of these are wired up yet to the fabrication elements. So again, as we're looking at how to make this data all more digestible and consumable, and more rational and easier to understand, we're looking at ways to modernize those efforts to make it more streamlined for your use.

So again, a lot of different things looking at as we look to evolve this whole flow and friction and pressure drop workflow, considering all those things as we make those iterations and investigations.

So the last slide here is related to, how do you get involved with [INAUDIBLE], and how do we solicit your feedback? So there's a variety of different vehicles by which to do this. The first one is, the Revit idea site, right? So if you're-- again, there's QR codes on all these to take snapshots of those and maybe save those links for future viewing.

But the first one is Revit ideas. And this is just kind of a playground by which we see all sorts of users kind of contribute their ideas. Might be very small ideas, might be very large ideas. There could be some engagement and conversation, dialogue between different users building upon one another's ideas. But also, there's the opportunity to vote on different ideas there.

And this is really important for us. Even on the support organization, sometimes support will drive people to the idea site. And they said, well, we just have this conversation with you in support. Why don't you provide that to the product team?

It's a very good point but, also, the other side of this is also additional data that we don't get when it's just surface to the product team, and that there is that additional dialogue. There's the customer voice in there. And then, also, we see the voting on it so we can see how big of an issue is this for the community. Are there hundreds of votes, or is this something that's in the dozens of votes type of thing? So that's one of the reasons we have the ideas is to use that as a gauge of how important different things are to the overall user base.

The next one is the other way around is the Revit Public Roadmap where we provide information on what it is that we're working on. This is kind of broken down into different kind of levels of horizons. So there's things that are in progress. There are things that we're working on now. Quite possibly things that you could see in our preview site, which we'll get to in a moment.

Then there's the what's next and later. So those are kind of longer term things that might be a bit more speculative. But still, there things that we're looking at ramping up and starting to work towards. And then, of course, there's the launched, which also just provides a summary of things that have been recently come out.

So again, reach out, check out the Revit Public Roadmap. There's roadmaps there that are broken down for MEP for structure, as well as for architecture. So again, save that link and refer to it back periodically to get the latest information on what it is that we're working on.

Then there's the Autodesk Research Community, and this is a place where you can go and sign up to get involved in potential future research activities. So all the work that we do, all the various teams, development teams working throughout Autodesk are involved in various forms of user research. And so this is a great way to get involved and help us understand what it is that you're interested in, what's your role, so that when a team is doing research on maybe friction loss or flow, or whatever the case may be, that they can kind of distill from your profile what it is that you're interested in, how often you're interested in being contacted, and be able to reach out to you to-- whether it's a survey, whether it might be a user interview to deep dive on a topic, or there's a variety of different other ways to get involved. And that's kind of the gateway into those kind of conversations.

Then the next one is, as I alluded to before, there's the Revit preview site. This is where we actually post a monthly build of functioning prerelease software of Revit, and other products, of course. But being our focus today is on Revit. And so again, this is a place where, again, you could sign up and periodically come in and see what's the latest and greatest. There's feature summaries of things that we're working on. And again, it's a great place to engage in conversation, provide feedback, and see where things are going.

And then the last two bullets there we don't have QR codes there because they're related to some of the other entry points. So inside the factory, these are in-person events, as well as virtual events, where we have customers involved in hands-on testing, whether that's in one of our offices, or else like I said, also periodically, we do these virtually. And it's an opportunity for people, again, to get face-to-face time talking with the developers, talking with the various team members on where things are going, how they're working, and to gauge are these things ready to ship. Are there any major concerns, or are there things that we need to be aware of, so on and so forth. So it's another place to get involved.

And again, where we post solicitations about inside the factories and let people know about the opportunities that are coming up are by way of the Revit preview site. So through the Revit preview site, we'll have postings on there when there are activities coming up for inside the factory. And then the Revit sprint demos are another vehicle by which people can get involved.

This is kind of a more ongoing venue where a particular development team will have-- they work in what we call two week sprints. And at the end of each of those sprints, we have a feedback session where we involve a handful of customers to provide feedback on that feature set. And it's typically where we want people to be involved in an ongoing basis to maintain context to see how the functionality evolves over time, and get that ongoing feedback.

So again, there's not a QR code for that. People typically get involved through prior research identified through the research community or, again, through the Revit preview site. So please reach out, please get involved in those various ways. Let your voice be heard, and we look forward to hearing from you.

So thank you very much for your time today, and that wraps it up for us today.