In this article, we’ll look at how to make virtual reality (VR) a reality in your design firm — with a small hardware investment and some inspired team members. Starting with understanding the different possible solutions, we’ll look at how to use the technology throughout all phases of design and construction. We’ll also look at how good model management is critical to the success of the virtual environment. By the end, my goal is to have you feeling better about deciding on a road map to implement VR at your practice.
Where It All Starts
Deciding on where and when to start down the path of creating virtual reality content is probably one of the hardest decisions to make when looking at the current emerging technologies market. The reasoning for this is because of how rapid the technology is advancing, how many companies are getting into developing hardware, software, or both, and the seemingly limitless ability we have to tie in different systems, such as user tracking and construction technology, into the virtual environment.
Choosing a Direction
So, with such a rapidly changing environment surrounding VR, how do we choose the best direction to move forward? I wish I could tell you that there is a simple answer for this, but sadly there is not. The truth of the matter is that you need to have an idea of what you would like your end-product to be as well as do some examining internally of workflows and processes to determine the best solution.
End Products
Depending on your familiarity with VR, you may be asking what are my options when it comes to end products. Let’s examine a few of the different possibilities and see how they can relate with each other.
“Photorealistic” vs. Abstract
One of the first decisions I like to have a project team make is what is the quality of the image that you are trying to produce. A lot of times this is driven by if you need an environment early in the project or if you are looking for an end-product or what you are trying to get the client to understand about the project.
Let’s say you are at a conceptual phase. You most likely don’t have your materials picked out or even much of your design figured out, but you are playing with masses spatially. This situation would call for a more abstracted experience so the client is paying more attention to the spatial qualities of the project rather than the details. Something with a museum box style may be best, mainly shown in white and gray tones so you are easily able to understand the spatial aspects of your model as well as things like how light feels within the space.
During design development, you may want to consider getting a little less abstract but not yet realistic. In this phase, you would show things much like a shaded or consistent color view allowing for you to visualize material palettes conceptually and where they are placed but without getting into things too deep like considering patterns, reflections, etc.
When your design is “finished” or you are producing a final environment is when you would want to get to the more realistic views. This allows the client, users, donors, etc., to get behind the idea and more fully understand what the final product will be. This issue with getting to photorealistic is the sheer amount of detail that needs to be produced within the model. Things that aren’t always modeled in a Construction Documents set, such as flooring transitions and material properties, can make or break an experience. That being said, there are still multiple levels of realistic environments. Diving into a photorealistic environment is something that needs to be vetted thoroughly as it is a quite costly and time intensive activity, where you may consider having it done externally.
Mobile vs. Tethered
Deciding on how you will be showing your content can make the decision for quality come more easily, but it is only a matter of time until that will no longer be the case. With the advancement of technology such as processors for smartphones, things like Oculus GO, and better wireless display technology, soon we will be at a point where we can run fully rendered photorealistic models in a mobile system. But until that day comes we remain tethered to a PC for higher end experiences. In the meantime, companies such as HP have created more of a mobile PC environment with their Z VR Backpack PC allowing for the quality of a tethered system while having a much larger area to use for movement.
Deliverable vs. Workflow (Offline vs. Real-Time)
Tying in once again with what the rendering output will be we need to look at deliverable or offline vs. workflow or real-time solutions. There isn’t necessarily a perfect or even right choice here; rather, it depends more on what you are trying to show and why you are wanting to use it.
Let’s look at the deliverable solution first. These solutions can tend to be more realistically rendered but that doesn’t need to be the case all the time. You can utilize lighter modeling techniques and abstract rendering techniques to provide a lightweight, low cost takeaway. This can become a useful tool for recruiting, interviews, conferences, or just showcasing some of your work, but does not work when you are looking to use VR as a design tool.
You can take this a step further and tie in either panoramic or full 360 renderings, photos, animations, and videos that are viewable via the web or things like Google Cardboard as an extra layer of information to provide. Where these aren’t normally seen as VR, the same principles apply to static or pre-animated environments as navigable ones. Tools such as Revit’s 360 cloud rendering, Lumion3D, and others can be used to produce these more “static” environments.
Now let’s look at the workflow-based solution. There are currently plenty of tools out there that can be lumped into workflow-based solutions and a few that work in both deliverable and workflow processes. These solutions require more time being spent in understanding how your teams work. We’ll dive into that a little later (see the section on workflow). What we have found is programs such as Fuzor, Enscape, and to a degree Revit Live work internally for us. There are plenty of differences between all three of these software solutions, which are explained in the next few sections.
Getting up and Running
Now that you have decided on a style or styles that you would want to pursue, you’ll need to make sure you have the infrastructure in place to create and view your projects. So, what is it that you need? The first and possibly the biggest expense is getting a computer that can handle modeling, rendering, and displaying the project. Second, getting your team a headset or headsets to view the work in. Finally, purchasing software to create the virtual environment.
While the computer requirements for VR can be broad, we can create a baseline that can handle modest scenes as you develop your VR offering. There are a few options to consider when deciding what to do when purchasing a computer. You can consider VR-ready computers, available in both desktop and laptop, upgrade a current computer, or build one from scratch.
Let’s first look at the VR-ready devices available for purchase. These devices are generally geared to consumers with stronger specs when looking at the graphics cards but the processer and memory are not always the best available. That being said, you can upgrade the machines after purchase. These machines are great for viewing models but aren’t recommended for the modeling or the creating of the environments.
There are a few devices that are available that fall into the VR-ready category but are geared more to the prosumer or professional market. HP’s Z VR Backpack PC is one such device, with higher specs than the other pre-assembled computers available. Another positive of this device is the mobility that it affords: with dual hot-swap batteries your range is limitless on the size of space that you let your users go wild in.
Depending on what you currently use in your firm, it’s possible to upgrade those machines, possible relatively cheaply to optimize for VR. The reason for this is that many computers that are built for the AEC industry tend to be spec’d out pretty well, with better processors and more RAM than traditional machines. You may only need to upgrade to a better graphics card and possibly a larger power supply. Currently most software recommends no less than the Nvidia GTX1060, but I would recommend GTX1080 or better for the majority of uses and possibly going to the professional series depending on your needs.
Building a machine from scratch is always an option but can take more time and require you to have some knowledge in building a computer from scratch. Much like if you were getting a machine for your modelers you would want to make sure you have a high-end process in terms of speed. 32 GB or more is recommended for RAM, minimum of a GTX1060 but a GTX1080 or better is recommended for a machine you will use when creating the environment, a solid state hard drive, and a power supply that can handle the load of all your components.
Workflow
As we dive into using VR it is important to make sure that we are constantly examining our workflows to determine the best solution. It’s best to look at a few different projects of varying scales to document the workflows that are currently being used as they may vary slightly but will give you an understanding of how your users work. Once evaluated, you may notice that the workflow you are currently using ties into a VR solution easily, or you may need to either modify what direction you will be taking or modify your current workflow.
Here’s what we like to think our design process looks like:
A project initiates, multiple concepts are looked at through conceptual models, those are then narrowed down and a few have visualizations created for them. One is selected to move forward and design continues through your documentation process, with multiple visualizations being done throughout. Eventually the project is signed off on and built.
Now we know that this isn’t how it works. In truth, things get fragmented easily and we end up with multiple tangents that need to be managed and resolved down the line. This is pretty normal for the “traditional” design process, but it leaves us with a pretty inefficient workflow and having to reproduce work. Let’s look at how these fragments and tangents are created first and then we will consider a workflow that minimizes if not removes the extra work that we have been doing.
The “Traditional” Design and Documentation Process
What we’re defining as the “traditional” design process is really a process of splitting the model from one program into multiple. This splitting from a Single Modeling Environment to a Multiple Modeling Environment causes the fragments and tangents mentioned before, which cause more time managing the models as well as a considerable amount of rework.
So now that we know what the underlying cause of this fragmentation is, let’s look at a real-world scenario about what goes on when we go to a Multiple Modeling Environment. The time where these tend to appear the most is during the middle of the design process where the design is still ongoing but documentation has already begun. During this time people are more likely to break the model into an offset, allowing them to develop different options. We see this happening when people are looking at new floorplan(s), different material palettes, and furniture layouts. If not managed correctly it can cause havoc on the project timeline and quality of the final product.
The “New” Design and Documentation Process
The “new” design and documentation process isn’t something that is actually new; it has been around for quite some time. What is different than the traditional process as described above, in that there is an optimization component to the process, allowing for design iterations to take part of the documentation process without causing potential breaks.
Since this is more of an idea, there is no right or wrong way when talking about what software or technology is used. That being said, some software is more tailored to this process. Software such as Kallochtech’s Fuzor and Enscape are two of the leading-edge software solutions that allow you to visualize as you design. Each of these allow you to have a separate view (a new program window with Fuzor and a new window within Revit with Enscape) that enables things that aren’t possible in a standard Revit 3D view. While you are working within your Revit model these views are updated live.
While not having Live updates, Autodesk has also come out with Revit Live which is a one-click solution to VR. What makes Revit Live unique is that it uses the cloud to create the VR environment and notifies you once complete so you can download the model. From there you can modify the file in 3ds Max Interactive (previously Stingray) which is a game engine built on the 3ds Max environment. This allows you to modify things that you couldn’t in other software such as light switches, fans moving, etc. While you are working within 3ds Max Interactive, you can be continually advancing your Revit model. At any stage, you can “re-export” the Revit model and update your 3ds Max Interactive file allowing you to keep all the extra work while continuing to develop your project.
Chris Kelusak is a highly motivated individual that has a real passion when it comes to technology in architecture. Currently working as a job captain and BIM manager at Taylor Design, a California-based design firm that utilizes a human-centered design-based approach to solve challenges in the healthcare market, gives him not only the opportunity to develop incredible models but allows him to be continuously looking at new ways to improve the process, increase performance, or sometimes do things never seen before.