The Now and Future of Virtual Reality for Design Visualization

JEFF MOTTE: So good morning, everybody. Thank you all for coming to the very first 8:00 AM session at AU. I know this is a tough one to make it to. Obviously, we've got a really cool panel today, kind of all focusing around VR and the AEC space, the architecture space. Curious how many people here are either using VR, MR, or MR in production right now? Wow, OK.

And of those who haven't put your hands, up how many plan do that within the next year? So it's like the remainder of you, good. So first of all, let me just introduce our panelists here and then I'm going to have each of them do a little intro. So we've got Joel Pennington here from Autodesk.

Scott Dewoody-- sorry, it's early in the morning here. From Gensler. Anthony Cortez from ARUP. And Gaspard from piranha in New York City.

GASPARD GIROUD: Hey, everybody.

JEFF MOTTE: Cool, thanks. The reason I'd asked you all that question earlier because we did a survey on CG architect I guess July of this year and we asked the same question of whether or not people were using VR at all. You can sort of see on the right hand side here, it's kind of the make up of who answered the survey. But when we dig into the details here, we asked which HMDs everybody was using.

So you can see, obviously, the Gear VR, the easiest kind of entry cardboard, the vibe, obviously really high. And we asked whether or not they're using it in production and then, of course, whether or not they plan to experiment with these same HMDs in 2016 or into 2017. And then I guess this is the part that I found really interesting, is whether or not people were actually using it in production.

And right now about 70% said yes and then 77% said that they would plan to do it next year. So I think it really kind of shows how timely this VR is and how applicable VR is to architecture and how excited everybody is about this space. So I just want to go down our panel here, maybe just quickly because I want to get into the questions, introduce yourself and just kind of what your role is with VR and maybe what you're working on right now.

So Joel?

JOEL PENNINGTON: Sure. So I joined Autodesk about 6 and 1/2 years ago from Disney. So I come from the film and game world. I spent many years doing virtual production for [INAUDIBLE]. And before that I worked at Electronic Arts building motion capture work flows for all the sports games that folks here hopefully play and love. And that kind of was a springboard to architectural visualization in a roundabout way. But the challenge that I was given was, can we take what we're doing in the entertainment space and make it available to folks in AEC.

And can we do that, not so they have to know how to use the game engine, not so they have to know how to go and do special technical direction work, but rather can they work from Revit, stay in a workflow that makes sense to them and get into an interactive experience very quickly. So tough, but we're cracking that nut.

JEFF MOTTE: Scott?

SCOTT DEWOODY: So my name is Scott Dewoody. I've been with Gensler for about nine years. My background is actually not in architecture. I studied 3D illustration. So I started at Gensler as a visualization artist. And now I'm what we call the firmline creative media manager. So I'm overseeing all of our rendering technologies, R&D into VR, AR, MR. Overseeing workflow processes and development of software that way, and tools internally. So that's my role at Gensler.

ANTHONY CORTEZ: Good morning, my name is Anthony Cortez. I've been with ARUP for 13 years now. And I work in the visualization group. And we support all of the different [INAUDIBLE]. Structural [INAUDIBLE] and acoustic [INAUDIBLE]. So my job is to basically work with them. Gather all of their calculation models and create an integrated model, which then traditionally [INAUDIBLE] the renderings or animations. But lately, we've started using virtual reality, augmented reality, to be able to help our clients visualize what their projects would look like, would perform like years before they get built.

So I've been doing that for a while now. It's a lot of fun, it's like playing with LEGOs all day. You know, I hope the share-- I'll be sharing with you some of the projects that I've worked on. I've been doing the VR stuff for quite a few years now. So I'm looking forward to sharing some of the insight with you on some of the projects and workflows. Thanks.

GASPARD GIROUD: My name is Gaspard Giroud. I am an architect by training. I founded a company with my business partner nine years ago. He's an animator. So we always did work between the world of architecture visualization and commercial work. And today, seven years later, we have a company that has four divisions. One is advertising or communication, a creative studio. One is realistic branding, marketing and visualization. We just opened it. It's called Public Square.

We have an aerial filming division. We do a lot of visualization for architecture, so we did a lot of aerial filming and now we're doing it for TV and film. And finally, we just started a division for virtual reality and we're working on a VR platform. And that's how I joined.

JEFF MOTTE: Excellent.

JEFF MOTTE: So I think a lot of people that have just started to get into AR and VR, they're trying to figure out how they use this in this space. So I'm wondering, you know, one of the reasons we use it is to improve communication within the architectural space. I mean, do you think that's an important part of this? And how do we improve that process?

JEFF MOTTE: So guys are sharing the mic here, I guess.

SCOTT DEWOODY: I'll go and kick it off. It's really important in the communication process because in my experience, a lot of clients sometimes in architecture have a hard time visualizing even with a rendering but even when you look at a 2D drawing, a plan, you don't get the sense of depth and space that you do with VR. And so everyone knows when they put a headset on what they're looking at because it is almost like looking at reality in the same way, especially if the renderings come close to photorealism.

And clients just, they seem to automatically get it because you know what you're looking at. And what I mean by that is that we all know-- ceiling heights, we all know our experiences in the real world. And so moving into an architectural space that hasn't been built, you try and make the same parallels and they come very easily. It's much like watching a CG movie with a human character.

And if it's not done right, you can tell something isn't right with that character because we look at humans all day. And so it's that-- I don't want to call it the uncanny valley but that is kind of what it is.

GASPARD GIROUD: So at Piranha we don't use VR for design, as a design tool. We use it for a marketing tool. So I don't actually have so much experience in using it as a design tool.

JOEL PENNINGTON: One of the things that we've learned is that we made some assumptions at Autodesk around visualization. And we thought, OK, virtual reality could be used to help convince someone during a pitch or a pursuit process, yeah, we're the right architectural firm for the job.

But we actually have learned over the last year that the designers themselves to kind of riff off of what Scott was saying, is that they're actually going into virtual reality to do like a little mini design review. They want to see what the building is like before anyone else has done it. And in fact, we've done some tests with architects who are 20 year professionals. And they notice things once they get into virtual reality that they didn't see in the design package.

And that's quite interesting to me because this does seem to move from, this is just for a visualization to, oh, maybe this has some application in the design process.

ANTHONY CORTEZ: Yeah, it's pretty cool how virtual reality comes back every 17 years or so, right? Yeah, I remember, you know, from like Lawnmower Man in '94 to like 2000 when GPUs started being used. And then now with the advent of our new HMDs, the Oculus Rift, you know it finally came back and I think it really caught on now. And it's pretty exciting, you know, to be able to experience or witness that evolution of technology from back then when we were just using AutoCAD and 3DS to do our visualizations.

And then later on we've got Revit, we're doing BIM. So just like seeing the evolution. And I think the next step is VR right now. And I wanted to share it with you-- oh sorry-- one of the earliest VR projects that I've worked on. We did it for our client, GSA. It's for the courthouse in Mississippi, Jackson, Mississippi. This was done back in 2003.

And traditionally, GSA would build prototype physical mockups of these courthouse spaces out of plywood. And so we offered, you know, we told them that because we are working with you in multiple disciplines, like lighting and acoustics, we're able to take this and create a virtual 3D model that includes all of that lighting calculation information, the daylight lighting coming in, so you're able to walk around and see different sightlines of sitting in a judge's seat or sitting in a back row, if there's any glare coming into the room.

And it was very valuable for them to be able to experience that in 3D and virtual reality. So yeah, even how the real would sound. And it's the design iteration process of it that really helped them come up to the final. And this was actually presented at the Disney labs, the Disney cave in Los Angeles back in 2004 where you have a panel of judges. They're all sitting down in the same room looking at this virtual model. And they are able to experience what the design would look like.

And it pushed the project forward. Got approval. Got clearance. We got paid. And now there's the Jackson courthouse in Mississippi. So yeah, definitely communication is one of the key processes in making these projects go forward. And we've also been using augmented reality, as well, to visualize projects.

Over here we've loaded our models into iPads and tagged them with GPS coordinates so that we're able to load up a model of-- like an infrastructure model-- of water collector, a rainwater collector. And be able to view that in context on the sidewalk while our team was walking around New York City in the Bronx. And these, then, they can orientate the models. They can move it around. And all that information goes back to the office, which is, again, communicating the design in real time.

And there's other examples that I'll get into later. [INTERPOSING VOICES]

JEFF MOTTE: Yeah, so your last slide there, with the judges in the courtroom, kind of segues into the next question. I was at an event, an Oculus Connect event, I guess about a month ago. And Mark Zuckerberg kind of introduced this concept of social VR, which really plays into the concept of collaboration within VR. And I'm wondering if any of you guys have run into experiences where you've tried VR in a collaborative environment or even developing something that kind of plays into that and how important that is.

SCOTT DEWOODY: I would say collaboration is extremely important because every time I introduce a new partner Gensler to virtual reality, their first question to me is, so can we put multiple people in this at the same time? And I see collaboration sort of being the next new buzzword around VR. Where it's like now that we've got these singular experiences and we kind of know how to start developing for it, it's like now we need to put more than one person in it.

Because we'll do presentations to clients with four or five of them in the room. And so we may get four or five year Gear VRs and have them all sort of look the same project at the same time. But if you could have everyone in actually the same project at the same time, someone can curate it and walk everyone through and say look to your left, look to your right. We all kind of know where they are really looking.

I think that's going to be key because again it goes back to communication. So now we can communicate better because we're all in the same virtual space.

JEFF MOTTE: Is that something that you guys have solved yet? Or are you still working on trying to figure out how that works, technically.

SCOTT DEWOODY: Technically we know how it works. I just have to start putting it into the actual motion. A couple of years ago, we started developing our platform around unity and got a multiplayer server up and running internally. And were able to connect load in models in there and have people walk around, just like you would a normal video game. But taking that to the VR level shouldn't be too difficult in there now.

GASPARD GIROUD: I do think VR will get adopted. There will be a mass adoption at some point. But I think it's a very isolating experience. And it'll take a lot before you get to a place where it's become collaborative and people can share experiences in there. I think we have to realize that in order to do that you have to create avatars. And human-like avatars are extremely-- that's another step that's going to take many, many years before we can do that and yet it will probably be very uncomfortable.

So I think that's a very big obstacle that VR has to overcome and is probably going to take 7-10 years.

JEFF MOTTE: Is anybody going to that that event at Top Golf? It's like, I don't know, there's supposedly this place down the street from here, it's called Top Golf, where you're there. And you're surrounded by screens. And you've got a golf club. And you're swinging the golf club virtually. And the whole environment is around you. And it's all CG.

But it's, you know, more and more I think there's going to be more opportunities to be able to create these sort of experiences so that you can have multiple people, multi players, working and playing together. A huge, huge phenomenon that happened this past summer was the whole Pokemon Go. It's AR. It's not VR. But the just the phenomenon of it was that it was released. 100 million people downloaded it.

After a month or so, it generated $250 million after one month. And you had 100 million people playing it every day for, I don't know-- together, collaborative. Together. So I think it's just-- the technology is there and we're just trying to find the right applications for it.

JOEL PENNINGTON: So to add on to the concept of collaboration, I don't think it can only happen in VR. I think it has to be device agnostic. So if you've got a principal at a firm who is just not going to put goggles on his or her face, that person needs to be able to experience that collaborative environment from perhaps a desktop or a tablet.

Or if you have folks that are on the field, maybe they're in some form of augmented reality because they're looking at a situation where HVAC is intersecting a structural member. Who knows? But I'm thinking it can't just be a bunch of people in VR, a bunch of people in AR, a bunch of people in MR. But it's got to be across those boundaries.

JEFF MOTTE: Have you guys ever read that book, Ready Player One? Anybody? Great book.

JOEL PENNINGTON: That should be required reading for everyone in this room.

JEFF MOTTE: Yeah. I mean, that is like the ultimate-- the killer app for virtual reality, social virtual reality, right? You know, being able to load up different worlds and interact with each other. I think Second Life had a little bit of that going in two dimensions. But again, it's something that could evolve into the Ready Player One in real life.

Right now, how are you guys using AR, VR, MR in your own firms and agencies and studios. I mean, maybe give some specific examples of applications that you guys are working on. And then even to that end, something that you've seen out in industry that you think is kind of, that's it, you know, that's where it should be going. This

SCOTT DEWOODY: That's a long question. So at Gensler we've started developing our own viewer app that designers and clients will be able to use. You can check a version out. It's actually on the app store now. It's Gensler VR. So we've been doing that. We do a lot of just still render 360s, which is what that app supports. But we find that just the still renderings is sort of like really good bread and butter to get the idea across.

But even then, now we're diving into more like the room space VR with the HTC Vibe and the Oculus Rift. On the augmented reality, we're working closely with Trimble on the HoloLens. And so we're trying to figure out, kind of again, going back to that can someone be out in the field wearing a HoloLens device, make some markups, and in real time feed that data back to someone at an office sitting in front of the computer who can make those changes to a model in real time.

That is a very interesting use case to us. And we're exploring all different other kinds. Like can I be in a room with the HoloLens looking at a model on the table and have it render in real time via a server or something along those lines. It's really interesting where that technology is going. And we try not to have the two compete.

Like they definitely need to work together. But the use cases for AR and VR are very specific where I always tell people, it's like VR is where you want to completely immerse someone somewhere else, that's the VR. When you want to bring your 3D model into the real world and see it in place, AR is perfect for that.

And we've been doing that. Our LA office is expanding their campus to another tower and so we're building a sky bridge. And so one of our first test goes with the HoloLens was we took a Revit model of the bridge and we loaded into the HoloLens and was able to put it in place and see where the bridge was connecting. And designers were able to walk up and go, really? This is six feet? We need to make that higher. And automatically started making designs decisions and changes looking at the actual model in place.

ANTHONY CORTEZ: OK, so just getting here, I just want to give another example of how it could be useful and is being used in real life. Just getting here from the airport, I hop into a cab and the cab gives me directions, turn by turn directions, via GPS. It brings me out to the Venetian, the front of the hotel. But then when I get outside the car I go inside the hotel lobby. I want to go to the registration area, right?

And I must have gotten lost, you know, like walking through the casinos. Made a left instead of a right. So I think, you know, like way finding is going to be a huge thing for augmented reality, virtual reality, mixed reality. You know, we've developed a couple of applications using the Google Tango where the Google-- has anybody used Google Tango here? Yeah. So it works like-- it does area learning. And what it does is once it recognize and references where you are in the environment, then it loads up virtual information on top of that.

So we've used a combination that and some scripts to be able to do turn by turn navigation indoors using SLAM, simultaneous localization and mapping, that's built inside the Google Tango. And now, I think, last year Google Tango and Lenovo had partnered up together. So our next generation phones are going is going to have this kind of technology in it. And you know, those are just going to be more uses for this, I think.

JEFF MOTTE: Gaspard?

GASPARD GIROUD: So we are using VR mostly for sales or for some of our clients when the product is already finished. So it's rarely useful as a design tool, although I see enormous advantages of using it as a design tool. Some of our clients from the advertising or real estate world ask us to model and represent their space for selling or marketing purposes.

The other thing that we're developing, so it's a more traditional [INAUDIBLE]. It's not so pioneering. But our main focus is on the platform that we call a VR browser, geography VR browser, which we look at it more as a platform as opposed to a VR expense itself because we're going to invite other vendors to put their expense in there.

And we started by creating a scale model of New York City where you can walk around and then get access to all these experiences. So this is mostly what we're focusing on. And so I think this goes way beyond architecture visualization and this would require funding. And it's a very ambitious project but that's something that is inviting people from all different type of industries to join in the VR world, right?

JEFF MOTTE: Joel, you're at Autodesk but maybe talk about what some of the customers you're starting to see, how are they using this technology now?

JOEL PENNINGTON: Well, so we've got-- we have customers from all walks of life. And they are building their own. If they're a large enough firm, they're able to build their own. They can make their own experiences which are compelling. Lets them win projects, lets them communicate design intent and so forth. There are other firms that aren't big enough to be able to do that, so they're leveraging, perhaps, game technologies and hiring kids out of school, out of film schools or game development schools.

And then there is what we're working on right now, which is Autodesk Live. It's not a sales pitch but really information for y'all to understand that this is using Stingray, which is game technology, and allows us to go from Revit today into an immersive environment without having to hire the folks from the game school, without having expertise or domain knowledge at your firm.

So we're seeing that as the first availability. And then me, specifically, I get to look a couple of years ahead of that. Where are we going to go with products like Autodesk Live? Where is our thought leadership? And collaboration, augmented reality, machine learning and expanding beyond the confines of just architecture. Getting deep into construction. Getting into combining the content from the manufacturing world into architecture are all areas that we're starting to seriously look at.

ANTHONY CORTEZ: Can I give another example of how we're using it in the field?

JEFF MOTTE: Sure.

ANTHONY CORTEZ: So this here is an image of-- we worked with Virginia Tech to create a virtual reality cube on their campus. It's a 50 by 40 foot black box. And it's a research center that blends science and engineering with art and virtual reality. So the idea is that using virtual reality, they're able to go inside this space and collaborate with different models. They can load up a DNA strand and walk around the DNA strand in this room.

They could go inside and load up a tornado and go inside and experience what it would be like to be in a tornado without getting killed. So it works off of the same technology as motion capture rigs. There's a bunch of sensors all around the different catwalks. And then the Oculus Rift has an antenna on it that records your position and orientation within that space.

And the cool thing about this rig is that 24 people can be in this virtual world at the same time, walking around that DNA strand. And then another aspect of this, also, is integration of sound. Around the theater, the space, there is about 64 loudspeakers that are set up. And we're using this technology called wave field synthesis to pinpoint and locate sound anywhere in this room, whether it's above you, next to you, over to the corner.

And it's another way to immerse your audience. You know, you have your visual. You have your audio. And those are the two senses that makes you feel like you're there. So that's another example.

JEFF MOTTE: So we've got virtual reality, augmented reality, mixed reality and everybody's kind of experimenting a little bit in each one of these areas. Do you think, ultimately, there's one technology that's the direction we're going? Or is it always going to be split up between these three areas based on use case? What are your thoughts on that?

SCOTT DEWOODY: I think there are going to be-- I don't think it's going to ever combine because I think each has a really good, nice, specific use case. But I see applications taking advantage across all three, depending on what we were talking about earlier, where someone in the field might be using AR. And then someone back in the office might be using VR. And then someone might be sitting on a 2D screen modeling off to the side. I mean, the possibilities are endless. But I don't think that those three are ever going to converge into format or experience.

ANTHONY CORTEZ: I think some of the tools we're using will help that process along, particularly with the capturing tools like using Photogrammetry, using LIDAR scanners. All of that goes into the pipeline of-- that helps set up your virtual and augmented realities. And we're all here because of the AC and probably a lot of you are into the building information modeling. But I think the ultimate goal, at least for our industry, is to be able to take those models, those big models, and overlay it on top of the real world and have your ultimate killer app, which would be like x-ray vision inside this room.

And be able to, I guess, like the ultimate stud finder, to be able to see behind these walls. And using your BIM model that you've already captured, you've already scanned, overlaying it on top of this and selecting different objects in this room and getting specification data, cut sheets and stuff like that. All of that, whatever can help you enhance your visual perception makes you smarter, makes you have-- make quicker decisions as well. And that's one of the huge values that we're going to need to see to keep pushing this forward.

JOEL PENNINGTON: Yeah, I think it boils down to being device agnostic. So use the right device-- the right tool for the right job. And if a virtual reality is the right thing at the right time, go for it. If it happens to be your phone and you want to use your phone as an augmented reality device to look at a piece of conduit in the wall for whatever reason, and that's what you have, great. Use that.

But I don't think we should feel we have to use something specific if it-- today, for me for example, I kind of don't even have a line between my phone email and my laptop email. And it's all kind of blurred. So I expect we'll have a very blurry future with MR, AR and VR.

JEFF MOTTE: So we're starting to see hardware now that everybody can actually use and do productive things with it. But I think one of the biggest challenges is still workflow. You know, all of these development platforms all came from the gaming industry and of course we're all in the AEC industry. And even though many of us are very knowledgeable about how to operate in 3D space and how to develop tools in 3D space, that process is still very cumbersome. And I'm just wondering what your guys' thoughts are on that. How do we fix that problem and how do we make these tools work in a way that's more conducive to working in design and architecture and architectural marketing.

GASPARD GIROUD: I think that Stingray is doing exactly the right thing by trying to get from your 3D model to VR. I think it's a very exciting direction. I think they also were very successful at doing that in the early days of 3D, you know, when silicon machines were not accessible. And then you were able to have a 3D-- even before 3D Max on a PC. And then they [INAUDIBLE] democratize that use and became a mammoth in the industry. So I think they're doing the right thing.

No, I think what is still a very big obstacle, is that building 3D is not something that is so easy to do. That will be the next step, would be making that model making more democratic. But I think you'd be a better person to answer that question.

JOEL PENNINGTON: So anecdotally, when we were making films like Avatar or A Christmas Carol or other classics like Mars Needs Moms, which no one here should ever watch, we would spend over $100,000 a day to let the director be immersed in the virtual film production world. And that's not a tenable way of working, to say the least.

So we're already leaps and bounds ahead, in my opinion, ahead of where high end film and game development is. I think we're starting to see a shift where folks in the AC world are leveraging technology that is so powerful for what it costs. And to riff off what Gaspard is saying, we're seeing things come out of the woodwork now, out of hackathons, from like tangential industries, showing the idea of not just being in your immersive environment to look at something but being in your immersive environment to start to create the thing. And that's really interesting.

So what does that mean for massing? What does that mean for conceptual design? Do we do that while we're in an immersive 3D environment? If we're in HoloLens, in a HoloLens environment, and we're actually out on site, can we start to push and pull and start to shape a proposed design and see it in context immediately? See it lit as though it was real, as though the sun was directly lighting the actual CG elements that you see through the lenses. These are all starting to bounce around our heads a little bit now.

And if you look at the new version of Microsoft Paint, it is now Microsoft Paint for 3D. So it's a wake up call for us at Autodesk at the very least.

GASPARD GIROUD: I am actually not aware of a Microsoft Painting. Explain a little bit more about it.

JOEL PENNINGTON: So yeah, Microsoft Paint's like the very basic, like very, very, very, very basic version of Photoshop. Lets you draw stuff.

GASPARD GIROUD: I know, I'm talking about the 3D version that you just mentioned.

JEFF MOTTE: It's coming out in their next service pack next year. They're calling it the creative addition, or something along those lines. They're upgrading it to support more direct VR and HoloLens capability in Windows 10. But they're throwing in 3D into Microsoft Paint now. You can import 3D models into paint and I don't know. [INTERPOSING VOICES]. It's like, really?

SCOTT DEWOODY: Is it kind of like Google Tilt Brush?

JOEL PENNINGTON: It's similar to Tilt Brush in a way, and also Rif's experience. What's the--

JEFF MOTTE: Medium.

JOEL PENNINGTON: Medium, yeah. It's very similar to both those.

SCOTT DEWOODY: Yeah, that's really cool, especially if you could bring your 3D models into this environment. I think the key is, though, to be able to do your adjustments, your edits, your tweaks in your model in 3D, right? But have those, whatever you do, gets recorded and dynamically linked back to your original Revit files so that those are updated.

JOEL PENNINGTON: We've experimented bringing Revit models into Tilt Brush and have had an interesting experience with that. I mean, we were successfully able to do red marks and markups and save screenshots and things of that nature. Now, getting them out in 3D format is not possible yet. I mean, you can start to see, like as soon as you do that and bring it in, it's like your mind as well is like, yeah. I can actually be in VR and start making red marks.

And having the ability to maybe eventually push those back to Revit would be awesome, right? And so you can start to see the building blocks. And I think we're just going to-- all of us are going to start requesting and doing hackathons and making this stuff happen. And that's the first time I've seen architecture. We've always adopted other things from movies and film and rendering and things like that. But this is the first time where we're adopting gaming technology. But we're consuming it faster than we've consumed anything else before.

And at some points we're actually moving faster, even faster than that industry, where it's like, hey, why can't we do this? Why can't we do this? And they're like, we just don't do it that way. And I'm like, why not? Like, we need it that way. We could do this. And it throws that industry through like, oh, well, I guess we could. And so even now, if you look at it like Nvidia's developed the new Vulcan API for gaming, moving away from like direct [INAUDIBLE].

You can now have billions of objects on the screen with virtually no hit to performance. And that's going to help us because now when I load in a massive building with thousands of chairs and work spaces and lights and [INAUDIBLE], my designers aren't going to have to worry about occlusion calling and optimizing. It's like, yeah, now you can't just throw everything and the kitchen sink and it should work without too much effort there.

So we're starting to see some of those developments and I think over the next year, six months, I mean it's going to happen I think faster and faster, especially with everyone in this room in our industry just doing more demanding and going, hey, you're not moving fast enough. We're going to go ahead and do something ourselves, which is what sort of we've done at our company already.

It's like we saw some issues where people hadn't filled the gap. And so like, you know what, we just need to get in there and do it.

JEFF MOTTE: I'm curious, just because it was brought up, how many people here have been using Tilt Brush or even Medium if you have early access to it? And using that in a markup or a kind of early sketching phase? I'm curious, just with a show of hands.

JOEL PENNINGTON: I spent a good two hours the first time I was in Tilt Brush and I wasn't even marking up stuff. [INTERPOSING VOICES] It's like, really, painting in VR? And then as soon as you start doing it, like, two hours went by. And I was like, oh my Lord. Like I have to get back to work.

SCOTT DEWOODY: I think there's also a connection now to it where you can put music and connect it to music. Your paint strokes will have audio.

JOEL PENNINGTON: It will pulse to the beat of the audio.

JEFF MOTTE: There's another guy doing a talk this week, Paul Nicholls from Factory 15. He's doing a talk this week, so if you don't see it, I would stop in. But we were having dinner with him last night and he was talking about how for the last couple of months he's been spending eight hours a day in Tilt Brush actually designing worlds and stuff as part of projects that they're working on now. So they've been doing that for quite a bit.

JOEL PENNINGTON: Because you can actually export your paintings as FBXs now and import them into [INAUDIBLE]. [INTERPOSING VOICES]

JEFF MOTTE: Yeah, he was actually-- what I found really interesting is they were experimenting with how they combine real world and virtual world. So they actually built like a foam mockup of space. And then he would trace that space in Tilt Brush just so he'd have reference in the VR space. And then build around that so then when people were walking around in the VR space and they went to touch what they saw, they were actually feeling something in the real world. So it was actually kind of an interesting use case.

SCOTT DEWOODY: The void, has anybody tried to the void?

JEFF MOTTE: Has anyone been to the Void in Utah? Do you guys even know what this is? OK, so this is a company built out a warehouse. They built physical walls for like a virtual arcade. And so you put on their backpack and their HMD. And so as you're walking around, you could actually run into a wall. And if there's mist or smoke, they actually blow mist and smoke on you as you're walking through. So think of it as like a real life maze but with a VR overlay on top of it. And they just got like the Ghostbusters--

SCOTT DEWOODY: Yeah, the Ghostbusters one in Times Square in New York. It's pretty cool.

JEFF MOTTE: Have you seen it?

SCOTT DEWOODY: Yeah. It's pretty cool.

GASPARD GIROUD: Can we ask, how many people have the [INAUDIBLE], just out of curiosity. So there is a significant amount. Have the [INAUDIBLE]. How many people have tried the [INAUDIBLE]? Everybody, pretty much?

SCOTT DEWOODY: I also I have another question for the audience. A lot of you raised your hands when you're doing VR in your work flow. I think there's like a couple-- two types of VR that is out right now. I guess that the 360 degree panoramic stuff that you would see from like the Google Cardboard. And then there's the fully CG environments. So are you guys doing one or the other or both? [INTERPOSING VOICES]

JOEL PENNINGTON: Chaos group teamed up with a company that's developing that stuff, too. The [INAUDIBLE] renderings, they're 360 renderings but they're a little more dynamic. You get that parallax feel. So if there is a column in front of you, you can actually look around it instead of just being at a fixed point. So if you guys haven't seen any of that, I've looked that up. So I think it's like [INAUDIBLE] was the Chaos group one.

GASPARD GIROUD: That's what Dead Room is using, right? I don't know.

JOEL PENNINGTON: And then Nvidia with their [INAUDIBLE], they've got a version of it that's not commercially available yet but you can see stuff online with their campus. They did some light field renderings and you can kind of walk around a little bit and check it out. So that's going to be like the next, I think, small evolution. They're very computational heavy at the moment. But as things get faster, that should be more and more common practice.

SCOTT DEWOODY: So going back to the question, the original question about painful workflows, we do a lot of recording using the 360 degree camera balls. And those recordings, you have 10 go pro cameras and they're all recording 4K movies. And you're recording and saving 10 different shots from different locations and just managing all of that information together is the stitching part. That is the most painful thing.

And we're trying to figure out better ways to be able to organize and manage that information.

JEFF MOTTE: So this probably brings up a good question. I mean, there's so much technology out there. And I've spent most of my year traveling around to all the VR and AR events that I could this year just trying to keep up with it. And opinions that I had formed and things that I've seen, even three months ago, I think I would already changed my opinion of them because it's moving so fast.

And for companies that maybe don't have the resources that you guys have and they're looking at all of these options out there, how do you even know where to start? I mean, obviously not everybody can go out and buy every single HMD and every single authoring platform and you know spent a year developing and see what works. I mean, how do you decide what's going to work for you?

GASPARD GIROUD: I think that disruption is the new stability and is going to keep accelerating. So it's a very, very big issue. And I think it's going to bring enormous problem because even from investment purposes, the visibility that you have is getting shorter and shorter. So it is, in effect, a blindness almost because to a certain degree, there's no way-- at a different time, where railroad was installed through America. You're talking about years where you can put your investment, whether it's financially afford or anything else but you have visibility. That visibility is shortening by a tremendous amount.

And I think there's going to be a very big problem as we keep moving into this in the next few years even. So I think this is a very important question. And I totally agree that-- sorry-- will be [INAUDIBLE]. I'm really, like you said, Paint 3D that I was not even aware of, it's fairly typical that you come across technologies that are very important that you're not aware of. And I think that train is not going to slow down. It's only going to accelerate.

JOEL PENNINGTON: So we have, at Autodesk, tried to stay on top of all the new technologies and from the big players like Valve with HTC for the Vive, with Oculus, with which they have Vive port with Oculus and the Rift, as well as Sony with their virtual reality solution. And then Google, and so on and so forth.

But there's always someone that comes like right out of left field. And I suppose that my group, because we're attaching ourselves to a game engine, that the game team is strongly motivated to constantly update and support devices which are consumer friendly. And that becomes a cost effective way for folks like yourselves or our other engineers and construction professionals to get into the virtual augmented reality technologies because we've tested. But then we weave that and we test again and we develop.

And hopefully we've landed on some technologies which have a bit of staying power. And it is definitely true that the technologies shift. But perhaps we're seeing, like what we see with Google's Daydream and with Microsoft and their new virtual reality headsets that will be coming out, those are leveraging technologies that were developed with Tango and with HoloLens respectively, which was born out of Kinect.

And so they do have roots. They do they do have paths that they go along, if you're willing to stick it out and follow the direction that they're taking. So maybe on the hardware side, the cycle is a little bit slower. On the software side, the cycle is way, way faster. And so I am now trying to understand, trying to reconcile what does machine learning mean for all of this?

I mean, imagine if you could have an AI partner while you're in your immersive environment helping you design, giving recommendations. Or that is the designer and you're a consultant at that point, asking the AI to make choices on your behalf.

JEFF MOTTE: It's kind of scary.

JOEL PENNINGTON: Hopefully it's a good opportunity and it's not scary.

SCOTT DEWOODY: Skynet. That's all I think.

JEFF MOTTE: So somebody then is looking at an authoring platform or an HMD, is there one that they should-- should they just go for the big ones that have the big names, the Vives, the Oculuses because they have the most visibility now? Is that the safe path? Or is there another way to look at it?