Pivoting Game Dev Skills into XR Utility for Enterprise, with Extality’s Arash Keshmirian

XR for Business

English - April 21, 2020 10:00 - 31 minutes - 28.6 MB - ★★★★★ - 12 ratings
Technology Business Homepage Download Apple Podcasts Google Podcasts Overcast Castro Pocket Casts RSS feed

Previous Episode: Exploring XR Collaboration Statistics in Real-Time, with National Research Group’s Lauren Xandra

Next Episode: Shaping the Digital World with Our Hands, with Clay AIR’s Varag Gharibjanian

Today’s guest got his start in the world of game development. But soon, Arash Keshmirian saw the writing on the wall that XR’s current usefulness was better-suited to the worlds of industry, retail, and journalism. Arash and Alan discuss how he made that transition.

Alan: Hey, everyone, it's Alan Smithson here, the host of the XR for Business podcast. Today we have Arash Keshmirian, co-founder of Extality. His personal goal is to create powerful content that delivers results. We're going to dig into using Magic Leap and Hololens and mixed reality headsets as a tool for business. So all that and more, coming up next on the XR for Business podcast.

Arash, welcome to the show, my friend.

Arash: Thank you for having me.
Great to be here.

Alan: It's my absolute pleasure.
You guys have done some pretty cool stuff. I was on your website
playing with a shoe. What is Extality?

Arash: Were a lot of things to
many people. So we built Extality out of a long, 10 year experience
in the games industry, building mobile games. Did a lot of games,
including Zombie Gunship, which ended up being this kind of worldwide
sensation of shooting zombies from an airplane. That company --
Limbic -- we ended up doing a lot of XR stuff. And kind of around
2016, 17, 18 we built a game for ARKit, called Zombie Gunship
Revenant. And that ended up being a huge hit across the app store.
Apple featured it a whole bunch of times. It was one of the 2017
games of the year. And it spread ARKit to a lot of people, trying new
things on their new iPhones. And we later did a project called Zombie
Guns Raptor with Oculus and Oculus Go, Gear VR. But really kind of
around that time -- 2017 -- we started to feel like it was getting
way too crowded in the games business, and we were starting to look
around and try to figure out what we could do with our experience in
high-performance graphics and making cool experiences, immersive
experiences. I started talking to a guy named Ryan Peterson, who's
the founder of a CEO called Finger Food out in Vancouver. And he was
telling me about all these exciting opportunities in AR and VR for
enterprises. He was talking about how they'd saved millions and
millions of dollars for a truck company that was looking to move
their design to virtual reality from using clay models. And this got
our head scratching, we were like, "You know, maybe there's an
opportunity to use all of our games experience, to help big companies
and do more than just give people an entertaining hour on their
phones." So we founded a new company called Extality. And we set
out to essentially discover companies that really wanted to explore
XR, be it on their phones, on headsets, iPods -- every type of XR --
and leverage our background in doing just really hard graphics
problems, building scalable global servers and connectivity, all
those hard things that you learn how to do making games, we quickly
realized that we're super, super applicable to building enterprise
solutions as well.

Alan: Actually, I know Ryan very
well from Finger Food, really great guy. And they've done some
amazing work in the space. What are some of the highlights that
you've done for enterprise? And first of all, I just want to say that
having a flying zombie shooter game? Pretty awesome.

Arash: [chuckles] Thank you.

Alan: The fact you guys had a
hit with ARKit is pretty amazing, because there's not too much out
there leveraging the power of ARKit yet.

Arash, welcome to the show, my friend.

Arash: Thank you for having me.
Great to be here.

Alan: It's my absolute pleasure.
You guys have done some pretty cool stuff. I was on your website
playing with a shoe. What is Extality?

Arash: [chuckles] Thank you.

Alan: The fact you guys had a
hit with ARKit is pretty amazing, because there's not too much out
there leveraging the power of ARKit yet.

Arash: Yeah. I mean, if you want
to talk about games for just a second, it's an interesting thing. I
mean, it gives people a totally different experience using their
phone as the controller and running around the room. We have all
these videos during our user tests of people crawling on the ground,
shooting zombies and jumping up and down. And it became a very
athletic experience, playing games this way. But what we quickly
found -- and part of why we started to look at where can we use AR
outside of games -- was that I think session times really suffer
because if you want to play a game, you want to be lazy, right? You
want to sit on your couch. You want have your beer. You want to take
your phone out and just play in the least movement, sort of laziest
way possible. And asking people to get up and run around for 30
minutes -- not surprisingly -- led to really short session times. And
so we gained a lot of people wanting to download the game to try out
AR, but we lost a lot of retention, because people weren't interested
in spending the kind of time they would normally spend on a mobile
game, running around their house shooting zombies. And so that was
kind of the impetus for us to say, "Well, we've got a lot of
downloads on this, but it doesn't perform in the way games were
supposed to perform." But we still just love this technology and
we want to keep using it in a place where it really will make a
difference for people.

Alan: This is why PlayStation VR
has done better than the other full room scale experiences, because
most people when playing a game are just sitting there.

Arash: Exactly.

Alan: And they want to play the
game from the comfort of their desk, chair, or couch, or whatever.
And PlayStation VR lets you do that. Most of the games are on rails
and you very rarely have to give it in 360 degrees.

Arash: Yeah, exactly. Yeah. And
I think that just sort of underscores the importance of knowing how
things are going to be used and building experience for it. I think a
lot of people get carried away with, "We have this great
technology. We can have a 10 meter square play space. Let's use every
corner of it." And the reality is, that doesn't always make
sense for what your user is after, for that particular game.

Alan: Rarely makes sense to use
it 10 meters by 10 meter. I mean, when is the last time you had a 10
foot clear space anywhere?

Arash: Exactly. Yeah. It's
really just outdoor experiences that can really compete.

Alan: Enterprise. What have you
guys done in the enterprise side, what solutions have you created?

Arash: Yeah, like I said at the
beginning, we've been kind of all over many different industries. And
so we got-- our main first break as Extality was leveraging some of
our game connections and things like that, to get close to Magic
Leap. And Magic Leap, of course, did an amazing job of evangelizing
mixed reality across the world. And one of the companies that they'd
been talking to for a while was CNN, Turner Broadcasting Cable News
Network. And what they wanted to do was essentially explore with CNN.
What does it look like to do news in mixed reality and how do we
augment the experience of essentially what you might do in the
morning -- which is watching headline news -- how do we augment that?
How do we use these new display technologies? How do we make that a
richer, more interesting experience? And so what we ended up looking
at with them was at first, really, how do we show news and what's the
benefit of having, for example, Magic Leap device on your head when
you're having breakfast and maybe getting ready for your day? And we
started to play around with the first thing that comes to mind, which
is, well, we can have virtual televisions on multiple different
channels, in different sizes all around the room. And so we started
to explore with building user interfaces to create just like a really
great channel selector. And you can have maybe a little TV with the
stock market over at the end of the breakfast table, then a really
big TV that's got a particular show or maybe a Anthony Bourdain
telling you something about food. Have that at a larger scale and
then imagine even potentially multiple users seeing the same TVs or
some shared TVs and some individual TVs. How do we make that a cool
experience, so that you can have a TV wherever you want it floating
in space? And that was definitely a new way to to consume media. And
that's where the conversation started, I think. But as we got deeper
into it, you realized that, of course, there's a lot of stories that
really benefit from being told with 3D content. And this isn't
something that's new for news, right? I mean, we've seen a lot of 3D
rendered imagery. I think it started with weather. But a lot of the
time, a lot of even local news stations do this now where there's a
rendered fly through of the crime scene. Looking around the world for
something or showing different objects in 3D, so they can pull apart
parts of it and show the pieces of the latest SpaceX rocket, or
something like that. And we started there, because a lot of companies
-- including CNN and Turner -- they have teams that are dedicated
toward building the sort of content for broadcast news, where they
prepare a rendering and send it along with the video feed. But we
thought, "Well, what if we can take that content and somehow
make it more interactive, and make it so that you can place it on
your breakfast table, or on your coffee table?" And so we looked
around for a good story that might have been a good sort of starting
point for maybe exploring these types of interactions in this kind of
presentation. One story that was really coming out in the news at the
time we were working on this project was the Thai cave rescue, where
these soccer players got stuck in a cave in Thailand and they were
stuck in there for quite some time. It was a situation where the
whole world was watching and they had to find a way to rescue these
kids. And there are a lot of technical challenges. There were a lot
of weather challenges. It was just a really interesting story. And I
think where we were really excited to show that with mixed reality
was actually showing the cave itself and giving people a sense of how
deep were these kids, and how far below the ground, and how quickly
did the water rise. And so we started to work with CNN's team that
builds this kind of contact, to create 3D models of the actual
geography and the terrain around where this cave was. And we even got
down to doing things like where did the water rise to, and where were
they at the time? And building these timeline experiences, where you
can drag a timeline back and forth and see what kinds of things
happened during the course of the rescue. But what was really
exciting for me was actually being able to start to make it in your
own home. And what I mean by that is we took a cross-section of the
cave. So there's some parts where it gets as narrow as 16, 15 inches.
That's really, really small. But you hear that number and you don't
know what would it be like to crawl through a space that large. And
so one of the interactive experiences we built for that story was an
actual cross section that floats in your room and you can actually
put your hands through it and you get a feel for like, "Wow,
that's going to be really tight to squeeze through. I can't imagine
crawling through two or three meters of that." These types of
things bring the news in in a different way. But what got interesting
with this sort of presentation of factual journalistic material in 3D
was -- this actually came up from the CNN creative director -- was
how do we know what we can show and what we can't? What I mean by
that is if you were to show in 3D in a fully explorable by users kind
of way a crime scene, there's a distinction between what we know--
what we know for sure. And what we may have added to kind of fill out
the scene. So, for example, let's take the raid on Osama bin Laden,
for example. If we were to create that on your coffee table, we'd
have to show the whole building. But it's possible that we don't know
what's in the other rooms. It's possible that we don't know what the
content is of one of the back rooms. And we have to make this
judgment call, because as game designers, we would have just filled
that up with all kinds of set dressing and filled that with stuff to
make it look good. But for the news, it's-- they're held to a higher
standard. They have to know that everything they put into that
building is actually based on facts, or based off a photograph, or
based off satellite data. So this is a huge concern for CNN. And
they're like, "We can't say that this is there, if we don't know
that it is." And so there was a lot of conversation around in
the future as this develops as a key part of the way media is
presented. Do we show it in some way? Do we draw cross hatches
through those areas? Do we say, "Hey, this area was added for
visual effect, but we don't know that it conforms to the news."
And so it creates a lot of actual questions around this new type of
media. How do we tell the news and how can we verify that everything
in it is true, because the user can literally poke their head in
anywhere and we don't want someone saying, "Well, I poked my
head in the back room of the Abbottabad compound. And I found that
there was -- whatever -- a table and chairs in the back. And so there
couldn't have been more than six people, because there were only four
chairs." And drawing conclusions that are completely unjustified
by reality.

Alan: What an ethics dilemma,
right?

Arash: XR touches everything.

Alan: It's funny how the average
user of this technology would never even consider these types of
things, and the fact that the CNN team considers these things, it
really does kind of speak to the real power that they hold as
journalists in holding themselves to a higher standard.

Arash: Yeah, exactly right.

Alan: I wish the whole XR
industry was held to a higher standard.

Arash: Yeah, absolutely. I think
they set a good standard for what is information, and how is it
presented, and what what sorts of assumptions do we project with
that, what's implied in the things that we show. So yeah, it was
cool. And that was really exciting for us as one of our early big
projects, because it did the exact thing we wanted as founders of
this company, which was to get a glimpse into how do big companies
run. What are the exciting challenges that they have under the
surface? What are the conversations that they're having? We wanted to
learn -- kind of like Alan Peterson had -- how are trucks designed?
How is news made? What are those conversations inside the conference
room and what does all that look like? So for us, just kind of our
personal curiosity as founders, this was super cool, to go into CNN
and talk to all the people that are involved in making the news, and
then present to them this incredible new way of showing their work.

Alan: I guess one of the things
the major a-ha moments for you and your team started to deal with
these big companies, going from a gaming background, sitting in these
board rooms with these big companies. What was something that was
kind of totally foreign to you that revealed itself?

Arash: Well, first of all -- in
terms of the sales process and decisions -- it's a real challenge,
because you've got so many different stakeholders at so many
different levels. And not everybody is gonna be in the conference
room with you. Not everybody's gonna be in the same conversation to
give you the chance to convince them that that XR is the way they
should go. Sometimes that happens outside and and people-- someone
you've never met and never knew will torpedo the deal because they
think, "Oh, my son had an Oculus Go and had a terrible
experience. And this entire technology has no future." One of
the things for us was just learning who the stakeholders are, and
dealing with just large scale of an organization that's trying to
avoid risk. And for us, learning how to -- in the same way that that
Magic Leap does so well -- learning how to really evangelize this
within an organization and deal with a lot of the objections was a
big challenge. And the other piece of it is that a lot of companies
at this point now have tried some form of XR. And many of the folks
that we talked to are like, "Well, we tried that two or three
years ago and it didn't work at all" or "We tried that with
this other vendor and it was too expensive." They have some
preconceptions about what this technology is capable of or how much
it costs or how much work it is.

Alan: And to be honest and frank
with everybody, it was expensive and it did not work all that well
three years ago.

Arash: Absolutely. And now in
this industry, we have to come back and say, well, it's different
now. Will you please try us again? [laughs] And then every year it
gets better.

Alan: We all have better
quality, lower prices right across the board. Just take 360 video,
for example. Four years ago when we started doing this -- or five
years ago, whatever it was -- was $10,000 a minute. Because you had
to hand-stitch-- you had to, first of all, 3D print the camera.

Arash: Yeah, exactly.

Alan: And then you had to use
software and manually stitch every single scene together. And this
was a crazy tedious process. And then all of a sudden, these
all-in-one 360 cameras popped out and you're like, "Oh, OK.
Well, great. I don't have to spend a month stitching this."

Arash: Yeah, handheld for $200.
Yeah, exactly. These kinds of things are huge, huge game changers.

Alan: I went from $10,000 a
minute, to a $200 camera.

Arash: So, yeah, I think a lot
of it's "Come back and take a look at what it is now. It's a lot
different from three years, we promise." And what helps a lot is
also that I think overall in this industry, we've done a good job as
a whole of recording a lot of really good case studies. There have
been a lot of case studies where I would have expected this to be
something that either the company or the developer would have kept
super secret. And this does happen from time to time where they're
like, "No, no. AR is our secret sauce. We don't want anybody
hearing about what we did. We don't want anybody to copy us, we don't
want our competitors to take any of this stuff."

Alan: And that actually, from a
business standpoint, makes a lot of sense.

Arash: It does. But then by and
large, a lot of companies are looking at this as "Well, let's
look at the other side of the coin, which is that, hey, we can
protect ourselves as a very innovative company." This is just
now and we're already working on the future. So even if you did what
we're doing now, you're still behind, which I think is great, because
it gives the industry overall a look at what's working, what's not
working, and what kinds of gains are people seeing. And then when we
go to talk to a company that's, for example, a construction company,
we can pull a lot of different use cases for where other companies
have been super successful. And that helps to turn that tide, where
companies that are super risk averse and really slow industries, they
can say, 2Hey, wait, actually, a lot of companies like us are doing
this already, and they're really happy with that. So maybe it is
worth it for us to give it a try." And Alan, I've got to say,
thanks for all the stuff you're doing to promote the industry. I
think it's important to have a lot of conversations around this
stuff. I think the more information that's out there, it keeps
developers like us doing our best work, knowing about all the
solutions that are out there to be leveraged and not reinventing the
wheel ourselves when we need to do something, and then helps
companies to see how much exciting stuff is happening in space.

Alan: Wholeheartedly. And
actually AR new business model with MetaVRse, it's a combination of
us realizing that we can't possibly build everything for everybody,
even though we're really focused on training and education and
learning. We can't possibly build everything for everybody. So what
we're building is a platform marketplace for learning technologies
and content. So, for example, you guys have built this player on
Magic Leap that you built with CNN. Well, maybe they want to resell
that player as a generic player for other companies to leverage that
technology. Well, there we go. We can now put that on the platform
marketplace and make it available to every company, without having to
pay the million dollars of development costs upfront. And this can be
done for all sorts of content. We're really focusing on training and
education and learning content, but having a platform marketplace or
managed marketplace of this technology will lend itself nicely to
scaling this technology. As you know, it's not easy to make, and
companies like Extality are few and far between right now.

Arash: Absolutely. I think
standing on each other's shoulders is really the way to go to the
next step.

Alan: I think there's room for
everybody. And if we all work together, then good luck if you're not
part of the fold. [chuckles]

Arash: Exactly, yeah. And I
think we're going to see a lot of technologies raise up as the gold
standard of doing certain things the right way. And I think the
benefit of having people just take things that have been established
as, "Hey, this is the good one, use this." and then build
all the other stuff yourself. It's kind of like web technologies,
right? Like there was a point where everybody was implementing their
own authentication software. And now you don't do that anymore. You
just use one that's established to be really good and you can make
sure that it's already secure, because there's people that are
dedicated to solving that problem. Versus you want to build an XR
solution and you have to build every piece of it yourself. You may
not be an expert in the security part of it, or the network layer
part of it, or the graphics part of it. And we can lean on people
that are, to really get to what's really driving value for the
client.

Alan: Yeah. I think the the
billion dollar question is which technologies serve which industries
best, and which products are best to deliver these results.

Arash: Yeah, absolutely. And
we've been all over in different industries and we're seeing pretty
much every single industry demands something different of us. I think
of ASICS. That was another client of ours where we 'were talking with
them about building prototypes around some challenges that they had
with essentially with their distribution model, where essentially
you've got folks that are buying shoes to sell them. So a company
like a Macy's or a large retailer in Europe. And one of the
challenges for ASICS -- and pretty much every other shoe company --
is that they have to come to these meetings with all the shoes for a
particular season. And if you consider all the different color ways
and all the different models, they've got a pile of shoes that are
being shipped all around, everywhere around the world. And it starts
to kind of give you this a-ha moment, if you're deep in the XR world
of "Well, why can't we just digital twin everything? Why can't
we build really, really beautiful, very realistic looking shoe
models, and then put these in the hands of the sales and distribution
reps, instead of having to actually send around giant crates of shoes
or whatever."

Alan: It's crazy. And the same
with apparel. Apparel's a little bit more difficult, but shoes lends
itself perfectly. We spent a lot of time modeling shoes. Shoes,
watches, handbags, glasses, and apparel was kind of the--

Arash: Exactly. The rigid goods.

Alan: Yeah, rigid goods. And
then things that are fashion forward, they're always changing, right.
So when ASICS shoe -- although they kind all at the same eventually
-- but the shoes of 2019 are going to be different than the shoes of
2020.

Arash: Exactly.

Alan: And 'having to create
physical models is kind of-- it's almost useless now. You don't need
that. For how much money just in shipping that stuff.

Arash: Right. It's crazy. And
it's just like, you can take down all kinds of-- it's not just money,
right? There's carbon that you're saving, and there's-- it does a lot
of good for everybody. They can do the meetings more quickly, it's
easier to show things, pulling it up is much faster. Like, oh, you
want the blue, whatever GT-2000? We can show you the model really
quickly, versus "Oh hold on, let me go dig this out of the crate
of 300 shoes." And the thing that's interesting is we have
companies like Audi, where they've they've taken the digital twin
throughout the entire chain. It's used both in the design side, to
the production side, to the retail -- or in their case, dealer --
side. I see that for a lot of different industries. I mean, with shoe
companies, it's the same thing, the same way that these shoes are
being sold to retailers and they can figure out for their buyers and
that kind of thing. You could imagine these same models showing up on
websites to be used in the retail side, on the e-commerce side. Use
it in a lot of different places. You can get a lot of value.

Alan: A lot of times big
companies, their marketing departments don't talk to their training
departments. And one of the things that we realizing is that it's
expensive to make a 3D model, but you can use it across multiple
parts of the business. You can use the same 3D model to share with
designers. You can share with buyers. You can use it on your website
for marketing. You can use it for training. All of these things. Yet
it's kind of take some time to re-educate businesses on the fact that
this content can be reused and really creating this digital catalog
of 3D content.

Arash: Exactly. And it goes back
to what you were saying earlier is where in some ways now helping
businesses talk to their individual parts. And it's almost like a
transformation challenge and we end up being agents in that way as
well. It's more than just delivering XR solutions. It's "Hey,
have you talked to your marketing department? Has the sales talked to
marketing? Does -- like you said -- design talk to marketing? You
guys are all using the same file here. You should have a system that
shares everything." And then you go back to the question of,
well, the decision maker for marketing is different from the decision
maker for design. And you start to get into some of these interesting
corporate challenges. And so, yeah, you asked originally how is this
different from games? Yeah, that's a huge area where we just learn a
lot about. "Wow, this is inefficient. And we're not just giving
them efficiency by giving them XR tools, we're giving them efficiency
by -- in some ways -- just changing communication inside the
organization."

Alan: Arash, you've done your
work with ASICS. You've done work with CNN and Magic Leap. Is there
any other products you want to discuss?

Arash: Well, there's a project
we did with a small refrigerator company -- well, I don't know if
they're small, they're pretty big -- it's a refrigerator company
that's doing refrigerators with solid-state electronics. And so
normally refrigerators got compressors, and it's got all these like
fans and stuff like that. These guys found a way to make a
refrigerator essentially with no moving parts. So it's electronically
cooled. And this was an interesting marketing project for us, because
we had this refrigerator. And it's unique, because there's no space
taken up by a compressor. You can fit it in a lot of different form
factors. And this was interesting, because their marketing team just
had this vision and they said, "Can we put this in people's
homes? Can we put this in stores where they want to use this? Can
they customize it on-site? Can we make it interactive?" And this
was an exploration in just, "Let's say yes to everything."
Let's make a really cool ARKit tool that you can download off of the
app store, so that anybody -- no matter whether they're a sales
person for the company, whether they're a customer, whether there's
somebody who just likes refrigerators -- they can try this and then
experiment with how do you work with this? Where can I put it? What
does it look like when it opens? Can I change the skin to say "ice
cream"? Can I change the skin to say "yogurt"? Can I
change it to something completely different? And exploring taking
models from CAD, rendering them realistically in ARKit in this case.
Can be headsets, can be anything. And that started to become a model
for us. Are there other companies that can really quickly give us
their CAD models or give us something that we can photograph and
create 3D models through photogrammetry? And so that's kind of been
our focus in recent months, is looking at companies where they're
just ready to go. There was a bike company, same thing. They have a
really innovative bike platform, it's an electric pedal combo bike.
And they just wanted to show like, "Hey, this is a really cool
new carbon fiber bike, and you can get it in a lot of different
colors." And creating these just really easy -- kind of goes
back to the thing we said with shoes, versus in handbags -- can we
create a really easy model, where people can navigate it to their
heart's content, where with photos you're like, oh, I don't know what
that part looks like. I can't see that angle. But with a 3D model,
and especially if it's on the web or in an app, you can just rotate
to whatever angle you want to and peek at the smallest, most minute
detail and get whatever answer you're looking for. Kind of see it
however you want to see it. So that's that's been interesting. And
looking at how do we get that data? Does it come from CAD? How do we
get materials? That's been an area we've been really focusing. But
then the other really, really big area for us has been construction.
We've been working with Unity's Reflect team. And for people that
haven't heard of Reflect, it's basically a tool which allows
construction companies and architects who are working in a tool
called Revit -- which is kind of one of the industry standard .BIM
building information modeling tools -- to export those .BIM data
files, which are usually huge and contain millions and millions of
triangles, transfer those onto a mobile device so that you can
actually look at a model while you're on the job site or somewhere
else in an office in a meeting. And it creates a two way
relationship. So if you were to change the file in Revit, it
instantly updates in mobile. And we're looking at a lot of challenges
in construction and other large manufacturing organizations, where
there's just a lot of data, a lot of 3D data and getting that on to
mobile devices, onto mobile VR/AR devices. It's just a horsepower
challenge. And by being able to do these operations off-line, over
the Internet, we're talking a lot about remote rendering. We saw some
cool stuff coming out of Microsoft with that, the Apollo demo and
things like that. I think that sort of starts to paint the future of
how we can use this in larger scale industrial operations. And so
what we're working on, building tools around Reflect, that's been a
really big area of internal development for us with some of our large
construction clients. Visualizing things like build sites, build
process, how does the 13th floor come in, and what are the crane
swing operations that we have to do. And helping them really just
communicate better, being able to put the building right in the
middle of the conference room table and have everybody, all the
stakeholders looking at it in one place. With a model that's that
heavy, normally that's just not possibly, you've got to look at it on
a high-performance PC and you can't do it on an iPad or a Magic Leap
for something like that. So we're seeing a lot of big technologies
that are starting to push us in that direction, which is really cool,
making that possible.

Alan: I'm enamored by these
programs that are taking CAD files and BIM models and transferring
them into 3D for AR and stuff really simply, because that was a big
problem. How do you take these CAD models that you have that you've
spent millions of dollars on and convert them to 3D? It's in 3D, just
not the right format. So there's like a-- and then having this mess
of 3D versions and 3D model formats, you've got FBX, and glTF, and
OBJ, and all these different-- and then you've got Apple with USDZ,
and there's really a lack of standardization across industries for
this. And I think we're starting to see these things work themselves
out of it, which is great.

Arash: Absolutely. It reminds me
of the early days of images, too. We got BMPs and PCXs and every kind
of-- there were so many different formats, if you remember early days
of graphic design on the PC. And those all sort of consolidated. And
really, I think it ends up being, what are the major formats that
people support? What are the major formats that are becoming
interchange formats for different operations? And we saw it kind of
in the games industry, where things all kind of centered around FBX
and COLLADA. And I think we're gonna see that sort of with USDZ and
glTF. I think there's just a lot of tools like SketchFab and things
like that, where your experience with certain formats is lot easier
and we see people gravitating toward those formats.

Alan: It's interesting you say
that, because everybody in the world was moving towards glTF, until
Apple said, "Oh yeah, we got our own one!"

Arash: "Oh, by the way."

[laughs]

Alan: [laughs] "By the way,
we're Apple. So you can deal." But there's a pretty cool program
called Meshmorph -- meshmorph.io -- that will convert file formats
into USDZ.

Arash: Perfect. Yeah. I'm
speaking with Apple, and they're investing a lot in making that
process a lot easier. And again, those types of investments are what
will make people gravitate toward USDZ over something like glTF.
Yeah, let the best man win.

Alan: The biggest problem I see
is-- I've got a 3D graphic designer, who built just an AR business
card for us. Sent it over to another group that were posting the same
graphic on our website in 3D. And back and forth, back and forth. And
like oh, the textures weren't right and all of the bumpmap wasn't
right. It's like, God, this is-- there needs to be a standard, how
hard is it to send a bloody file? But it's apparently hard.

Arash: Yeah. I mean, the
challenge that I think we see is as we work more and more with this
is, that there's so much more data to 3D than there was with images.
In an image it's like we got pixels and what color's the pixel. But
here it's what are your materials, what is the shininess, what is
this thing, which map is being allocated to which channel, and where
you're dealing with these photorealistic physically based rendering
models? And it's important to get it all right, because different
renderers are going to use the bumpmap a different way. They're gonna
expect a normal map. I mean, there's all these kind of technical
gotchas that you can get wrong if you're importing from one format to
another. And again, I think that just underscores the importance of
this stuff starting to get more standardized. But at the moment, I
think I agree with you. It's a big mess and different exporters are
exporting differently. And you'll see two different ways the same
file formats are being interpreted. I mean, we had challenges with
FBX where it's just like, "Hey, you can store animation
different ways, or you can store the materials different ways."
And people aren't getting what they expect and leads to clients
getting on the phone and being angry.

Alan: Yeah. And when you're
dealing especially with retail, when you drop something in AR, it
needs to look like -- or even on 3D -- it needs to look like the
product that I'm going to buy. People are used to seeing a photo of
the product I'm going to buy. If it's skewed in some way or a
different colour or different shape or whatever, then you've failed
in doing that. And then, of course, the challenge becomes, "OK,
we made it look photo-real. It looks great. How do we make it
smaller, so that it works on all devices fast?"

Arash: Right. Exactly, yeah.
Still lot of model data is quite large. We're getting better with
compression. But yeah, you're right.

Alan: Oh, yeah. It's getting way
better. And then you've got 5G coming. People are like "Oh, 5G
will unlock this!" I'm like "Yes, but if you can get
compression great now, when 5G comes along, then you can just do
more." It still doesn't shortcut the fact you need to get
compression down.

Arash: Yeah, things need to be
efficient.

Alan: They do. Well, Arash, what
problem in the world do you want to see solved using XR technologies?

Arash: Oh, that's a really good
one. So me personally, I'm a pilot and what I'm looking forward to is
being able to wear a headset while I'm flying an airplane, and have
everything just laid out in front of me out in the open, kind of like
I think F-22 pilots have this with a $20 million headset. I want to
see every pilot being able to see other planes, all their runways,
all these kinds of things. And that being extended later to drivers
and that kind of thing. I think there's so much opportunity to give
lots of information to people operating vehicles.

Alan: Interesting you said that,
I was just at the I/ITSEC conference last week--

Arash: Oh, yeah. How was it?

Alan: --the
Interdisciplinary/Industry Simulation and Training Conference[sic].
It was amazing. I got to fly a helicopter. I don't know which one,
but it had the articulating propellers. So it actually can fly like a
plane as well.

Arash: Oh, right, yeah.

Alan: But I was in a giant
multi-million dollar simulator flying helicopter at one point. I
was-- I shot things. There was lots of things to shoot. I also got to
drive a tank, that was pretty cool. Out of about 400 and-- I'd say it
was about 450 booths there, over 100 of them had VR at them.

Arash: Yeah.

Alan: The fact that there's that
percentage, at least a little bit over a third of every of these
booths had VR in them. Or AR. I mean, I tried a gun range using the
Magic Leap where you're shooting a gun. But when you're looking
through the Magic Leap you see a full range in front of you, with
grass and trees. It was like, whoa. And it was a one to one exactly
to the gun. So when I looked down the barrel of the gun and shot it,
it was very accurate.

Arash: Yeah. That's no accident
that even early days, like in the 90s, 80s, we had heads-up displays
on airplanes, like it's just-- it's huge, huge game changer to have
information exactly where you're looking.

Alan: You've got kind of
pioneers in the space. People that have been working this since the
60s, 70s in these heads-up displays. And you don't really know what
you're missing until you have that data in front of your face.

Arash: Yeah.

Alan: Do you know Tom Furness?

Arash: I don't think I do.

Alan: Tom Furness was one of
those-- he's known as the grandfather of AR, because he was building
heads-up displays -- or HUDs -- for the US Navy and Air Force back in
the 60s, 70s.

Arash: That's amazing.

Alan: He's never left the
industry. So he's actually one of our mentors with XR Ignite. Yeah,
it's pretty cool.

Arash: Good stuff.