The REAL TIME Digital Revolution | EPIC GAMES | PI Apparel

Hello, everyone. Thank you. Welcome. Thank you for being here and having us. My name is Sally-Anne Houghton. It was really close. He did really well. And I lead partnerships and business development for fashion and retail at Epic Games. So who are Epic? Everyone kind of knows they're Epic, but Epic is— it's a software company. It's a gaming company. It's an ecosystem of tools and platforms. And dare I say it, it's a metaverse company. Sorry, Facebook has kind of ruined it for all of us. So we develop tools to empower creators and unleash creativity. And that creativity may live in a game on a gaming platform like ours, Fortnite, where you can play shooting games, play with LEGO. You can have a racing game. You can have a music experience with your friends. Our ecosystem is full of diverse and engaging experiences that are made by us or creators. And our tools can be used to make TV, film, animation, amusement park experiences— any interactive 3D experiences. And these can be found on low— online, in store, on mobile, and in-house. And importantly for us here today, these tools can be used to revolutionize important creative workflows. For your business, the design process, the approval process to visualize for R&D, making 3D environments collaborative and iterative. And the Unreal Engine is the pumping heart at the center of our ecosystem. It's a sophisticated game engine that enables real-time experiences with a visual quality that is unmatchable. So why is a gaming company at PI? So high fidelity, real-time 3D lets you make the right creative choices earlier, quicker, and first time. And that is good for business. So my role is to be a conduit between Epic Games' tools and the fashion and retail space to ensure that Epic's development of open creative tools allows— and creative tools and the fashion industry's development work hand in hand, side by side, so that all the opportunities of real-time can be actualized. And this is a really special time, because we're at an inflection point where technology and fashion are able to finally collide with spectacular results. Visuals, textures, simulations can be delivered at scale. But not necessarily easily, which I'm sure many of you here know and have experienced. And I'm hoping— I'm sure many of you have been part of your internal 3D pushes, proving out 3D, showing its value and its benefits. And I've seen an explosion of small teams building parts of pipelines, exploring integrations, and creating bespoke solutions. And we want to help you to continue to do this, to push the boundaries, to expand the use of 3D, and build more creative and more robust pipelines. So how are we going to do that? So no one can do it alone. We need an ecosystem of users, partners, and integrations, and only then can we scale and make 3D more accessible and accepted. And only then can we build complete and compelling real-time pipelines. So that is what we need to discuss, and that is what we are going to discuss now, today, here. So I have an incredible group of people here to join me on the panel, and I'm going to hand over to Sophia, who is going to introduce himself and the rest of the panels. And thank you very much for being here, and thank you everyone for being curious. And, um, I will sit down. Sophia, thank you.

Salyan. Good morning, everyone. My name is Safir Belhally, and I have the privilege of facilitating this, this conversation that I'm sure will be extremely interesting. I am a consultant. I spent 18 years at VF Corporation. I led innovation at Vans, digital innovation at the group level. I'm also an educator and I am Really excited to hear what everyone will be able to share today. Shannon?

Awesome. Hi, everyone. I'm Shannon Molden, Director of Digital Product Creation at Athleta under Gap Inc. Brands. And I work really closely with the technical design team, design teams, and product creation teams, as well as our vendor base to create digital workflows that replace physical samples in our design and development pipelines.

Hi, I'm Holger. I have been working quite some years in automotive and aeronautics, and afterwards I joined Adidas to build the digital creation pipelines that are still existing over there today. And afterwards I started to join a company called 4D Pipeline, where I'm right now the director of client services, and we are building bespoke applications around 3D computer graphics. We are trying to support our clients with really solving the challenges at hand. And for example, we are building parts of Adidas pipeline but we are also part of the Epic teams, and we are working closely with the Twinmotion and Datasmith teams over there. And again, our core is to really support on the technical and tricky parts of 3D and computer graphics. Over to you.

Thank you, Holger. My name is JR. I am French, based here in New York, and I lead the technology team of Theodo. We are a software consultancy, even though I am a mechanical engineer by training and started my career in the car manufacturing industry. At Theodo, we spend our time building the plumbing between product creation systems, so getting data in your design application into your PLM flawlessly, so really improving operational efficiency of the teams involved in product creation so they can really focus on using the tools they love instead of half the time inside the tool they love and half the time sending emails or doing like double or triple data entry. So yeah, data plumbing is the short way to put it.

All right. Love the term. You should patent that. Let's talk a little bit about the DPC journey. We have, you know, a couple of really interesting stories. You know, Shannon, you're currently in an organization, Athleta. Holger, you spent quite a bit of time at Adidas. When did your journey, your DPC journey start? What were your biggest hurdles and what have you learned so far? I know this is a big topic. It's a meaty one. Right, it's a meaty one. But how would you summarize that?

So we started, I've been with Athleta for about 4.5, 5 years and came in in a 3D role. So we've been working in the space for about that long. But the first couple years were mostly around POCs and sort of project-based work. We've really been trying to scale and make a larger impact with our 3D workflows over the past 3 years. So from sort of focusing on selecting a tool, building our foundations, enrolling teams, working with our vendor base, just really building a mature process for digital sampling within all the way from design to product approval. Challenges, there are so many. It's hard to know where to start. But I think that one really relevant one that I'll touch on, because I think it's really sort of meaningful to the discussion that we're having here, is around once you've reached that level of maturity in digital product creation and workflows, how do those assets go to work for you within the brand? So we spend a lot of time right now, and I hope I'm not alone, rendering and bringing renderings onto Miro boards. And those renderings very quickly become irrelevant if there's a style change or a colorway change. So I think that's one of our biggest challenges today is just how do these assets connect and stay live and stay relevant so that they can really work as hard as we need them to for us in the business.

You're not the only one.

Yeah, I'm glad I'm not alone.

Hugo?

Yeah, now at Adidas, I think we started at 2005, 2007 introducing 3D mainly for sample reduction, mainly run by our factories back then. So they were doing all the 3D modeling, creating shoes in 3D, creating garments using V-Stitcher in 3D. And at some stage when I joined the company, actually 2011, we started to pull everything forward and then trying to really introduce 3D into design and development, not only doing sample reduction, but really trying to integrate 3D as an essential part into the workflow. And when we started with the 3D introduction, it was primarily on design. But at some stage, we realized this needs to be an integrated approach. So that was already the first challenge that we saw, that you cannot just do it in silos. It needs to be an end-to-end integrated process because actually it starts with marketing, it starts with What is the vision? And then how do you want to take the information that you need to transport from step to step to step? And at some stage you start enriching it using 3D. Yeah, it's not the paradigm itself to have 3D, but it needs to be also placed at the right, right point in time. And that is literally what was done was building a complete end-to-end digital creation environment that transported data and 3D assets from start into the factories, into different applications downstream, so that in selling, in e-com, in marketing, all those assets could be used. But the biggest challenge I saw was adoption, because the tools are— it's something different if you use a pen or if you use Illustrator and you're just using something that you have grown up with for years rather than switching to a 3-dimensional space, right? I mean, 2-dimensional versus 3-dimensional, it's a bit of a different paradigm. It's a different level of complexity. And just getting people trained, understanding how you can create in 3D, how you can get the same creativity in 3D because you're more constrained on a piece of paper. You're not constrained in 3D. You are sometimes constrained because you can't do certain things. You can't do stretchy looks or overamplify certain features. So this was That's the tricky part of really balancing it and then scaling it. I mean, doing this in small prototypes, in small teams, and with a few people that works. But doing this for hundreds of designers, hundreds of developers downstream into the factories, getting everything integrated, that was one of the biggest challenges we saw. And I think it gets solved.

Yeah. I think, you know, most of the brands here are if not all, are somewhere on the DPC journey, right? At a certain point, there's a shift that's happening from the POC stage where you're trying things with, you know, smaller collections, things like that to prove things out. And when you start thinking, okay, how does this work at scale? So I was wondering, you know, this might be for JR and Holger, Have you established pipelines? So you started speaking about that, Holger, but I think the key point that you mentioned was end-to-end. Have you established pipelines, full pipelines that allow organizations to scale the DPC efforts? And if so, what role does real-time 3D play in that?

I'll give it a stab at it. So, full end-to-end. So personally, in our experience at Theodo, we haven't really spread it fully end-to-end. We see a few companies that have this intent and sort of this vision to, like, from the beginning of the effort, that this is going to be an end-to-end thing. But in our experience, companies actually first put the finger on one element of it without seeing the bigger picture. And that's quite interesting too. You have to solve that first step effectively for them to maybe think, oh, actually these two steps around it could be also integrated and so on. Not everyone has the global vision. And I'm sorry, I forgot the second part of your question.

The role of real-time.

So real-time, we really see— so like you said, I think the best example is what you said. You use Miro, you have renders in there, and they're outdated. And I think that's the critical point of like the data needs to flow. Everything we're talking about is just representation of product data. So if your data does not flow anymore, at some point you're going to make a decision on something that's outdated. So you are going to look at the line plan that is actually not up to date, so you are going to make a decision. So real-time is really critical and extremely hard to achieve, but made easier with things like Unreal Engine. Because if you have an end-to-end— like, that is really the dream. You have an end-to-end integration between all your stakeholders in product creation. Data flows seamlessly between all these silos, and data can be represented at any stage in real time. It means literally, if right before this line review meeting, a designer made a change that was approved, it is going to update the— when you look at it, the line plan is going to be up to date. So that is really the dream. If you do not have real time, you have really only solved part of the issue, because you did not want to do 3D just to do 3D. You wanted to do 3D because you wanted to make better product decisions. So if you have 3D, you still have not solved the problem at all. In a way, if I want to be provocative, you have not really solved the problem at all. You will only have solved it when you have real-time 3D.

Holger?

Absolutely. I think one of the biggest challenges is that you want to make decisions very fast, not only as a group, but only specifically if you are designing your products, you want to have interactivity. And most of the tools that existed before 3D do this. Photoshop, Illustrator, drawing something that's pretty much real time. And you cannot take that away. We've seen examples where this was taken away and that just led to non-adoption because it basically means that people do not want to use that software or that solution. So you have to be somehow integrated. You need to be able to have the same performance, the same speed, the same interactivity. And that is what you gain with 3D and with real-time 3D specifically. And then you have the second challenge, as you mentioned. Yeah, then you have all your designers doing, all your developers doing everything in 3D, and then you need to generate the images. And there are companies right now that render literally millions of images every season in a couple of days because you have your season cutoff, right? You have your handover. At that point in time, you need to generate for all products, all latest images. And luckily there's technology like cloud and you can scale that, but it still takes hours and days even though it's a massive operation. And introducing 3D at that stage, so in real-time 3D that you don't have to do all these cumbersome renderings at some cloud, big, big, big cluster, right, that spawns literally thousands of nodes. I think that's going to be one of the biggest advantages that we will see very soon once something like Unreal gets introduced into those toolchains.

So yeah, maybe to summarize it, like, if you think that by introducing 3D you are going to make the company move forward, you might be extremely wrong. Because if you just introduced 3D, all you did was make everything more complicated and make those iterations go even slower because you asked everyone to do an extra job on top of what they were doing. And that's a big challenge. But if you have real-time 3D, that's when really you go back to that same speed of iteration of like a team around the table scratching things, like doing things on paper and in like real-time interactions, real-time iterations on product ideas.

Yeah, this was one of our biggest learnings at VF, you know, granted with the complexity of having multiple brands, multiple product categories and trying to streamline the processes. We very quickly understood the power of an end-to-end, as you mentioned, end-to-end pipeline that at scale is powered by real-time 3D. Because that's, you know, when you're in POC mode, you can manage or you can do things manually. But when you're dealing with tens of thousands of SKUs, you cannot introduce manual processes that are not going to stay up to date and that are going to jeopardize the whole thing. Let's talk a little bit about internal versus consumer-facing. We talked about end-to-end, concept to consumer. Fashion is adopting 3D workflows internally. But there's also a really interesting application that is consumer-facing in terms of consumer experiences, whether they're brand experiences or product experiences. How do you see these internal workflows and the work that's happening internally also benefiting from the real-time nature of games and interactive 3D?

I mean, for us, I think it's very much about the consumer appetite for newness. And as long as you're relying on photography of physical assets to present newness to the customer. It's extremely limiting, time-consuming, and expensive. So I think that there's a huge opportunity for us to leverage what's coming out of our 3D development design workflows and make those consumer-facing. But it's just so important for it to be exceptional from a visual standpoint so that it's not obvious and disarming for the customer. So I think that that's sort of where we're looking at it. We're not there yet, but definitely the direction we're trying to head.

Yeah. We had— it's interesting because one of the things that you realize very quickly also is that once you've built these assets, right, they don't just sit there to be rendered or to be visualized. There's so much you can do with them, right? Jara, I was wondering if you— what thoughts you had around that, so how these assets can really benefit from these technologies.

So one thing I really like to talk about is, again, this idea— so I look at things from the data angle because that is what we do. I think it can be eye-opening. When we think 3D, we think of a 3D file that contains colors, textures, etc. When I see 3D, I just see a white model that I want to apply a BOM to, onto. So I really want to apply the BOM at the moment when a designer is going to look at it, at the moment when there is going to be a line review meeting. I want to take that 3D asset and apply the BOM that I would get in real time from the PLM. I can switch between my BOMs. The thing we try to apply with our clients is really have this workflow where they can have one 3D asset that is the golden source of truth for what is the shape of my product, the same way they have one golden source of truth for what is the bill of materials of my product. And when you want to look at it, you look at it, you just apply the bump to a 3D model. It's kind of the same idea of, in a spreadsheet, in an Excel spreadsheet, the pivot table is one way to look at that other sheet that's full of data. And you can have many, many pivot tables. They're just different interfaces into the same data. So to me, whether you're looking at your product in Adobe Illustrator in 2D, in a fancy real-time view somewhere, or in a render on your e-commerce platform, it should just be different representations of the same data. And so ideally you get as close as possible to the, that real-time data. So ideally every e-commerce website has an Unreal Engine representation of the product that's super realistic and that's up to date. Your designers that work on color and materials, they look at a real-time representation that is the combination of the geometry that may have been updated 1 minute before and the materials that— maybe there's a new view scan that just came in, and they just look at it in real time. So that's really the idea of separating each aspect of a product— geometry, materials, pricing, environmental impact, whatever you want. These are just data points about a product.

Sounds like a dream. Love it. All right, let's talk a little bit about what next. I want to make sure we give ourselves enough time for Q&As. So the consumer expectations in today's world are changing, right? They're changing really quickly. Consumers are expecting more from the brands, more from the product, and then more from the point of view of the story and the experience around the product, right? So what is kind of the next step for an organization, a brand with an established DPC pipeline in terms of scaling these efforts and meeting the expectations of the new consumers? Shan, do you want to—

Yeah, I mean, I think just like JR said, it's a matter of being able to have one version of that asset that sort of serves all of the purposes that we're— and problems that we're trying to solve for internally with 3D, but then also having that be ready and available for consumer-facing assets. Right now, with everything that we do with a digital asset, we render it in at one certain resolution for Miro, and then we render it at another angle and resolution for throw— a consumer insight test. And it's a completely different asset if it needs to go consumer-facing. So it's how do we sort of create that one ideal asset that is the living version of where that style is in its CC assortment, in its build. It's just sort of an up-to-date representation of the style so that it can be used for also consumer-facing consumption.

Love it. One last question I wanted to ask. So we have, we have very similar backgrounds, J.R. Halliger and I, in that we started in the automotive world and then we transitioned into fashion and technology. Well, into technology and sometimes fashion. But I was wondering, Sayan, what can we learn from other industries? You know, I'd say the We talked about this yesterday, right? The automotive industry went through this digital transformation way before all the other industries. What can we learn from what's happened in other industries? And what is holding us back? What is holding the fashion industry back?

So first of all, we should take comfort in the fact that other industries have done this. Like, it feels, you know, it feels like, you know, we've talked, Shannon, about the fact that, you know, it feels like you're in the trenches. It's like, it's hard. Like, newness progression is hard. But other industries have done this. They have created end-to-end pipelines. They have fully implemented real-time product creation. And they've done it like piece by piece. And they did begin with proof of concepts. And they did then evolve those proofs of concepts. But they had to get buy-in across the company. And it took time. And I really think that it, it, it's about, it's about people and it's about bringing 3D to the people of your company. And that cannot be asking every single person to be a 3D operative. Like we don't, we don't want that. And so I think that part of the solution is, you know, proof of concepts, then slowly scaling, showing the value, showing the worth, proving it out across the company, um, piece by piece. Piece. But also, I do think that we need to start thinking about how 3D can go further and wider than the 3D department. And that comes from company-wide interface. How can we bring— like, most people have invested in a 3D viewer, which I think is great, and it lives on their e-commerce site. Every company should have a 3D viewer. Everyone on their desktop should have a 3D viewer. Because if you're making a product, and that is your work all day, every day, and it's being made in 3D, then everyone should be able to interact with it. And I think it's just thinking company-wide and thinking of how people can become invested in 3D will allow your 3D to be end-to-end and to be cross-company.

I love that. This is actually a great, great place to stop, understanding that, yes, technology matters, but ultimately, Ultimately, it's about the people. That's a great message to take home. I'm looking at the timer. We're at 0 minutes.

It's a really quick question.

Do we have time for one question? One question. Do we have anything on Slido? All right, we have a few hands back there. I think we should just continue for another 30 minutes. Can I?

Let's crash the agenda, huh?

Yeah, right? Real time.

Hi, my name's Katie.

Sally-Anne knows me.

Hi, Katie.

My big question as a technical designer is, what do you see the biggest challenge and hurdles that you have with technical design and solutions that have implemented waste reduction and other things within that chain?

I can take that one. I'm also a tech designer. So I think the biggest unlock for us, and we're extremely lucky, has been standardizing the processes that we have with our vendor base, building standards that drive consistency you see in how the samples are built that we receive so that they're user-friendly for our tech designers to work with. But there's also a piece that you have to sort of acknowledge that everyone learns at a different pace. So for scale, we rely very much on our vendor base and then the internal processes that we do. So building a concept from scratch or actually fitting something in 3D, it's much more reliant on the user readiness, openness to learn, comfort with the tool, and the product category that they work on. So I think it's important to set up the standard processes with your vendor base that feed the machine, give you the assets that you need when they need— when you need them to service the business. But then also make the space and learning opportunities for your teams to engage with the assets in a deeper, more meaningful way. Way without the pressure. Because if they feel like the business is counting on them to build X amount of their styles in 3D, they're going to sort of feel suffocated by that challenge at first. So it's important to leverage your vendor base and, and also meet people where they are.

Thanks for the question. Great question. Well, I think that's, that's all for us. Thank you all. For your time in sharing your thoughts with us. Enjoy the rest of the conference. Thank you.