When we head over to Gamescom each year it’s usually the done thing to go and check out all the games. After all, there’s more than a few of them there, and here at Dealspwn we do like our games. However, we were presented with an opportunity to chat with some of the people behind the Unreal Engine 4 – the latest version of Epic Games’ proprietary engine – to discuss how it has been doing in the hands of developers, and in turn learn more about the community-driven Unreal Tournament project. In a day full of presentations to other developers, I was lucky to sit down with Technical Artist Zak Parrish, Lead Programmer James Golding, and Senior Designer of the UT project Jim Brown, to find out more about one of the industry’s leading development tools.
To start off, I had to ask about Epic’s newly opened UK office. Comprising mostly of the team at Pitbull Studio, last month’s announcement was a huge boost to the industry here in the UK. I asked the team what it was like officially bringing them into the Epic family and setting up shop on our shores. “I moved back over here in January to help set up the UK office, and that’s been super exciting to be back in the UK and seeing Epic make a commitment to the UK and embrace all the talent and stuff that’s going on here,” said Golding. “We already have a lot of developers using the engine in the UK and interested, so that makes sense, and we’ve had this relationship with Pitbull for a couple of years now, working super close with those guys and we’ve been treating them like peers, like part of Epic, for a long time, so it feels very natural.”
That’s not to say Epic didn’t have a presence over here before that, but it was significantly smaller. “For a long time it was just me and Mike Gamble, who’s our Territory Manager, so that’s two employees at Epic UK and now it’s increased somewhat overnight!” In terms of what the new base of operations was hoping to achieve, a centralised location to liaise with European studios was a huge part, although they are also looking to the future. “We want to grow some of those studios and take on some of the talent that’s coming out some of the schools in the UK and universities and so forth,” Golding continued. “I think both of those are good reasons to do it, and really make a commitment to the UK. I’m super excited to be back here and see that slowly happen!”
The conversation continued onto what exciting things the team have caught their eye recently. “I’ve been really excited to see some of the stuff that’s been happening since UE4 has been out from GDC,” replied Golding. “I’ve been a lot more involved since then looking at community projects and what small teams are doing, and stuff that’s not necessarily announced yet but looks really exciting, of all sizes so there’s a lot of really cool stuff around Europe – around the world, really.” Brown then interjected by stating that this sort of approach was the basis of the UT project. “We’re starting with the foundation of the game and letting the community build it, and now that the tools are out there people are doing stuff,” he said, mentioning how there were “old classic games that are getting re-made” as a result, along with new things “using our stuff as a launch board” for their own games. “It’s been really cool to see everything, from even just basic concept art to full games getting done,” he added.
Thanks to the open nature of the UT project, it means that community is helping to input alongside larger studios, something Golding gave an example of. “I’m on the forums a lot,” be began, “and some of the things people are doing, you see someone making a full mantling, ledge climbing, Assassin’s-Creed-style climbing and sharing it out with the rest of the community, and I don’t necessarily know where these people are! Like, they have a handle on the forums, so I’ve not met them yet, but we had a developer [visit us] a few weeks ago, we had teams from around Europe come along and bring their games, and it was really cool to see some of that stuff I hadn’t really seen before. Stuff is happening so fast that it’s really hard to keep up with what people are doing!”
While an official presence has only been in the UK for a month, Brown pointed out how Epic Games have been proactive in the UK development scene by supporting the Epic School at Staffordshire University for some time now. “They hadn’t made the switch over from UE3 to UE4, but they’re now using our [new] stuff,” he explained. “We were out there recently and did a talk to some indie developers, and I’ve stayed in touch with those guys as well. They’re actually starting to develop classes using UE4 specifically for the purpose of adding stuff to Unreal Tournament. So they’re actually teaching people who are going to do stuff that is going to be used… in theory! Or at least it’s available to them, so they’re getting professional experience right out of the gate which is great!”
Over the last few years we’ve been wowed by the tech demos, such as Elemental and Infiltrator, highlighting what the Unreal Engine 4 can do. To that end, I asked what sort of response it had garnered from developers. “I think the studios look at that as kind of the high bar for graphics end result,” said Parrish, “but again that was a cinematic tech demo, and that’s a lot different than actually powering and driving game logic at the same time. It’s a great way to showcase the visual capabilities of the engine, which is exactly what we designed it for, but in terms of setting up a game it doesn’t even really scratch the surface.” As he continued, he explained that the community were helping to demonstrate what the engine was truly capable of. “We’re making games, [and] the community is really starting to come together and show how they can use Blueprint to make their own games and game systems. So the real demo, the real power of Unreal Engine 4, has only been exposed in the year since we did Infiltrator, and it’s because we’ve been able to put this tool into everybody’s hands.”
The news earlier this year that Unreal Tournament would be returning as a community-driven development was not only music to my ears (being a huge fan of UT ’99) but it also provided a great way to get involved with the game from a foundation level. I asked what the initial feedback from early adopters of the project had been. “I thought it was a great idea when we first started it, and I’ve actually been really surprised and pleased by how well it’s gone so far” replied Brown. “Looking back, that was my in into the industry. You know, modding on YouTube was one of the reasons I got hired at Epic, and when I look back at the team that’s building this current UT, almost every single one of them… actually, I take that back – every single person on that team got their in into the industry because of the original [UT] except one, and that’s Steve Polge who’s the project lead who built the original UT!
To further the point, Golding then pointed out that Polge started out by modding the original Quake.
“Exactly!” Brown exclaimed in agreement. “So our whole point was to go back to get that community feel, and it’s great that we have people that came from that background so that they kind of understand it, and it’s nice to see some old faces that are creeping up into our forums, and it’s really interesting because we grew our community really, really quickly to the point where we overwhelmed the Unreal Engine forums and had to split off our own! But people are downloading UE4 to build stuff for UT, and so again this is all like a loop where one team is trying to outdo the other, and it’s been really good.”
In terms of interest, Epic knew there would be a big crowd wishing to get involved, but even they were surprised by the numbers of people who joined initially. “The UT forums overtook the UE forums,” explained Golding. “Like, the UE forums were busy, and then I think it took them a week for you guys to overtake us and all the people on our forums were like ‘All those UT people keeping coming over here and posting their Flak Cannon in the Physically-Based Rendering forums – put them somewhere else please!’”
Brown went on to praise how passionate the community was being about the project. “This is our chance to tap into that and just harness that passion [while making] them feel like they can actually contribute, to be useful to the game,” he said. “I mean, our team is only six of us, and we’re building a triple-A level game that in the past we’ve never done with less than fifty to one hundred people, but we’re just taking all of that from the community so it’s almost like we are managing instead of creating content or curating and training those people. It’s cool!”
“It’s pretty cool, because the source code’s all up there, and I’ve seen that people have taken that, made a distribution,” Golding explained. “Like, there’s that NeoGAF thread. Someone sent it around and it was really interesting to read that thread.” Brown even explained one of the more ingenious things some of the community had put together. “There’s a whole IRC channel right now where you hit “!UT” and it downloads the entire game and installs it on your computer and you ready to start playing! For free, just like that!”
So has the entire process has been a smooth, argument-free process for everyone? Nope, but that has been a good thing overall, as both Golding and Brown were about to explain. “It’s so cool that the movement is such a contentious thing, and we’ve had so much debate over the years” said Golding. “Like, I worked on UT 2003, 2004, and UT3, and so seeing the feedback over that period with the different games, the different loves and hates, [with] this one the feedback has been pretty positive to the movement speed stuff so far.”
Brown continued the point by explaining how the community has split into three groups – UT ’99, UT ’03, and UT3 in terms of preferred gameplay. “Everyone has their favourites, and we set out a basic line for what we think feels right to us,” he began, “but we’re putting everything in Blueprint which is our scripting language so anyone can go in. You don’t need programming knowledge or even artistic knowledge – you can go in, and we just put that out there and so people start creating demos. Like ‘this is what it looks like with this movement speed’ and ‘this is what it looks like if you jump this far’ or ‘what if I turn the gravity up here?’ They get into these weird philosophical arguments on the forums, but it’s great because those are the same arguments that we typically have in the office and so instead of it being just this small group of people, the entire community is having this conversation.”
“In theory I think we’ll end at a point where it’s more satisfying to more people because more people have had input, if that makes sense?” he continued, with Golding adding that thanks to the transparency of the development it should mean that “even if they don’t love where it ends up, they’ll know how it got there and why it’s there.” Brown then highlighted how sometimes the community can focus on the details a little too much. “Honestly, it’s hysterical when people get into these massive flame wars towards ‘the value should be 50 or 50.5!’ and we’re like ‘it really doesn’t matter!’ [laughter] ‘Oh yes it does! [they say.]’”
One of the big features of the Unreal Engine to have been touted in recent years is the scalability – specifically with mobile devices. With more and more developers looking to take advantage of the growing market of app, I asked if the interest in UE4 had seen a strong increase by smartphone developers. “There’s two parts [to that],” Golding began. “One, building rendering like what we can show here [points to the big screen showing off development tools] like the fact that our 3D high-end rendering scales to mobile, where mobile is now such as iPhone 5 & 5S, it runs really well on those things. It’s very forward looking, but you know, you look at K1 stuff that nVidia showed and it’s clear that that’s where mobile processors are going. They’re just going to get faster and there’s going to be more convergence on what you have in your pocket to what you have on your desk in terms of [visuals] so that’s super exciting. It shows that unlike UE3 where there was a very different pipeline for building Infinity Blade on a phone versus some desktop content or console content, now it’s a much more unified pipeline. You build it up once and it’s scales automatically, but that feels very forward looking.”
“It feels like the right way to go, but at the same time we’re also seeing a lot of feedback from the community building simple content,” Golding continued. “That’s why we did the Tappy Chicken game which began a joke but it actually became quite helpful in looking at the performance of that, looking at how we can make it run better, looking at all sorts of games that people are trying to make. Very small teams [and] 2D games [are] an area that isn’t our traditional comfort zone, and now we’ve got one of the guys, Michael Noland who’s been working on a paper 2D plugin for the engine in his own time for a while. Now we’ve really pushed that to the forefront really giving him the time to work on that, and we’re trying to make sure that the engine will let you make 2D content and run just as well as if you’d written it from scratch or a specific 2D engine because people think of the engine as about rendering.”
“That is a large part of the engine, but there’s so much more to it that you get when you use it,” Golding explained. “You get all of our Blueprint scripting, you get all of the debugging tools, all the profiling, all the loading & saving, and all of the gameplay. All that stuff you can still use to make a 2D game or something much simpler than you would traditionally think as mobile, so we’re trying to make sure that the rendering stuff works well there but you get all the other benefits of using Unreal.”
Returning to the topic of the Unreal Engine 4, I put to the trio that over the last year there have been other proprietary engines that have appeared, gaining popularity and utilising similar subscription methods being used by Epic. With that in mind, I asked them what truly sets Unreal Engine 4 from the alternatives available. “I think it’s the openness of the development of the engine,” Golding answered. “I think the fact that when you get the engine you get the full source code, that we make our full roadmap available for our developers who are on the forums. We’ve tried to start from there – everything else comes from that. Everything comes from a change in attitude which is saying the people out there are smarter than us. We need to learn from them and understand where we’ve gone wrong, understand what we can do better, and we’re going to do a better job by being very honest about where we’re going with stuff so we can change if we need to. It’s already paying huge dividends. The fact that the source code is out there, that we’re completely open, every time we check in is out in public. There’s little bit of stage fright to that, but the ability to point to a bug fix if you make a forum post, I’ll maybe check that every day I go and look at some stuff and I fix it, I can reply to the thread and say ‘here is the fix – I just checked in.’ Even better is that people will find the bug, fix it, they’ll submit the fix [or] optimisation back to us.”
Golding continued by highlighting more of the ways the community has surprised him. “That sort of two-way communication with the community…. There are some amazing smart people who are maybe working at other games companies, maybe working at other technology companies who do this in their spare time, and the whole Linux port! The fact that our tools are on Linux is a community project. There were a passionate group of people who wanted the tools on Linux so they just did it, and we worked with them and integrated those changes and I think that is the biggest difference. The fact that we really want people to get into our code, take it apart, fix it, improve it, tell us what they think of it, and we really do want to find a way to make our plans adapt to what people want.”
As the discussion went on, Brown & Golding explained how they consider the current version of UE4 to be a beta stage that was still growing, and how it was strange for Epic to be building tools that they were not using ahead of others. “That’s the thing – In the past it was difficult for us because if we didn’t use it,” said Brown. “It would languish or it wouldn’t be fully featured, and we would be like ‘well we’ll start the implementation’ and then offer it to people, but now other people can help us finish that. We have really shifted – and it’s an entire shifted company focus – away from building our own games to building a engine that we can use as a studio but so can everybody else.”
“We’ve seen some really cool benefits to that,” added Golding. “We’ve seen things like making the engine more accessible means even huge teams have an easier time getting new people onto the engine, so if a big company like a Lionhead or something, they like it when we spend time on accessibility. We thought that might be a bit contentious but they’re like ‘oh no-no-no! When you make tutorials for content that makes our lives much easier [so] when someone new starts we don’t have to have someone spend three weeks training that person. We can just point them to all these video tutorials that accessed online or whatever, and also what we’re seeing is there are people who are using maybe their own tech or using another engine, who grab our engine in the evenings and start to play with it.” Golding finished by stating it was strange not knowing who was using UE4, but that the trade-off was “you’re always surprised by who has had a play with it and the feedback you get,” which he said was “definitely exciting.”
I ended the interview with my traditional closing question – in your opinion, what is the most badass thing about Unreal Engine 4? Usually developers keep their answers pretty short, but not these guys – the three of them were clearly excited by what UE4 is capable of, and it showed in their responses. Parrish was first up, “Oh that’s easy – that’s Blueprint. To have a visual scripting language that gives you all of the flexibility that Unreal Script gave you in Unreal Engine 3, but to have that kind of power [given] to someone who isn’t necessarily a programmer, and to be able to actually visualise it as you construct it? It really opens the door for gameplay creation to a lot more people. I’m not a programmer by stretch of the word but I’m very comfortable sitting down with Blueprint and making my own game systems or entirely complete games, and no other system has ever allowed me to do that.”
“That for sure!” agreed Brown. “The only other thing I would add on to that is just general accessibility, whereas I think UE3 was very intimidating to a lot of people, ‘cos you know we used to create these great big console games, but now UE4, with the new business model, with the new tools, just with the ease of use of giving in to it, anybody can do something with it, and again with UT that’s what we’re seeing. We’re seeing somebody who’s never built anything before but was a passionate player can now contribute to the game, so anybody can get in and build something which is great!”
“I’d say the users…” began Golding.” There you go, there’s a cheesy answer, but I think it’s true. To me, the most exciting thing is seeing either what the guys at Epic do or the community does because it’s always surprising. It’s always pushes the engine in ways I hadn’t thought of or we hadn’t considered, and then we can respond to that, so I think for the most badass thing about working with the engine is what users do with it. I know that sounds kind of cheesy but it’s true!” “Well that’s because in the past we would build a game and then the engine would pick up steam,” replied Brown. “Like, people would see what it was capable of, and this time we haven’t really released any full blown big game on UE4. We’ve done samples and demos but other people are doing that – they’re filling that void. It’s become an actual community instead of people using the engine.”
To finish off, Golding spoke of a topic he hopes to see more of with UE4 in the future. “I think something I’m excited about going forwards is more procedural content,” he said. “I think one thing the Blueprints do is they enable designers to build smarter things in their levels, so our process in the past was about putting a ton of little objects in the level, meshing out a level, and it was part of the process that Jim knows a lot about. I like the idea that we’re moving away from so much of a labour intensive process that we can really empower artists to make their levels smarter so that they can be more free to do stuff, so if they want to build a big world they can built a more procedural world and they have those tools at their disposal, and I think we’re going to add more and more. That’s where I see a lot of our Blueprint and tools going, [allowing] artists to be more powerful so that they don’t need 400 people to build the world, that you can start to empower that stuff.”
“I think that procedural is an overused term,” Golding continued. “I think we’ll see some games come out that use the word ‘procedural’ that are actually kind of boring. I think you need to have the procedural tools to be in the hands of someone who wants to give it personality. They need to be an enabling thing to let someone who has a real vision and real sense of design & narrative express themselves through those tools. I think that’s going to be a really powerful combination, so I’m looking forward to that in the coming years.”
A huge thanks to Zak, Jim and James for taking the time out to chat to me! For more information about Unreal Engine 4 you can head over to the official website, and for more on Unreal Tournament by checking out the official blog and its associated forums.