Battlefield 6 Studio Technical Director Interview: “We Learned a Lot from 2042”

ComicBook sat down with Battlefield 6‘s technical director, Christian Buhl, to talk about how the team at Battlefield Studios is preparing for the massive launch of its long-awaited shooter. Battlefield 6 is shaping up to be one of the biggest games of the year and fans are chomping at the bit to play the final product. It’s a moment of redemption for the franchise after Battlefield 2042 crashed and burned, leaving fans hopeless for the series. However, some are still waiting to see if that shoe drops and the game fails to deliver on expectations. The Battlefield series, even in its finest hours, is known for having big issues.

Battlefield 4 was crippled by network issues, 2042 had server problems for a while after launch, and the games are known for being titans that push hardware to their limits. With that said, fans who played the Battlefield 6 beta were surprised to see that the game had pretty impressive performance across the board. That’s the result of technical director Christian Buhl and his team working for years to create the most stable, performant Battlefield game to date.

Not only that, but they’re also making efforts to make sure Battlefield 6 stands out as one of the most fair shooters on the market with the studio’s extensive anti-cheat efforts. Although some of their methods have caused frustration, Buhl describes it as a necessity to creating a fair game that keeps players engaged. You can read our full interview below.

This interview has been lightly edited for clarity and length.

ComicBook: As far as I gathered, this is your first Battlefield game that you’ve worked on, is that correct?

Yes, sort of. So I’ve been back at EA for about three years. I actually was at EA for about 11 years prior and during part of that I worked on the Frostbite team, so I was actually a technical director for Frostbite. So I think I’m in the credits for Battlefield 1, but as as part of the engine, not part of the Battlefield team.

Was there anything you wanted to bring into the technical side from the outside coming in?

The answer to that is absolutely. I left EA about seven or eight years ago. I worked on League of Legends for a number of years and then came back and one of the reasons I came back was I was excited about the opportunity to help build Battlefield in a way that was focused on performance and stability. It’s not necessarily the sexiest thing, but obviously we learned a lot of lessons from Battlefield 2042, and I wanted to kind of help start Battlefield 6 as a game that was going to be performant and stable the whole time so that when we launched, people wouldn’t talk about [stability].

I always tell people that if players are talking about stability, we failed, but that’s what I want. I want nobody to notice our stability because it just works. Obviously, it’s not going to be perfect, and I want the game to be performance on whatever hardware or platform you’re playing on, and that means different things if you’re on a mid-spec machine versus a really expensive, beefy gaming machine.

But we want to make sure that whatever we expect the experience to be at that level, that’s what you get and you get that reliably. So for the last couple of years, we’ve put a lot of effort into building up huge test farms with all of our mid-spec, higher-end spec, Xbox, PlayStation, running the game over and over and over, capturing crashes, capturing performance data, doing a ton of internal play testing.

We have play tests scheduled literally every day in every studio that we’re making this game. So we have four or five or six play tests a day at least. And we’re obviously gathering feedback on if the game is fun and those things, but also just collecting a ton of data on whether the game’s performance and stable. Then we just put a massive effort into making sure that it is right and making sure that when this game goes out, you can just play the game and enjoy the awesome gunplay without worrying about whether your game is going to crash or whether the frame rate is going to stutter.

There’s been some rough launches in the past, some ups and downs with Battlefield. How are you learning from previous Battlefield launches to best optimize this game?

Well, I mean, like I said, we did learn a lot from 2042, which did not go great, right?
But we learned that we need to, maybe to some extent, the best lesson is that you can’t just fix everything at the end, right? It’s something you have to maintain the whole time. So for example, a while ago, we built levels that we called technical target levels, which is basically, we took a level, we built a level that was just full of as much stuff as we wanted to make sure got into the game.

It wasn’t built to be fun or anything. We just threw a bunch of stuff in the game. We want this many vehicles, we want this many soldiers, we want this many buildings and this much explosions and destructibles, and we just loaded a level. We built a couple of these levels, we just loaded them with that and then use that to make sure that we were on track for performance. In some cases we weren’t. In some cases the game was crashing or it wasn’t performant. We used that to kind of drive a lot of effort with our level designers and technical artists and engineers to optimize the build the game, the maps.

There was no ray-tracing in the beta, is there ray-tracing in the final game?

No, we are not going to have ray-tracing when the game launches and we don’t have any plans in the near future for it either. That was because we wanted to focus on performance. We wanted to make sure that all of our effort was focused on making the game as [optimized] as possible for the default settings and the default users. So, we just made the decision relatively early on that we just weren’t going to do ray-tracing and again, it was mostly so that we could focus on making sure it was performance for everyone else.

Talk to me a bit about Secure Boot. This was a thing I had never heard of prior to the beta, and ironically, I didn’t know about it either because my PC already has it. So it was one of those things where people are like, this is the worst thing ever, or something like that. And I’m like, I have no idea what you’re talking about. And I was just curious, why was that the optimal solution for this game?

To your point, a lot of people, especially if they have a new machine, it’s probably on by default and you didn’t know it, it just works. But there are a lot of people, if you have older machines or if you turned it off for some reason, there are a lot of people who don’t have it on.

For some people, it is a barrier to entry for the game and I wish that it didn’t have to be. I wish we didn’t have to turn on Secure Boot. But again, a couple of years ago, we had a lot of deep, deep discussions about anti-cheat and what we wanted our approach to be. We had discussions about how much do we want to focus on friction versus fairness? And the fact is, the more anti-cheat you do, the more friction you add, but hopefully the more fairness you add.

We made a pretty deliberate decision that we wanted to focus way more heavily on fairness. If it’s super easy to play the game, but you get in and everybody’s clearly cheating, that’s not fun. You’re not going to have fun and you’re probably going to stop playing the game. So, it’s unfortunate that there has to be some of that friction, but we did make that decision deliberately.

Secure Boot is a super powerful tool for us to help prevent cheating. It’s not foolproof. There is no foolproof anti-cheat solution. There are always going to be cheaters, but this does really help us in combating cheating and making the game fair for our players. And so that was the decision is we looked at our options. We said, this is going to have a huge impact on fairness, it’s going to add friction.

We’re going to do everything we can to put FAQs out there and help people walk through turning it on. The fact is there are some people who won’t be able to play because of it, and that sucks. I hate the cheaters for making us do this, but I think it’s going to be better for everyone else to have fewer cheaters and more fairness in the game.

Following some of the outcry during the beta, are you making any change to how it’s implemented?

Yeah, I mean, the short answer is we’re not doing a lot to change the implementation. We have taken some feedback on it was confusing. We’re trying to clarify some of that. We’re trying to give people more information, more tools to turn it on. So that is what we’re doing, but we’re not necessarily changing the implementation.

What else did you guys learn from the open beta, and is anything major being implemented based on any data and feedback you gathered on the technical side?

Well, I guess I’ll start by saying we ran the open beta, but we’ve been running Battlefield Labs since the beginning of the year, and that has been just hugely invaluable. We started putting it in front of players before we hit Alpha, so we knew we had crashes. The game wasn’t finished, sometimes we put levels in front of players that were clearly unfinished or had unfinished parts, but we really wanted to just start gathering feedback from players as soon as we can.

Both for the kind of technical things, we’re gathering performance data from players, whether you’re on PC or console, we’re gathering crashes and fixing them. We’ve been doing that from BF Labs all the way up through open beta. And so for open beta, obviously we had way more players than we had in any of our BF Labs events, and we were kind of running it on it for a longer period of time.

So we found some crashes that we hadn’t really found at BF Labs and great. So those crashes are top of our list to fix before we launch. So hopefully the game will be even more stable than it was in open beta. And then obviously we gathered a ton of gameplay and player feedback on different things.

I mean, if you played the beta, you probably saw that the shotgun was kind of overpowered, so that’s obviously something we’re going to fix. But also we realized that our tools that allow us to tweak that more quickly weren’t working the way we expected them to. So that’s something, it’s less important that we fix that one weapon and more important that we fix the system so that in the future, if a weapon comes out and it’s overpowered, we can quickly tune it rather than saying we weren’t able to fix it in the open beta.

But those are examples of places where we gathered feedback from players. We’ve taken a lot of learnings and lessons. We didn’t like some of the win rates on some of the maps where it was defenders having too high of a win rate. We’re working on balancing that. We got a lot of feedback on a ton of different things. We reread it. We see all that negative sentiment as well, and we’re taking all that feedback and listening to it. It doesn’t mean we’re going to do everything somebody wants, but we certainly hear all that feedback.

Does bringing the game to a console like Xbox Series S then help you when it comes to optimizing the game for more budget PC configurations? Is there any sort of comparison there?

I mean, certainly our mid-specs across the board and Series S has less memory than some of the other platforms. So optimizing for memory to make sure that the game can run on a Series S means that most PC players, who I think our PC min-spec has more memory than the Xbox Series S, means you should be in a pretty good state. You shouldn’t run out of memory and you should be able to play the game fine. So yeah, I think having the Xbox Series S as our mid spec helped us optimize the game, and it raises all boats, so to speak, as we optimize. I mean, sometimes it doesn’t, there were times when we had to do things specifically for one platform or another, but we did a lot of testing and a lot of validation.

I noticed that the PC settings have a balanced and performance mode, like a configuration, which is typically seen on consoles. What sparked that decision to bring a very simplified settings menu?

Well, I think that’s it. It’s a really easy way to tell players, ‘Hey, here’s two giant levers you could set. You could make it as performant as possible or you can make it look really good.’ And by the way, when you make it look really good, it still should be pretty performant, right? We’re still going to run it 60 frames per second, but if you want to maximize frame rate and reduce input latency and everything else, go the other way. I think it’s just a good idea, right? Yeah. We saw a lot of console games do it.

It makes sense and it’s just a good idea to put in there, but also we’re still giving you all of those other options. You can start with one of those settings or just go and pick and choose what you want as a PC player to customize and optimize where you care about the experience. But it’s also super confusing for people. It can be intimidating to open up like, ‘Oh, here’s 50 options. I don’t know what does what.’ So we just give you one big button, boop, make everything perform great. You can do that.

There are over 600 customization options in the PC version of Battlefield 6. Why so many? Obviously, it’s great to have customization, but I’m curious how much does all of that really change the game?

I guess it depends on who you are and what you care about. Obviously for most players, most of the settings aren’t going to matter, but there are players for whom each one of them matters. We actually, a couple years ago within EA started putting together what we call the PC standards list, like console and Xbox and PlayStation have their standards. There’s obviously not really a PC standards, but we started putting that together like, ‘Hey, we think this is what PC players expect. Here’s the level of customization they want. They want to be able to turn these things on or off. They want to be able to go to widescreen, they want to be able to add an Xbox controller or some other type of setting.

I think networking stuff for Battlefield has always been an issue at launch. Anything you’re able to share today about how you’re preparing for those things? I am assuming you’re aware of these things and preparing your best, especially after seeing how many people came to the beta.

I’ll start by saying the biggest problem we had in the beta was we had more people that we originally anticipate, which was a good problem to have. But I hope you’ll notice that the servers didn’t crash or anything. We had a log-in queue, that’s one of the things we put in place. We wanted to make sure that if more players joined than we thought we could handle, we would rather have a good experience for players that were in the game and put people in a queue rather than crash out our servers or systems. So we’ve put a lot of effort as well into optimizing our servers, optimizing our services.

We’ve made many adjustments to our projections thanks to the success of our open beta. All of the stuff I said about stability and performance also applies to server stability and server performance, so we’ve been putting a lot of effort into that as well. I mean a little bit of behind the scenes, we run a test where we basically simulate whatever our target is, whether it’s a million or four million.

We set a target for the number of players we want to have connected to our systems at a time or at launch and we run a simulation with simulated players. And guess what? The first time we run it, it doesn’t work, right? Some system falls over, something breaks. Great, but it broke then, right? It broke during our testing and then we fix it. We add machines, we optimize that system and we keep repeating that until we hit the numbers that we feel confident that we think we need to hit for launch. So, that’s the approach we’ve been taking to that.

So is the server queuing going to be part of the final game?

We’re always going to have that as a tool, but we are planning on not needing it for launch. We have projections and then we’re preparing for more than our projections to make sure that as much as possible, we don’t have to have server queues for launch. Now, just to caveat, we also use the server queue for rate limiting. So the moment we turn the game on, if a million people connect in the first minute, we’re going to have a queue just as we move people in.

But other than that, we’re not planning on [using queues]. We have a cap, but we’re setting the cap much higher than we are hoping we hit. I mean, it’d be great for me if we go over that cap as well. But yeah, we’re doing everything we can to avoid having to use the cap during launch, but we will always prioritize having a good experience for our players over just opening the flood gates and letting too many players in.