Microsoft Studios

In the shadow of greatness: How Certain Affinity co-developed Halo 4’s multiplayer and Forge modes

In the shadow of greatness: How Certain Affinity co-developed Halo 4’s multiplayer and Forge modes

Seven years ago, Max Hoberman was working with Bungie on creating the follow-up to their smash hit, Halo 2. He had moved across the country, from Boston to Chicago, Chicago to Seattle, Seattle to Austin, working on the game remotely. The move to Austin was the fulfillment of a promise to his wife, who was from the city, but it was also the perfect opportunity to split from Bungie and create something new. Thus, Certain Affinity was born. Hoberman is its current President.

Certain Affinity currently employs “about 75 full-time staff, plus a whole bunch of contractors” according to Hoberman, and that relatively small team has been behind some of the biggest multiplayer experiences of the past decade: Call of Duty: Black Ops, the Xbox 360 port of Left 4 Dead, and now, Halo 4. The company has been operating under a co-development model where despite considerable contributions, they receive little outside recognition.

They’re the Navy SEALs of game development; they get in, complete their work, and get out, likely to never see the spotlight they deserve.

The unknown soldier in the trenches

“Our involvement is usually brushed under the rug, so to speak,” Hoberman told the Penny Arcade Report. “And it’s, you know, it’s understandable. We’re working with these teams that are really sort of… if they’re not brand-new, then it’s their first time working on something big. So in the case of Treyarch – Activision was very frank, very up-front with us about it – they were trying to set them up alongside Infinity Ward at the time, and they really didn’t want us sort of stealing their thunder. They wanted them to get all the credit.”

Co-developing a game is a difficult task. How do you work on a game that has sold 25 million copies – the best-selling game of all time – and not want to stand up and take credit? Certain Affinity’s Production Lead, Phil Wattenbarger, told the Penny Arcade Report it’s all about where to place focus.

“We can’t always be in the spotlight,” Wattenbarger said. He said that the studio’s latest project, Forge mode and War Games in Halo 4, was rewarding in its own right. “Really, it’s the work we focus on. We feel like the credit comes – like right now, everybody’s seeing exciting reviews and that’s the reward. We were part of helping a great franchise.”

Wattenbarger and Hoberman both agree the company’s model as a co-developer is a relatively unique and uncommon one; most companies in Certain Affinity’s position are more separate from the main development studio. Work is handed off, completed, turned back in, and that’s the end for most partnerships, Wattenberger explained.

Take Deus Ex: Human Revolution for example. As Eidos admitted to Edge Magazine, time constraints forced the team to outsource boss battles to Grip Entertainment. Grip Entertainment worked independently from Eidos, remotely developing the boss fights. The results, while true to Eidos’ vision and instructions, felt disjointed and poorly designed. That’s not what Certain Affinity wants.

“When I say ‘co-development,’ what I really mean is it’s been an embedded co-development engagement. We’re not downstream from these guys, we’re really in the trenches with them,” Hoberman said. “We treat it as if we’re one team and we’re all in it together. Our involvement [with Halo 4] spanned an 18-month period, a year and a half, and I think we peaked at about 85 people on the project, so it’s just a huge engagement for us.”

Trial and error

Creating great multiplayer experiences is what Certain Affinity does best. Their pedigree speaks for itself, but if there’s any team that can explain how to make FPS multiplayer work, it’s these guys. I asked Hoberman and Wattenbarger how they know to direct players and the flow of a match.

“We always have theories starting out on what’s going to make a fun map and a fun flow through the map, but it’s not until you start testing it and playing it and actually observing empirical testing you really figure out how to make it great,” Wattenbarger said. He explained the company has four phases to anything they create.

“It starts out with the idea, and at the idea phase it’s basically… there’s paper maps but also visual concepts. And visual concepts won’t be very specific at all. It’s just like, ‘Oh wouldn’t it be cool if we did an asteroid field’ or something; just raw ideas to fuel the visuals. But on the paper maps design side, it’s like we’re trying to make a medium map with vehicles. The first phase is what we call the concept phase.”

Wattenbarger referred to the second phase as the “block out” phase. Blocking in this context could be seen as virtually synonymous with the theatre term of the same name. In that definition, blocking refers to the placement of actors on a stage; in a video game multiplayer map, the objects, objectives, and structures are not unlike actors, taking place in a defined area. At this stage however, the map looks almost nothing like the end product.

“I’ll use Longbow as an example,” Wattenbarger said. “We had a block out that had multiple bases and was lots of fun for vehicles and infantry, but did not visually look pretty at all. That was the block out of the map, and separately we had this idea generation and visual concepts that—‘Oh there’s this mass accelerator that’s gonna jettison these canisters up to a moon base.’ That was the big idea visually.”

“From there, slowly, these ideas between game play and visuals get integrated together. So the next phase is the second phase of block out, where we basically start trying to actually make it look pretty. First, we slowly rez in with the visuals to make sure we’re not breaking anything on the fun side, and then once we feel confident yes, we’ve made it fun, we’ve added all the visual key themes in there, we go to town and go to full-force art production. At the final phase we have a final art polish.”

Hoberman jumped in to elaborate. “By the time artists really dive in on the map, it’s 90% of the way there from a design standpoint,” he said. “Their job is to make it look pretty without breaking things that work really well that are fun.” Hoberman said that the team has so much experience now and knows their way around so well that they rarely make significant changes once a map goes beyond the concept phase. It still happens on occasion, though.

“There’s a map in [the upcoming War Games map pack] where the original concept… it was originally envisioned as a Slayer map. And we put it together and played it, and it was fun in Slayer, but at some point we tried out a game of CTF on it and it was like, ‘Holy shit, even though this wasn’t really designed as a CTF map, it’s actually more fun in CTF than Slayer,’ so we actually had to change our priorities. Our priority was Slayer first, then CTF and we inverted them along the way.” Changes didn’t stop there.

“And then, along the way, we also said ‘You know what? This map wasn’t designed for flying vehicles, but I bet it’d be really fun with flying vehicles!’ And again, it was like, ‘This works really well!’ Those discoveries happen. I wouldn’t say they happen that often, but one out of every four or five maps, you get some crazy, unexpected thing like that.”

Halo: full circle

Certain Affinity’s first project was the Blastacular Map Pack for Halo 2. They also created the Defiant Map Pack for Halo: Reach and have even worked with 343 Industries prior to Halo 4 with Halo: Combat Evolved Anniversary. This is a series they are intimately familiar with. It might be surprising then, that many of the changes and new modes for Halo 4 come from these series veterans.

Dominion, an evolution of Halo: Reach‘s Invasion mode, came from Certain Affinity: “How do you make a strategic game, how do you make a game that has depth, but is highly accessible?” Hoeberman asked me. The answer, turns out, is you do what gamers do best: play. You play and play and play, taking in as much information and observing player behavior as you can. Wattenbarger said you can’t get enough of the data, and watching playtests is the biggest help in figuring out what works.

“I think on the turrets, specifically, what was originally in there… there were turrets, and when the base would fortify, the turret would show up, and people would destroy your turret and it was as simple as that,” Hoberman said. He told the Report that the game was fun, but not action-packed enough. There wasn’t an interesting balance between risk and reward, since teams tended to bundle up inside their base and play defensively, not offensively.

“A lot of our philosophy is – and this goes for anything we do – Slayer and Deathmatch-style games are—you know, they tend to be the most popular, so any game that we build, killing should be a good thing. There shouldn’t be times in a game where it’s bad to kill your opponent. The way you make things accessible is if you’re running around shooting things, that should be a good thing to do. We start with that and layer on top.”

“So we kind of evolved it from there to this idea where the thinking was, ‘Okay, we don’t want people just camping the switch inside their base, we want people to come outside their base and sort of risk their neck, so how do get them to do that? How do we get them to risk their neck and come out of the base? Well you know what, let’s make it so they can respawn their turret and then eventually we took it even further and evolved it to where you have to actually go out to that turret to cause it to initially spawn.’”

Forging a path of ideas

Certain Affinity’s other major contribution to Halo 4 was Forge, the map creation mode where players can create their own battlefields to play privately or share across Xbox Live. It can be used to create all manner of custom games and insane map challenges. Since much of the groundwork was already laid out for Certain Affinity from Forge’s implementation in previous Halo games, Hoberman described Certain Affinity’s additions as “low-hanging fruit.”

The magnet feature is one such fruit. Before, lining up two items to be placed side-by-side was an almost surgical maneuver, requiring players to carefully push thumbsticks with slow and precise movements. Since items from Forge could clip through one another, it was up to the player to manually observe and check for any gaps between objects. Magnets automatically lock one object’s face to another’s, reducing the system’s complexity and speeding up map creation.

While magnets were a suggestion from the Halo fanbase, Hoberman explained that some features were more like happy accidents than anything planned or suggested. He brought up player trait zones, where Forge tinkerers can set variables like player speed, shield level, gravity, damage output, and more. “That was an idea that came about from one of our programmers; it was his idea,” Hoberman said.

“We were actually shutting down features and starting to get a lot more conservative, saying ‘Okay, we’ve got X amount of time before we have to close it all off,’ and it was kind of the last day when it was like, ‘Wait a minute, here’s this great discovery; what if we can get this in?’”

“I remember when he told me about it and I was like, ‘Oh my god, that’s freaking genius, why didn’t I think of that? We have to do that!’ Ideas come from all places. Whether it’s listening to the community, listening to our partners at 343, or listening to our own staff, a good idea’s a good idea regardless of where it comes from.”

Hoberman and Wattenbarger don’t want to take away from the work that 343 Industries has done with Halo 4, or the work that any of the developers they’ve worked with have done. But Hoberman is right: a good idea is a good idea regardless of where it comes from. If Certain Affinity – a small independent house working alongside giants of the industry – has a good idea, it’s only fair to recognize their hard work.