It took four years to find the FPS fun with Spectre Divide | interview

Nate Mitchell had a good run with a venture-funded game company before now. He was a cofounder of Oculus, which Mark Zuckerberg bought for a few billion dollars and then launched his Meta Quest virtual reality headsets.

After leaving, Mitchell got the urge to try it again at a time when there were more game-focused venture capital companies. He cofounded Mountaintop in 2020, raised $30 million and went to work with fellow creators like Lee Horn, one of the designers of Apex Legends. They were both 20-year fans of Counter-Strike and toyed with the idea of making a similar tactical shooter.

They built a team of 70 people, and enlisted the help of gaming influencer Shroud, an expert player who helped them with features that players would love. Their goal was to build a tactical shooter game that people would play for 10,000 hours. After four years of development, they came up with Spectre Divide, which recently held its closed beta test.

In Spectre Divide, Mountaintop created Duality, where you have two bodies controlled by one player. You can play as one character in a 3v3 match, and then place your second character in a strategic or secret spot by tossing a puck at the spot. Then you can instantly switch between characters. If one of the spectres is killed, you automatically switch to the other one. You get the benefit of one mind, two bodies. It’s an innovation that just might enable the game to grab its own audience of dedicated fans in the highly competitive FPS market.

I talked to Mitchell and Lee Horn, game director, after playing the game and the launch of the beta. The game is now launching as a free-to-play game on September 3. Here’s an edited transcript of our interview.

GamesBeat: How did the beta launch go?

Nate Mitchell: It went really well. It exceeded our expectations. It was great to see so many big streamers jump in and try the game. The level of excitement in the Discord community was top-notch. People are feeling good. The biggest thing we have to deal with now–everyone says, “I saw it yesterday and I want to play it today.” A lot of people are asking for keys. That’s a good problem to have.

GamesBeat: I wondered about the period of time where you were finding your game, or finding your fun. How did you get to some of the decisions about what kind of game to do? Three against three is an interesting number to wind up at. 

Lee Horn: Nate and I are 20-year CS:GO players, hardcore tac shooter players. A lot of people at the studio have been playing for a long time. We play a lot of shooters in general, but we were especially inspired by small team format shooters like Halo doubles, Destiny’s trials, things like that. We noticed the benefits of 3v3. I worked on Apex. At the time that was a big change, going to 3v3 from 4v4 in battle royale. A lot of that came from–it’s easier to find friends to play with. It’s easier to coordinate. Much better communication. You do more in the match.

That was our starting point. We started with a one life tactical shooter. There were two bomb sites, one life, three players. It was fun, but we could tell that the depth wasn’t there. That wasn’t a game you would play for 10,000 hours. We enjoyed playing it, but it didn’t set the hook in you. Part of that problem came from–imagine you have a defender on each site and one in the middle. Three attackers rush and they only have to kill one defender. The game is kind of solved at that point. They know there’s no one else there. They run in and plant the bomb and it’s easy.

The question was, how do we get enough density to create uncertainty? We tried some other mechanics. One night I had an a-ha moment. What if I could just be here in this corner, and also in this corner? If I could do that, that would solve the problem. The next day we tried a very simple prototype. That solved it. It worked. Then it was about a two-year journey of taking duality from that idea and molding it and weaving it through the entire game. We have pairs of weapons. We have areas where only spectres can go. We have ability kits with specific dualities. It’s about making that the core of the game, rather than just a piece of the game.

Mitchell: The first year and a half was just the iteration and prototyping. Lots of different ideas. Throwing everything at the wall, playing it, and seeing what worked. This ended up being one of those ideas after a year and a half of prototyping. Lee pitched this one and when we tried it, that was it. It was the thing we’d been looking for.

GamesBeat: The way the spectre is limited in terms of things like being able to climb, how do you get to that decision?

Horn: We played the spectre with a placement ability for a while, as well as the throwing ability. We found that the throwing part made it dynamic in ways that you could use. You could use it in clever ways that you would want to use in a tac shooter, things you normally can’t do. The best example, you want to cross an opening. Throw your puck across and you can get across safely, but they know you went across. They see the trail. That adds interesting gameplay and choices. 

Then it evolved into, how can we make map traversal interesting? A big part of that was creating spectre-only areas. Depending on the number of bodies alive on the other team, you can make decisions around how likely it is that someone might be up in that area. Will they take that path? Late in the match, when you’re playing against some opponents where it’s three people alive, but with only one body, it’s very unlikely they’ll be in the upper areas, unless they were defending there. What do you think? Maybe you don’t check those angles.

We took away things like team boosting to make sure it was really tight around information you could understand. You could look at the top HUD and see that there’s a pair alive. They can get up here. Versus, there are two players, now they can boost. Maybe they’ll be up there. You don’t really get any information. We wanted to play around that mechanic, so we adjusted it. That’s why you have the puck throw only with two bodies alive. In the future we’ll continue to add sponsors. Maybe one will break that and we’ll allow you to get up there somehow.

GamesBeat: How much play time does it take before people start properly making use of the spectres? I kept forgetting about it. At first I was just using it as my second life.

Horn: That’s why we’re super excited about the mechanic. At the baseline, say you forget about it. Your experience is exactly in line. It’s a second life. That just means you get to play more tac shooter. You learn faster. You’re in the action. We love that the floor has been raised for tac shooters. You just get to play more.

We see that it probably takes around three games. That’s where people start to think about. They’re not using it perfectly yet, but they remember to bring it up. They’ll put it in a more aggressive corner. Things like that. Over many games, you see that they learn to rotate with it. As they come up from another bomb site they’ll bring it with them. The key is that you need to understand the map. That naturally takes time. Once they understand the map, people anticipate where the enemy is. You can place your spectre based on that and make cool plays. Generally, though, it’s three games. Then it starts to become a part of the strategy. They remember to throw it. It’s part of the journey.

GamesBeat: How does it create new challenges? You have to have a map with a lot of hiding places to put the spectre. It’s not much fun when someone discovers your spectre and just shoots it.

Horn: If you look at our maps, there’s a fair number of nooks. We naturally nudge you that way. Hey, this would be a good place to throw it. If it’s in the backfield near your spawn point it’s probably protected from grenades. You’ll notice that when you spawn, you don’t just spawn in the open. Your spectre spawns in a little cubby. That makes it a bit more resistant to flanking and things like that, if you forget or if you just want to leave it there. It feels a little safer.

We’ve spent a lot of time trying to make sure each area you push in the map, the team asks, “Where would I stash my body?” We try to make something reasonably safe. As you get deeper into the map it gets less safe, because now you’re in the combat zone. But when you’re further back, we want it to be safer. You only have three sets of eyes. You can’t watch as many angles. It’s important that you have some safe spots there.

GamesBeat: I would guess that 5v5 doesn’t work as well. It’s far more likely you just find a hiding spectre that way.

Horn: Yeah, it’s chaos. It’s too dense. You have to do really big maps, a much bigger scale. Then you end up in situations where, once it gets down to just a few people, you’re dealing with great distances. This is the magic density of players. We’ve crafted maps around players being effectively 6v6. Everything is custom built, and it needs to be custom built. We spent a lot of time finding all the little nuances.

We have reveal walls in the game. That’s a big thing related to having 3v3. If the defenders go through them at the start, it’ll reveal them, like a map hack effect for the first 40 seconds of the game. That’s to stop them from flanking. With only three people you don’t have enough eyeballs to watch for flanks at all times. We want attackers to be able to push up a little bit more safely, execute their plan, and then worry about flanks. It’s things like that. You need to craft all the little pieces to make it work.

GamesBeat: How do you try to keep people happy as far as–some people might just see this as one mode, and they want 10 more modes to play.

Horn: We’re trying to make a 10,000-hour game, basically in one mode. We have ranked and we have unranked, but that’s just a question of whether you want to track your progress or not, rather than a mode that plays differently. Some of that is focusing on systems that have a lot of depth. A map with a lot of nuance, where it depends on–did they do A? Well, then I can do B, but if I know they do that I can do C, but if they know that they can do D. That level of thought, depth, map knowledge. It really goes deep.

We tried to raise the floor, as I said. You have that second life. But if you really want to go deep and coordinate your spectres, coordinate with your team, coordinate your abilities and your weapon pairs, all those things, we think that there is a game that you could spend 10,000 hours in and keep it interesting. We’ve been playing it for two and a half years, many times a week. We playtest regularly every week. The design team playtests five times a week. Five times a week over multiple years, and I still spend all night playing. The depth is there. The excitement is there.

We don’t want to go broad. Will we experiment with fun modes, alternative modes post-launch? Sure. But we’re focused on making that core mode the best. We want people to know that they can throw their whole gaming time into it and see that pay off.

GamesBeat: Do you have players who are interested in cross-play and controller support yet?

Horn: We’ve been developing with controllers since day one. That’s important, to make good controller gameplay. We made a decision to go with ADS gunplay, too, which is good for controllers. It’s almost there. We have some bugs to fix, but we’ve been playing with controllers for a while. We have a contingent of people on the team that only play with controllers. They die a lot to give us that feedback. We’re focusing on a smooth PC launch, but we want to move to console. We want to have controller support. Those are near term goals. The game plays well that way. We think the audience will enjoy it. There aren’t many tac shooter options on console, especially ones that favor ADS gunplay.

GamesBeat: Were there things about, say, Counter-Strike where you thought, “Everyone loves that, so we should keep that in”? Versus areas where you felt like you had to make this game different.

Horn: Probably the mode: two bomb sites, attacker and defender, round-based. That’s the core of tac shooter gameplay that we didn’t want to disrupt. You earn money each round. That kind of feels like the power curve. If you lose, you go down the power curve. You feel that balance. That’s the inspiration in the genre that we didn’t want to touch very much.

We did make a change in that we went down to seven rounds. We wanted to be faster. You’re not committing to a 90-minute play session. It’s more like a 30-minute play session. That’s nice these days. But otherwise we stuck to that core.

GamesBeat: Is there a time limit in the rounds?

Horn: We play seven rounds a side. First to eight rounds wins. The rounds are two minutes. They very rarely go for the full two minutes. Maybe one round in 15 goes to the very end. But it’s a fixed time. The average match time is around 30 minutes. It helps with matchmaking, because the pool is refreshed much more frequently. We hear a lot from our players that it helps them play more socially. If your buddy longs on you can say, “Hey, I’m in a match, but give me 20 minutes.” They might wait that long for you. If they have to wait an hour and 20 minutes, they’re not waiting. We see a lot of benefits. If you’re having a bad match you get out and you get to play with someone else faster. If you have a good win, you get to get in and go for it again. The community feedback has been very positive on that change.

GamesBeat: The appeal to the Counter-Strike crowd is obvious. Do you find that it appeals to players of different games as well, like Apex or Call of Duty?

Horn: The ADS gunplay is a massive change for the genre. If you liked tac shooters before Spectre came out, you always have to stop and shoot. It’s a very specific type of shooting model. Some people love it. Some people just love tac shooters and they’ve never really gotten into it. They hate stopping and shooting. Our big appeal to those audiences is ADS. Our weapon designer is from Apex. We worked together on Apex. He designed a lot of the Apex guns. If you like the Apex gunplay, you’ll like what we have to offer, because the same person built those things. Same with Call of Duty. That’s a big ADS community.

The other one is that it’s true to cluster accuracy. You see the dot and that’s where your bullets go. In traditional shooters the bullets don’t follow the cluster that much. They go above when you’re on full spray, which can be confusing. Some people just don’t get it. They don’t understand why. That’s the biggest change for those people. Again, the second life also just lowers the stakes a little bit when you’re learning. Learning some other tac shooters, you might run in and die in the first three seconds. Then you’re just waiting for a minute and a half. You’re not learning and getting attached to the game in that time. In Spectre you’re back in and getting another shot. We think that will also smooth the onboarding process.

Mitchell: Yesterday, if you look at the Twitch charts, a lot of the top streamers in the category–yes, we had Counter-Strike players and Valorant players, but Destiny 2 players made up a huge chunk. Overwatch players made up a huge chunk. The appeal to non-CS and other players is just as strong, if not even stronger. “There’s finally a tac shooter that makes sense to me.” That’s one of the things ADS really helps with.

GamesBeat: On a company level it’s interesting that you were venture-funded and you’ve gotten to this point. I haven’t written too many stories about venture-funded companies getting through to a launch. Was that a different experience for you, as far as how this progressed compared to traditional funding from a publisher?

Mitchell: I don’t think venture backing changed anything about our journey. One thing that’s great about our partners–two of our biggest investors, for example, Anthos and Andreessen, they are just insanely supportive every step of the way. They function as an awesome extension of our team in a support manner. They just push us all the time to make the absolute best game for players. They’ve never asked us to compromise. I do think that’s one advantage relative to traditional publishers.

Some publishers are better than others. Every publisher is different. That’s not a shot at publishers. But I do think there are a lot of special parts of Spectre that have required a lot of trust and iteration and craft and risk-taking. Some publishers would likely be uncomfortable with that. When you’re working with venture partners, they’re excited about it. That’s a great dynamic as far as creating a studio. It’s a great environment for very creative and bold games, which is what we wanted to achieve with Mountain top.

Horn: We landed on a very unique mechanic that’s in no other game in this way. You can’t compare something to it. You can’t really understand it that way. Being free to make these choices, having fewer stakeholders to manage, we were able to do it. It would be much harder when you’re trying to make everyone comfortable with a risky idea. A lot of that comes from–you have to be deep into the audience. You have to live tac shooters to understand why this on-paper crazy mechanic would work. When you get really in there, this is solving real problems. It’s improving real things that the audience cares about. Then you go on that journey.

GamesBeat: It feels like the whole mission of venture-backed game companies is to do the original ideas, the original IP that so many publishers are not doing anymore.

Mitchell: It’s a great thing. Obviously we’re biased, but we think it’s good for the industry.

GamesBeat: Was there influence you drew from Valorant as well, as far as seeing something new that could start up and get traction?

Horn: They’re a core tac shooter. They added ability gameplay to the genre and that did great things. I actually worked on Valorant in the really early days. It was cool to see that pitch at the time. But like I said, we’re inspired by all shooters, taking little bits and pieces and trying to improve.

GamesBeat: Are there any other important things you’ve learned from player responses?

Horn: You need top-tier players in the development process. I encourage every studio to seek that out. We partnered with Shroud to do that. We brought in a really high-level pro player early on, about two years ago now. That helped us identify balance changes, movement feel, weapon feel, recoil, all these things at such a high level that’s really necessary to come out with a product that’s ready to go. The alternative is to just say, “Hey, we tried, now the community can tell us.” That was integral.

We also have a smaller playtest group where we bring in players regularly, really high-skill players, to check out new features. They’re already playing sponsors that we’ll release in the future and letting us know if they’re fun to play, if they fit the right role. That’s something studios should always do. Make the community an active part of your development. They see a lot of things that you don’t see. They’re not in it every day. They give a lot of great perspectives. The diversity of skill levels is important. They might have different settings that you never considered. “Hey, they like to do X. We need to have settings for that. It’s important to this group of people.”

GamesBeat: Do you foresee needing to speed up development to some level of live service, where people are going to start asking about what’s next very quickly?

Horn: We’re going to be a free-to-play live service. We plan on updating the game regularly. We’re already working on future seasons and updates. We’ll iterate and change features and things like that. The magic of live service is that if you have a hungry player base, they always want more than you can do. It’s always about trying to deliver, set expectations, give them cool stuff, and continue to evolve the game. But that’s very much the plan. We’re committed. Spectre is ideally a 10-plus-year game that we’ll continue to grow and evolve over that time.

GamesBeat: Do you have a road map for when the next stage of the beta is coming?

Horn: The beta is through the 21st. After that we’re looking at–if everything goes well, we’ll try to roll into full launch sometime soon after that. This is a kind of hybrid for us. Players can stream now. They can fully see everything about the game. We’re inviting a lot of people. If you wishlist or you watch on Twitch, there’s a pretty good chance you can get in and play.

GamesBeat: How big is your team now?

Horn: We’re up to 80. 

GamesBeat: Is there any advice you’d give about having the right scope for a project, so that a smaller team can get something like this done? How do you figure out, if you’re in this group of venture-funded game companies–maybe they have their first round of funding. What can they do to contain their scope so they can get to the next stages?

Horn: Focusing on a genre you know and love and care about is step one. Stick to the core, or maybe 80%. Something like we did with two bomb sites, attacker and defender. That works. That’s exciting. Then put your effort into that 20% unique evolution. When you can get that right, it’ll feel more like 80%. But the mindset–focus on a few major needle-movers that really change the game versus trying to make every aspect of the game completely unique. That’s a massive undertaking that requires a massive team. It’s not even what players necessarily want. It’s not necessarily the best game. Stay focused on what matters, put effort into that, and then thread it through the game. That’s important.

GamesBeat: Did you ever worry that the shooter genre was too crowded for a startup?

Horn: Not at all, really. That might sound crazy. But the products I’ve worked on before faced a similar question, and it’s never come to fruition. When I was working on Valorant, CS was the dominant tac shooter. Valorant came out and grew the market. CS hit new peaks. When I was on Apex, PUBG seemed to have it, and then Fortnite came out of nowhere. With those two in control of the market, was there really room for another one? And then Apex and Warzone both came out. There was room for two more.

The key, if you look back at those, they’re all solving a different need. That’s why we’re excited about 3v3, duality, and ADS gunplay. That’s not solved anywhere, especially the ADS. There’s a tac shooter audience, a very large one, that has nothing. We’ll be something for them. Then there are all the tac shooters who want some new challenges. “I want to stop and take cover. I want to cover my own flank. I can’t do that. Where can I do that?” You can do that in Spectre. We’re excited about providing new opportunities for these players.