clock menu more-arrow no yes mobile

Filed under:

The Destruction (and Reconstruction) of Destructible Environments in Video Games

Almost 15 years ago, ‘Red Faction: Guerrilla’ and ‘Battlefield: Bad Company 2’ ushered in a short-lived heyday of digital destruction. Can recent arrivals like ‘The Finals,’ ‘Teardown,’ and ‘BattleBit Remastered’ bring it back?

THQ/Embark Studios/EA/Tuxedo Labs/Ringer illustration

In the summer of 1998, John Slagel received an assignment: Give gamers unprecedented power to blow stuff up.

Slagel was a programmer at Volition, a Champaign, Illinois-based video game developer known for the Descent series of fully 3D, flight-based, first-person shooters, which had started at Volition’s precursor studio, Parallax Software. When Descent 4 was canceled, some of the work Volition had put into the sequel was repurposed toward an original shooter, Red Faction. The new game, which was slated for Windows and PlayStation 2, would feature a rebellion of downtrodden miners on Mars. It would also aim to deliver a landmark technology: destructible environments.

“Some games did do the idea of, you could destroy a chair, or a barrel, or crate,” lead designer Alan Lawrance says. “And that was fun. We just wanted to take that beyond that and actually destroy the walls, or the ground, or bring down a structure.” The problem was, no one knew how to implement physics-based destruction on that scale, a much more complex undertaking than the more primitive environmental mayhem in 2D games like Dig Dug (1982), Rampage (1986), or Worms (1995). “When we came up with the desire to make a game with destruction in it, we looked to see what was out there, and there really wasn’t anything,” Lawrance says. “And we decided to take that challenge on.”

At first, we was me. “It fell on me,” Slagel, Red Faction’s lead programmer, recalls of his labors, which began before Descent 4 morphed into Red Faction. “Basically, I had a summer where they said, ‘Get this done.’”

When he was a teen in the mid-1980s, Slagel had designed a game on his IBM PCjr in which the player used the space bar to drop bombs from an ASCII art plane onto targets on the ground, creating craters. It was, in essence, a proto-destructible environment, but it wasn’t the kind he needed now. Slagel sought out reference books and papers that might give him guidance. He found a 1989 book called Geometric and Solid Modeling, published by Christoph Hoffmann, a computer science professor at nearby Purdue University. Hoffmann, who years later would harness his simulation expertise to determine why the World Trade Center towers collapsed, was hired as a consultant, and Slagel visited him to lay out Volition’s goal. “I was like, ‘Yeah, we’re going to do this in real time and blast holes in the world,’” Slagel says. “He’s like, ‘I don’t think you can do that in real time.’ I said, ‘Well, by the end of the summer, we’d better.’”

Sometimes a deadline does wonders. The system Slagel designed, dubbed “Geo-Mod” (short for “Geometry Modification”), worked well enough to wow reviewers when Red Faction debuted in May 2001. “Believe the hype: It is absolutely fantastic,” wrote Games Domain. “You can pretty much destroy anything you see.” GameSpot praised the game’s “moments when you say to yourself, ‘I wish I could do this,’ only to find out that, yes, you can.” Eurogamer reported, “Everything you have ever wanted to do in a first-person shooter is now possible.” IGN raved that Red Faction’s “total destruction” was “the ultimate in video game military fantasy.” Strong scores translated to strong sales, and Red Faction qualified for Sony’s PlayStation 2 Greatest Hits program. Much of Red Faction felt familiar, a mixture of Total Recall and Half-Life. But Geo-Mod “really helped the game become popular, because it was something that was so new,” Lawrance says. “People hadn’t seen that before, and it was really a cool thing.”

Red Faction’s success spawned three sequels over the next 10 years. It also helped inspire imitators. “Around that time, there were other games that came out that had some destruction in them as well,” Lawrance says. “It seemed like it was something that might become more commonplace in games.” Destructible environments, for a while, were all the rage, playing prominent roles in releases such as Silent Storm (2003) and The Incredible Hulk: Ultimate Destruction (2005); Destroy All Humans! (2005) and Destroy All Humans! 2 (2006); Earth Defense Force 2017 (2006) and Stranglehold (2007); Mercenaries: Playground of Destruction (2005) and Mercenaries 2: World in Flames (2008); Crysis (2007) and Far Cry 2 (2008); Fracture (2008) and Just Cause 2 (2010); Black (2006) and Breach (2011); and Battlefield: Bad Company (2008) and Battlefield: Bad Company 2 (2010).

And then, after roughly a decade, the heyday of destruction was over, as suddenly as it had started. “It definitely seemed like there was a very long time where destruction really was not something that games were trying to do,” Lawrance says. Although there are elements of destruction in powerhouses such as Minecraft, Roblox, Fortnite, and No Man’s Sky, few games have made destructible environments a core mechanic since the early 2010s. Some that tried to build destruction-based buzz, such as The Order: 1886 (2015) and Crackdown 3 (2019), overpromised and underdelivered. And franchises that were once seen as standard bearers for destructible environments, such as Battlefield and Just Cause, are now widely perceived to have dialed down their destruction.

Gamers have noticed, and largely lamented, the destruction decline. The last year or so alone has summoned well-trafficked Reddit threads titled, “What happened to destructible environments?”, “wtf happend [sic] to destructible environments!?!?”, “It’s a bit weird how environmental destruction came and went,” “games with fully destructible environments, why are they so rare?”, and “Why so few (FPS) titles have any degree of destructible environment,” as well as articles whose headlines asked, “Whatever Happened to Destructible Environments?” and “Why Did Video Games Leave Destructible Environments Behind?”

Lately, though, there have been some signs that destructible environments may be back on the upswing following that long lull. Teardown, a fully destructible heist game that entered early access in 2020, was released in full on Windows in April 2022 and sold more than a million copies by that August, when developer Tuxedo Labs was acquired by Embracer Group subsidiary Saber Interactive. Last November, Saber published well-received ports of Teardown for PlayStation and Xbox. Last June, an independently developed online first-person shooter called BattleBit Remastered, which boasts throwback, Bad Company–esque destruction, was launched in early access and became, per PC Gamer, an “overnight success,” selling almost two million copies in two weeks. And in December, Embark Studios’ free-to-play first-person shooter The Finals, a destruction-fest fresh off a buzzy open beta, launched with a lot of momentum and attracted 10 million players in its first 15 days.

“It’s always been a fun mechanic,” Mikael Högström, animation director for The Finals, says about destructible environments. “It’s something that players appreciate when it’s done right.” And perhaps they’re hungry for it, given that, as Högström adds, “not many other games are touching that space right now.”

Right now, no. But soon? We’ll see. Maybe blowing stuff up is about to blow up again. “People that green-light games, they’re a little bit risk-averse these days with the price of creating games, especially if you’re talking triple-A games,” Lawrance says. “And they want to see proof that this concept actually is going to have traction in the market. And so maybe the fact that we are seeing some successful destruction-based games, that will lead to more of them.”

Where did destructible environments go?

Part of the appeal of playing games is freedom from constraints: From double jumping to casting spells to, well, fighting mercenaries on Mars, you can do things in virtual worlds that aren’t possible (or safe) in the physical one. Yet gamers get used to certain restrictions that would seem onerous and strange in real life. Doors that don’t open. Impassable, indestructible shrubbery. NPCs that repeat the same stock phrases over and over. We accept (or used to accept) these things because we make concessions for the limitations of time and technology. When a game comes along that transcends a traditional barrier, though, the experience of playing it is magical and memorable. You remember where you were.

More realistic doesn’t automatically mean more fun, but over time, games have tended toward greater verisimilitude—and, by extension, greater interactivity. Take the evolution of shooters. After Star Wars: Dark Forces (1995) improved on Doom’s control scheme by permitting players to crouch, jump, and aim up and down, it was inconceivable that subsequent shooters would restrict players to standing and looking side to side. After WinBack: Covert Operations (1999) introduced a modern cover mechanic, anything less seemed half-baked. After Alien Resurrection (2000) defaulted to dual analog sticks, that FPS control scheme soon became standard on consoles (though not everyone took to it instantly). So it was, for a while, with Red Faction and destructible environments: The ability to impose your will on the world by reshaping it with weapons was so literally transformative that it seemed like it could become table stakes for future big-budget games, especially as computers and consoles leveled up.

That didn’t happen, for two reasons: technical hurdles and design hurdles, both of which applied (and to some extent, still apply) to any game that goes all in on destructible environments.

Here’s how Lawrance explains the Geo-Mod system in layperson’s terms: “You fire a rocket at the wall, there’s an explosion. We take a low-[polygon] shape that mimics the explosion, and that shape gets subtracted from the world. So you intersect that shape with the world, whether it’s a floor or a wall. And you, on the fly, modify that geometry and leave a hole there.”

That sounds simple enough, but in practice, it was pretty torturous, thanks in part to turn-of-the-century hardware. “We did a lot of things the right theoretical way, and then they would break, so then you had to start fudging it,” Slagel says. Early builds of the game constantly crashed because of “weird little tiny triangles that would go all the way through the world.” Sometimes, the seams would show at T-joints where two materials connected, creating holes where objects could pass through supposed-to-be-solid surfaces. Attention to detail was crucial. For instance, if the player shot a rocket at a wall, it would leave a scorch mark. If the wall was then destroyed, Slagel had to ensure that the scorch mark would disappear also instead of hanging in the air where the wall used to be. “While the game was developing, I was constantly fixing these things,” he says.

Making it all run reasonably well required certain sacrifices. In demos, it was possible to, say, redirect a stream of particles that looked like lava by blasting a hole in the chute that the lava was flowing through. With full levels loaded, though, this effect caused the frame rate to tank on lower-end PCs. Similarly, Slagel planned for the lighting on some surfaces to adapt to the destruction around it. That also slowed the game to a crawl, so those bells and whistles were removed. Early plans for an additive aspect to Geo-Mod, whereby players could use a “goop gun” to create lasting structures, were also axed.

The developers also “cheated” in subtle ways to improve performance. If Red Faction had provided players with unlimited ammunition, they could have caused an unlimited amount of destruction, which would have created an unlimited number of new polygons in the world and potentially exhausted the system’s memory. The workaround: “We just limited the amount of rockets or grenades that you could obtain, and that would put a ceiling on the amount of destruction that could have been done,” Lawrance says.

Limiting ammo is a common convention in shooters, so this solution didn’t stand out as shifty. Even less noticeable was another technique for conserving resources. If a player exceeded a certain amount of destruction in a large level, the game engine would “undo the destruction that was done at the start of the level,” which would free up memory for fresh destruction at the end of the level. Unless they backtracked, players would never know their messes had been miraculously cleaned up. “We never really even heard anyone complain about that,” Lawrance says. “So I guess that trick worked too.”

Making the game stable and smooth despite the destruction was one thing. Making it play well was another, because, Lawrance says, destructibility “affects almost everything else in the game, like level design and AI.” Computer-controlled characters typically rely on “pre-computed areas on the floor to know where [they] can walk. And if you can blow a hole in a wall or put a hole in the ground, you have to dynamically update what’s called a navmesh, which determines where the AI can move.”

At least AI characters go only where the computer tells them to. Players are less likely to adhere to developers’ wishes. And while Red Faction’s devs took technical shortcuts to make their code compatible with destruction, players could create their own shortcuts to exploit destruction. As Lawrance explains, the goals of laying out a linear path through the game, and allowing the player to vaporize large parts of those linear levels, often came into conflict. If “everything’s destructible,” he asks, “how can we create a scripted sequence that’s supposed to play out A, to B, to C when the player can blow a hole through a wall and go to C first?” (Teardown director, designer, and programmer Dennis Gustafsson, who confronted the same quandary decades later, expresses it this way: “If you’re given the freedom to remove all obstacles, then it’s really easy to end up where you don’t have any challenge.”)

The original Red Faction’s developers dealt with this conundrum by making metal doors or walls that the player couldn’t destroy, which allowed Volition to funnel players in the right direction. “We were never totally happy with that as a solution, but it seemed like that was the only thing we could do in the time we had,” Lawrance says. The third game in the series, Red Faction: Guerrilla, steered clear of indestructible objects but built in a provision for restoring mission-critical structures if a player had already destroyed them. “If you’re not trying to tell a narrative, it doesn’t really matter if people have laid waste to whatever,” says Matt Gawalek, a gameplay programmer on RF:G. “If you need things for a single-player, narrative-based thing, it does get really hard to design around all the permutations. So that was our little copout, was that we would bring certain things back.”

By the time Guerrilla came out for PS3 and Xbox 360 in 2009, technology had advanced, enabling a new kind of destruction—Geo-Mod 2.0. “It was more of a structure destruction game where whole buildings could come down and everything would break apart into its individual pieces,” Lawrance says, adding, “That was the next level of destruction that we couldn’t have pulled off in the Red Faction days because the cost of doing that kind of physics on that many objects was just prohibitive.”

It wasn’t easy in 2009, either. Geo-Mod 1.0 had enabled cutting into terrain, but as Gawalek says, “nothing really had weight to it.” Geo-Mod 2.0 incorporated a real-time stress system that, unlike a lot of other games with destructible buildings, could assess whether a structure should be able to stay standing. Whereas other games might not cause a watchtower to topple if three of its four support struts were destroyed (because one was still connected to the ground), Guerrilla “did perform this calculation, and so if you did take out the left half of the supports on the building, it would eventually buckle and fall, like you fell a tree.”

This was both incredibly cool for players and incredibly complex for creators, who had to relearn their craft. The environment artists’ first buildings “fell down in the engine, because they were used to just making static models [that didn’t] have to be structurally sound,” Gawalek says. “So now they [had] to go and actually put proper steel beams [in] and make a proper foundation.” And because building models were broken up into predefined chunks so they could splinter and crack more convincingly, those surfaces required extra textures, and more memory.

Even with more RAM and processing power than PS2s and PCs had provided eight years earlier, Volition had to make the difficult decision to confine Guerrilla’s destructibility to man-made surface structures. If players could have cut the terrain under structures, the physics would’ve stressed CPUs even more. Even so, Gawalek says, “Our textures were lower quality than comparable titles of that era, just to save some memory.” The game’s user interface cleverly employed a gritty aesthetic that allowed Volition to reuse the same few images. Even so, the game lacked the memory to load the world map, so Volition performed some sleight of hand to unload a vehicle from the game when the map needed to open—which freed up some precious megabytes—then repopulate the vehicle when the map closed. “We were literally stealing memory from other systems just to do basic functionality,” Gawalek says.

The Geo-Mod system carved out a niche in the gaming terrain for Red Faction, but the franchise failed to outlive the trend it kicked off. Fans of destructible environments still wax nostalgic about Guerrilla, which was remastered in 2018, but the 2011 follow-up, Armageddon—which added the ability to reverse destruction—wasn’t as well received. “If you looked at a graph of the sales of all the Red Faction games, you would see a downward slope,” Lawrance says. The original Red Faction was a moneymaker, but Armageddon incurred a “substantial loss.” Publisher THQ, which had acquired Volition in 2000, declared bankruptcy in 2012, and after multiple changes of ownership, Volition shut down last August. The dormant Red Faction IP now belongs to Plaion, a publisher owned by struggling Swedish holding company Embracer Group.

Slagel speculates that “focusing so much on the Geo-Mod,” as opposed to refining other aspects of Red Faction, “might’ve actually set our first-person games back a bit.” But it was fun while it lasted—and for some fans of Red Faction, it’s lasted for almost 25 years. Synchronizing destruction in online multiplayer modes presented technical challenges in 2001, but the unpredictable gameplay that resulted set Red Faction multiplayer apart from most other games of its era—or, for that matter, this era. “There’s still players today enjoying it,” Lawrance says. “And it’s really the destruction part of it that does that.”

With all due respect to Red Faction, there’s one multiplayer game that epitomizes the period of peak destructibility: 2010’s Battlefield: Bad Company 2.

Alan Kertz, a former developer for Electronic Arts–owned studio DICE, worked on every Battlefield game from 2006’s Battlefield 2142 through 2018’s Battlefield V. Battlefield 2142 featured very little destructibility, but the next installment in the series, 2008’s Bad Company, was the first game to use DICE’s Frostbite engine, which was designed with destructibility in mind.

“One of the big things was how do you differentiate yourself in, at that time, a pretty crowded first-person shooter space,” Kertz recalls. “There were a few games doing destruction, but nobody was doing it on the multiplayer front, and definitely not on the scale in the multiplayer that we really wanted to do.” Battlefield was known for dynamic showdowns on massive maps, so destruction suited the brand. “When you have vehicles, it’s quite unsatisfying to drive your 60-ton tank into a tree and the tree just stands there and says no,” Kertz says. Battlefield powered by Frostbite was “going to be the game where we have all the tanks and the planes and the jets, and if that sniper over there is annoying you, just bring the whole building down on top of [them].”

In the Bad Company games, destruction wasn’t a gimmick or purely an approach to product differentiation. It was an animating force for the creators. “The artists had all bought in on the idea that destruction was the thing that was going to take them to the next level,” Kertz says, continuing, “It wasn’t just a design or a producer idea [that] we should have destruction, but very much [an] across-discipline drive that … turn[ed] that into something that was at the core of the game more than just an element that showed up in the back of the box.”

Kertz recalls the moment when the gameplay potential clicked. The Bad Company team had created an internal tech demo intended to prove to EA that its investment in Frostbite and the studio was worthwhile. The demo featured an anti-aircraft gun set up in a clearing and surrounded by buildings, trees, and assorted set dressing. “You could just sit with this gun and just wipe the place clean,” Kertz says. “It was like you went from this pristine state to, everything’s gone.”

But not entirely gone, because Bad Company’s developers had realized that the key to truly dynamic matches was not just the capacity to demolish the map, but how engaging a playground the ruins made. “There’s got to be craters, there’s got to be negative space, there’s got to be rubble leftovers,” Kertz says. “There’s still places for players to hide.”


DICE doubled down on that deadly hide-and-seek in Bad Company 2, which was built with an upgraded, 1.5 version of Frostbite featuring “Destruction 2.0.” In the first Bad Company, buildings could be blown out but not fully brought down; to preserve spacing and pacing, internal walls would stay standing. In the sequel, players could destroy the structural integrity of buildings, which would trigger their collapse. Not only did this look cool, but it opened up fun gameplay possibilities: If an enemy was hiding in a building, a player could simply level it. Not everything could be atomized: Iron girders, reinforced concrete, and elevated terrain would endure, which was helpful for obscuring the borders of the maps. But even indestructible objects would sustain superficial damage, because the watchword at DICE was, “If the player shoots at it, something should happen.”

Initially, Kertz says, “We were super scared of, ‘OK, there’s going to be nothing left.’” But in the thick of the action, that rarely became a concern, partly because collapses created new cover, and partly because players proved to be so creative in deploying the destructive tools at their disposal. As Kertz says, “You open up spaces where, ‘Hey, that guy just blew a hole in the ground with a bunch of C4, went over there and hid in it, and then jumped up and ambushed me as I walked by. Oh, didn’t see that one coming.’”

Guerrilla supported 16 players in online multiplayer; Bad Company 2 supported 32. Like Volition, DICE had to pull out all the stops to get its game to run. “A lot of games are a lot about receiving data, but when you’re impacting the world, you’re sending that information to the server and then to other clients,” Kertz says. “So our bandwidth requirements were substantially higher than other shooters.”

Another source of strain was the lack of occlusion. In most games, pieces of geometry block (or occlude) other pieces of geometry. If a player is facing a wall that can’t be destroyed, the engine doesn’t have to render whatever’s behind the wall, because it can’t be seen. If the wall can be destroyed, in full or in part, then the game has to calculate and render whatever would be visible. To keep Bad Company playable, Kertz says, “There was a lot of optimization done to the way that we sent data. So, for example, building a bubble around a player and saying, ‘Everything within 25 meters is super important, send that data.’ And then looking at what direction the player was looking in and saying, ‘OK, everything in that area is more important than the stuff that’s behind you.’”

The immediate follow-ups to Bad Company 2 put their own spins on destruction, though their gameplay arguably wasn’t predicated to the same extent on organically laying waste to levels. Battlefield 3 (2011) ran on Frostbite 2, which featured Destruction 3.0, an update that made players susceptible to lethal damage from falling debris. Battlefield 4 (2013) debuted Destruction 4.0, known as “Levolution.” Levolution enabled large-scale changes to levels that were activated by certain predetermined actions players could take. For example, on Battlefield 4’s “Rogue Transmission” map, based on Puerto Rico’s Arecibo Observatory, players could snap support cables to destroy the large dish—which later happened to the real observatory, in the same year that the game was set.

Destruction is still part of the Battlefield series (which has had its ups and downs), though the most recent entry, 2021’s Battlefield 2042, doesn’t come close to rivaling Bad Company 2 in the destructibility department. Kertz, who now works for Raw Fury, an indie publisher cofounded by a former DICE producer, acknowledges that “There was a different focus from a product perspective” in later Battlefield releases, though he thinks the demise of destruction in the series has been somewhat exaggerated. The perception that Battlefield is less committed to destruction, he says, stems partly from the fact that the novelty wore off; it no longer stood out that a tank could roll right over a tree. “I also don’t think it was marketed in the same way, because it was like, ‘Yeah, of course Battlefield has destruction,’” he adds.

If destruction does play a less integral role, he says, it’s partly because of an emphasis on set pieces or the demands of different settings, such as the pre-destroyed World War I environments in Battlefield 1. Compared to the rural, underdeveloped European countrysides of the Bad Company games, urban environments tend to be flatter and more dependent on indestructible big buildings to delineate players’ paths through the level. “If the only building on the level is a two-story farmhouse and you can take it down, [then] logically, as a player, the perception is, ‘Well, if the only building on a level is a skyscraper, I should be able to take it down in the same way,’” Kertz says. “Someday, I think we’ll probably get there.”

It wasn’t a complete coincidence that the first wave of 3D games with destructible environments crested roughly 15 years ago. Nor was that fad—if in fact it was only a fad—solely a copycat response to Red Faction. “The PS3 and the 360 were the first … consoles where there was strong parity between the ability of a gaming PC and the ability of a gaming console to hit the same memory and tech abilities,” Kertz says. “So there was definitely a, ‘Hey, the tech is there.’” Högström, who also worked on the Bad Company games (and subsequent Battlefield titles) before becoming Embark’s animation director, adds that a technological leap “gives you high hopes that you can do a lot more stuff, and then maybe the big grand ideas come at the same time.”

Video game hardware has taken multiple leaps since then, which makes the creators of the OG Red Faction, such as Slagel and Lawrance, somewhat wistful. “The memory is so high on these machines, and the CPUs are so powerful, that destruction can be done without going to elaborate ends like we needed to do,” says Lawrance, who’s currently the technical director at indie developer Dry Cactus, makers of simulation-puzzle series Poly Bridge.

If that’s the case, then why isn’t destruction standard these days?

One might as well ask why people who are making much more money than they used to still often feel as if they aren’t earning enough. For one, things cost more than they used to; for another, the fancier your lifestyle, the more expenses you incur. The same paradox applies to game development, where there’s never a surplus of power: The more mighty computers become, the more game makers demand that they do.

Gawalek, who now works at PUBG Studios by way of Bethesda, explains, “With every new generation … everybody’s increasing the fidelity of the textures, there’s more NPCs, there’s more dynamic elements. … Developers are very quick to leverage those new resources to make the world more interactive and look better. So you’re still taking away some of those resources from those systems and reallocating them toward destruction.” As always, tradeoffs are required.

A destructible environment calls for dynamic lighting, which can illuminate a room differently if you knock down a wall and expose the area to an additional light source. But that tack is more “expensive,” from a processing perspective, than a static solution like lightmaps, or baked lighting. “If you compare a statically lit room to a dynamically lit room, if they look the same, the dynamic one costs a lot more toward the render budget,” Gawalek says. Opting for destructible environments with dynamic lighting might mean accepting some opportunity cost—like bidding farewell to 4K resolution at 60 frames per second—or delaying the game. “You can solve all those problems, but it definitely adds time to everything,” Gawalek says. And triple-A games are already taking a long time to develop.

If Red Faction: Guerrilla had been a huge hit instead of barely breaking even, Gawalek speculates, more developers and publishers may have been tempted to follow its destructible lead. As it was, he says, “I think the market largely saw [destruction] as a niche, and I think they decided to do more traditional, static environments.” Kertz comes to a similar conclusion about the wisdom of taking on a task that’s “orders of magnitude more complex,” saying, “I would wonder if somebody hasn’t done the calculation on a napkin about the return on investment. … If I can build two houses for the price of one destructible house and get two levels instead of one, is that worth more to players than one really highly destructible house?”

Considering those competing priorities, it’s no surprise that two of the three prominent new titles that rely on destructible environments have low-fidelity facades. BattleBit Remastered started as the passion project of three self-taught developers in three different countries who worked on the game on a part-time basis for several years. One of the three, lead artist Max Fink, says destructible environments were important to them both because the trio were fans of the “great destruction” in old Battlefield games—particularly Bad Company 2—and because “bringing down a wall is always cool.”

To pull off online play in destructible environments that support up to 254 players, the tiny team employed a low-poly look. “You still need some particle effects when something comes down,” Fink says. “You can’t go without it, it will just look too dull. But overall, the style definitely helps as it is very forgiving.” BattleBit is still in development, and its destruction has gotten more granular—with the size of its smallest wall fragments decreasing from big chunks to single bricks—even as it’s incorporated Levolution-esque events. “With BattleBit and the way we entered this market, we definitely put a light on it again,” Fink says. “That [a] destructible environment is something that is really great to have, even though it’s a pain to balance.”

Teardown uses cube-based voxel technology, which gives it a similarly blocky, LEGO-like look. Creator Dennis Gustafsson says the voxels “make destruction much more feasible in the game.” However, he’s had to make some of the same sort of allowances for gameplay and performance reasons that Red Faction designers did decades ago, including limiting both building and level size and making certain objects indestructible (in Teardown’s case, bedrock and the objects the player is supposed to steal, which it would be frustrating to destroy accidentally). When the game first surfaced in early access in 2020, it ran much slower than it does today, but Gustafsson subsequently spent several months optimizing its algorithms—most crucially, the one that determined when no part of the bottom part of a structure was still touching the top part, which was quite a time-consuming operation with millions of voxels involved.

One traditional challenge Gustafsson hasn’t had to deal with is the conflict between destruction mechanics and a linear narrative, because Teardown lacks the latter. Teardown started as a sandbox—which it remains for many modders who happily play in it—and became a heist simulator later. Because “the technology came first and then the gameplay was kind of an afterthought,” Gustafsson says, he “didn’t really feel much of those restrictions, because the gameplay was just already adapted to the limitations by the time it was invented.” Lawrance believes this latest group of games is better suited to destruction than many in the first wave were. “If the game is non-linear in particular, it really works better with destruction than a linear FPS like Red Faction did,” he says. “Maybe that’s something some people finally figured out: Start with destruction and then build something around that, as opposed to starting with a FPS and then saying, ‘OK, let’s add destruction into it.’”

Gustafsson, whose interest in game physics dates back to—of all the possible sparks—1998’s Carmageddon II: Carpocalypse Now, is intrigued by digital destruction but not by virtual violence. Unfortunately for him, those traits tend to go together. In a game with destructible environments, he says, “you’re given all these abilities to destroy stuff, and then if there would be people in that same scene, then it would be kind of a given that you could hurt people too. So it would have to be a game without people for that to not be violent.” In other words, it would have to be Teardown, whose unpopulated locations make it the rare destructible-environment game without virtual victims.

The Finals doesn’t have that quality in common with Teardown, but senior environment artist Joakim Stigsson, who admires Teardown’s technical wizardry, shares one sentiment that’s similar to Gustafsson’s: “We wanted the player to feel that this was a sandbox that they can shape.” In the case of The Finals, machine guns and grenade launchers do the shaping. In its realistic-looking graphics, its commitment to fancy physics, and its ties to DICE—whose former CEO, Patrick Söderlund, founded Embark—The Finals is an heir to the legacy of Red Faction (which Högström cites as an inspiration that’s “still like the staple of destruction for a lot of people”) and Bad Company (which he says is “also a game that’s been brought up a lot”). When Stigsson and Högström talk about how every department pulled together to make destruction a “core pillar” of the game, or how Embark’s artists had to think like architects to keep buildings upright, or how they made some elements indestructible and fine-tuned rubble for optimal traversal, they unwittingly echo Kertz, Lawrance, and Gawalek almost to the word.

The headlines that greeted The Finals last year were about as breathless as those 2001 takes on Red Faction. PC Gamer went with, “I almost don’t believe the destructibility in this FPS is real.” The Verge declared, “The Finals launches with destructive chaos like I’ve never seen before.” Someone on Reddit (an authoritative source if ever there was one) stated, “The Finals has the destructible environments we should have gotten in Battlefield.”

Is the positive publicity for these three games—along with the recent arrivals of lower-profile destruction-oriented titles like Deep Rock Galactic, Instruments of Destruction, and Lumencraft—enough to declare the destruction drought over? Are we entering a destructible renaissance? If fashion from the 2000s is back in again, why not game mechanics?

The time may be ripe, for a few reasons. As Gustafsson points out, better bandwidth has made physics-based online multiplayer more manageable. (Embark managed to lower latency further in The Finals by shifting the movement, physics, and destruction to the game’s servers, instead of each player’s system.) And with more games than ever being released, destructible environments’ viral appeal can help new titles cut through the clutter. “It definitely looks cool in the trailers if there’s things falling around,” Fink says. Sven Grundberg, Embark’s director of communications, adds, “It’s very rare to have a game that has such a standout feature. … So of course it’s been a huge boon for us in terms of being able to showcase something that is so concrete and that is inherently fun.”

More encouraging still, the barrier to entry from a programming perspective is shrinking. “The popularity of Unreal Engine 5 … sort of solves at least a lot of the initial problems,” Gawalek points out. The cutting-edge engine, which was formally launched in 2022—and which The Finals runs on—provides optional, off-the-shelf access to physics and destruction that used to require proprietary tech. “There’s still a lot of work to use those systems to build a game, but I don’t think it’s as scary, because you have a starting point, at least, with the available technology,” Gawalek says.

Granted, some games wouldn’t benefit from more destructibility. Gustafsson and Stigsson stress surprise as a selling point of destruction-based games: Teardown “gives the players so much freedom that it’s very rarely the same solution,” Gustafsson says, whereas “Every round, every minute of [The Finals], when you’re playing it or watching it, it’s different.”

Sometimes, though, that’s not what players—or established and aspiring esports—want. “Would Counter-Strike be better if I could blow a hole in the wall to get to the guys?” Kertz asks. Fink invokes Call of Duty to make the same point, adding that some BattleBit competitions have banned certain explosives so that attackers couldn’t destroy walls to expose an objective that a team was trying to defend. Kertz continues, “You can’t constantly change the game on people. … Destruction has an inherent predictability because it’s player controlled. So if you’re trying to build a game which is very competitive and it’s about the skill level of the player … and the strategy of it, you don’t want to throw those extra unpredictabilities into it, because it diminishes the experience.”

Overall, Gawalek and Gustafsson expect to see some increase in games that boast destructible environments. But because of the cost and difficulty, Lawrance expects them to remain something of a “niche thing.” So does Kertz, barring some kind of quantum computing breakthrough. “I’m surprised any game ever ships,” he says, semi-seriously. Add in the headache of destruction, and as Högström says, “You have to invest a lot, and you have to be stupid enough, like us, to do that.”

Gamers are accustomed to Moore’s law–like progress. They expect resolutions to get higher, draw distances to get longer, and loading times to get shorter. Destructibility hasn’t progressed so steadily. But while games like Guerrilla and Bad Company 2 were ahead of their time, that time could be coming.

“I think destruction is a lot of fun,” Lawrance says. “And I think players that want to seek it out find the games that are there for it.” Those players have a few new options. And one way or another, all of those games are bringing down the house.

Video Games

The Video Game Remake Boom Is Producing Nostalgic Hype and Uncomfortable Questions

Button Mash

The ‘Final Fantasy’ Party Draft

Button Mash

‘Final Fantasy’ Preview, ‘Suicide Squad: Kill the Justice League,’ and ‘Halo’ Season 2

View all stories in Video Games