Over the past couple of years, extraction shooters have become one of the most in-demand genres: new projects are releasing one after another, and playtests have become as mandatory a stage as trailers and launching a page on Steam. If earlier almost every new project was presented as a \"killer\" of Escape from Tarkov, then by 2025 developers had finally stopped blindly copying the original source. They began taking the genre's basic rules in order to build something original on top of them — with a different pace, risks, and raid structure.
To gather initial feedback and avoid shooting themselves in the foot at launch, teams are increasingly moving into closed tests or, conversely, opening access to everyone interested. We took part in the playtests of five of the most notable projects and are sharing our conclusions: what already works, where the genre is still getting stuck, and which games are truly worth watching.
PUBG: Black Budget
The developers of PUBG: Battlegrounds offered their own take on the genre, but their ideas do not seem innovative. Even in the opening video, the attempt to sit on two chairs at once is noticeable: on the one hand, extraction with stakes and the risk of losing loot; on the other, the shrinking zone from battle royales. However, these genres pull gameplay in different directions and interfere with each other.
The shrinking zone in itself is not the problem. The problem arises when the game simultaneously forces you to complete tasks in specific locations. While the map shrinks on a timer, the raid turns into forced fuss: instead of planning a route, you have to move chaotically between the nearest points. And if you fail to reach the required place, all that remains is either to evacuate early or start a new raid. As a result, it feels like a pointless waste of time.
Worse still, the choice of drop location works inconsistently. In almost all the raids we played, the game threw unpleasant surprises at us: in some cases it was impossible to choose a point, in others it bugged out at the start. For an extraction shooter, this is critical: starting conditions must be clear so you can build a route and make decisions, rather than begin a match thinking, \"let's see where it throws us this time.\"
There are complaints about the shooting too: it was transferred directly from PUBG: Battlegrounds — together with one of its most irritating features. In addition to normal recoil, there is also visual recoil: the camera shakes so much that you are fighting not the enemy, but your own screen.
But all of this pales against the main disaster of the test — the technical state. Regular crashes ruin the gameplay experience. They can happen at any moment. Once every few hours, the game would begin putting more and more load on the system and eventually close with an Unreal Engine error. For a mode like this, that is especially painful: while you are trying to return, your character can be killed by NPCs, other players, or the zone itself.
In the end, Black Budget left the impression of a raw hybrid. It tries to stand out by merging the rules of an extraction shooter and a battle royale, but their incompatibility clearly shows that this shooter has not even passed the concept check.
Exoborne
Exoborne is a hyperactive extraction shooter where movement works across all three dimensions: dashes, verticality, and constant repositioning set a frantic pace for the raid. The developers are consciously avoiding \"chess\" and complex tactics — the bet is on action and spectacular skirmishes. This model, of course, has its own weak points, but judging by impressions they are not critical.
What causes much more indignation is the studio's policy, which forbids test participants from sharing their impressions. The developers explain this by saying that the game is still in development, but the community does not believe these arguments and cites examples of other teams in a similar state communicating openly and not being afraid of criticism.
The shooter's main gimmick is the exoskeleton. It allows the character to run almost nonstop and jump high. In addition, each player has a grappling hook and a folding parachute. The hook helps you climb hard-to-reach areas and quickly change elevation, while the parachute is needed for rapid airborne transfers — to catch a tailwind and \"ride\" hurricanes. The exoskeletons come in several types for different roles: some are focused on survivability, others are built for close combat, and still others make it possible to turn invisible.
The overall pace of the game is comparable to Apex Legends: extreme mobility; armor and health reserves have to be maintained with consumables. At the same time, armor is divided by quality and durability reserve, but even so it does not turn players into walking tanks.
A distinctive feature of the game is the camera: all movement and interactions take place in third person, which makes it easier to set up ambushes. But when a firefight starts, aiming can be switched to first person — this makes shooting more accurate and understandable, without breaking the overall mobile dynamic.
The maps, meanwhile, are spacious: there is almost always a choice of where to go and what route to build for your current tasks. But it is precisely in open spaces that battles often feel almost like in an MMO — victory goes not to the one who shoots more accurately, but to the one who took the advantageous position and \"read\" the approaches earlier. The camera also supports this logic: the right angle lets you spot an enemy from cover, catch the moment, and start the firefight from an advantage.
The weakest point is the loot. Frank junk turns up everywhere, and the raid quickly turns into sorting through samey trash. At the same time, the inventory here is noticeably more modest than you expect from the genre, so paying off your gear in a single run is not always possible: there is little space, and useful finds are even fewer. The local artifacts are what really spark interest, but access to them is almost always blocked by enormous guard robots. There is no way around them, and trying to destroy them turns into an almost impossible task.
In the last two tests, the game managed to show its identity, but it is important for it to shore up the basics — stability and progression — and for the developers not to abandon open coverage of the project. After the last test, the team took a pause in public communications and promised to return only after noticeable progress. This silence has already lasted four months.
EVE Vanguard
CCP Games had for many years the idea of making the famous space wars of EVE felt not only in space, but also on planets — in a more grounded format where factions fight for resources and influence under the stars. EVE Vanguard is an attempt to create a shooter by the rules of EVE Online, where instead of ships you play as mercenaries.
The latest open playtest — Operation Nemesis — showed players its vision of the project, which in many ways overlaps with the extraction shooter genre: dropping onto a dangerous planet, receiving a set of tasks at the start of the raid, completing them during the sortie, gathering resources, and evacuating with the haul. The valuables collected can be used to create advanced weapons, ammunition, and equipment, which partially makes subsequent raids easier.
The risk here is reinforced by EVE's trademark idea of \"technological immortality\": the mercenary lives off shields and a health pool, heals using biomass, and can be reborn after death. But only if the squad has enough clone reserves. At the same time, the game encourages thoughtfully insuring your haul — for example, credits and chipsets can be deposited into terminals in advance, and then they do not burn up in the event of death.
In Operation Nemesis, the developers focused on content and convenience. They added clear tutorials for newcomers, expanded the set of weapons, and improved gear production: now you can assemble guns from blueprints, pick options, and customize them to your style. They also introduced elements of team interaction: voice chat with nearby players, the ability to revive a teammate, and a signal menu.
No less significant a part of the playtest is the separate PvP mode Insurgency: two teams of nine people fight for control over key objectives, gradually taking over the opponent's territory.
And for EVE Online fans, the developers also prepared an event similar to a hybrid battle in War Thunder: some fight on the planet, others in space. Together they hunt convoys in order to advance the event from two sides at once.
EVE Vanguard looks like an interesting support title for the original series. The game has a well-constructed weapon system, interesting design, and a competently organized raid process. But at the same time there are jerky animations, a meager set of repetitive tasks, as well as a heap of bugs that can cause you to lose gear while loading into a raid.
Release is scheduled for the third quarter of 2026. If the developers do not manage to solve these problems, then even the loud EVE name will not save another extraction shooter from failure.
Enginefall
In terms of setting, the impression is that the local extraction shooter was somehow crossed with Rust. The entire match unfolds inside a gigantic train rushing through a post-apocalyptic world, and players are not simply fighting for loot — they are fighting for the right to move farther along the train and secure themselves in new sections.
The cars function as separate zones: cramped corridors, forcing point-blank shooting, alternate with more open compartments where you can take a position and control the passageways. Because of the linear structure, advancing forward feels like a constant race. If you linger, there is no point in moving into the next cars.
Progression is tied to the level of the cars: the starting sections provide basic materials, while closer to the head of the train the loot becomes noticeably more valuable. The peak of this race is the driver's cab: the winner gets the right to turn key resources into fuel cores needed to develop their own shelter-train. In addition, from the cab you can address all players through the loudspeaker.
Along the way, the team can build small temporary bases: respawn points, places to store loot, and ways to secure themselves in a chosen section. Over time, the base opens access to a car through which you can evacuate, so building affects not only comfort but also the chance of carrying your property out.
The idea is strong, although the interface still needs more prompts and better clarity — especially when you need to quickly understand what can be built and how it will help in practice.
But in the Enginefall test, the game felt too raw: technical roughness is noticeable, as are uneven visuals and choppy animations that ruin the pace. At the same time, the game already has a good concept. If the developers improve stability and usability, this train has a chance to reach its audience.
Marathon
The first closed alpha tests of Marathon in April-May 2025 were noisy, but left a mixed impression: even supporters of the project noted that the game felt like \"just another extraction\" without a face of its own or compelling reasons to stick around for long.
And some time later, the project found itself at the center of a scandal over stolen design from the artist Antireal. Bungie publicly acknowledged the problem and explained that the assets got into the build due to an unnamed former employee. At the same time, many studio employees, including the art director, were subscribed to the artist.
The visuals were revised. The depth of light and shadows was increased, scenes became darker and more contrasty, and the textures more varied. Wet traces, dirt, and reflections are visible on surfaces. Dynamic weather and environmental effects add volume, but the overall artistic language remains the same: angular silhouettes, bright accents, and a fairly straightforward presentation of the interface — all this still may not appeal to everyone.
In terms of content, the game feels broader: more maps and more Runners — classes with unique characteristics and abilities.
The second notable shift is the attention to solo players. If after the alpha there were constant complaints that it was impossible to go on raids without a coordinated team, now Bungie has launched a separate queue for solo players.
Later, this direction was reinforced by the Rook class — an analogue of the Scav from Escape from Tarkov. It already has its own gear, and you can calmly gather loot on it without the need to prepare for a raid in advance.
The developers also reconsidered their attitude toward social interaction: voice chat appeared in the game, which the developers had refused for a long time. Previously, the studio believed that it would become not a communication tool, but a weapon for expressing toxicity.
Now the game has more opportunities for peaceful negotiations. However, unlike ARC Raiders, Marathon does not position itself as a \"kind\" extraction shooter.
The pacing and feel of firefights underwent important reworks. Character health was increased, while shield durability was cut, which reduces the value of expensive gear. Another subject of controversy was aim assist on PC: players noticed the Aim Magnetism feature, which allowed aiming with keyboard and mouse almost as if playing on a gamepad. After unrest in the community, this option was removed.
Today, Marathon is perceived as a project capable of fixing even unacceptable mistakes. The game has become richer in content, noticeably improved in world readability, and restored its basic multiplayer mechanics.
If the final version eliminates the remaining problems and does not end up at the center of new scandals, then this will be a rare case when a game was brought to a near-ideal state within a year. And if not, it will be forgotten as quickly as many other \"Tarkov killers.\"
Verdict
Playtests rarely show builds close to final. However, they honestly demonstrate the foundation: what the developers have been working on all this time, what the graphics look like, how clearly the gameplay is structured, and whether the project has a pace of its own that shapes its identity and distinguishes it from its neighbors in the genre. Based on test results, you can understand how well the developers' ideas work.
This works both ways. Some projects, after several sessions, leave the feeling that they have a future: even with problems, you can see where the authors are taking the game and why it may hook its audience. Others, on the contrary, quickly lose their credit of trust: either the developers fail to grasp the basic principles of the genre, or they do not believe in their own concept.
Tests are neither a sentence nor a guarantee of success. But as a tool of primary analysis, they are indispensable: they show what the developers have achieved and make it possible to understand what to expect from the release — and whether it is worth waiting for at all.