PEAK received its final major content update on 11 August 2026.
The game isn’t shutting down. It hasn’t been delisted, and its developers haven’t disappeared. They’ve released several patches since.
They’ve simply decided they’re done adding major chunks of new content.
I think we’ve become a little unused to that idea.
Somewhere between Minecraft, Fortnite, battle passes, seasonal roadmaps, Discord servers and years of Steam announcements, the expectations around games have changed. A game which stops receiving substantial new content can now quite easily be described as “abandoned” or “dead”, even when the game itself is still completely playable.
I understand where that comes from. If you really like something, naturally you want more of it.
But I think there’s still value in remembering what a finished game looks like.
Somewhere, “finished” became “abandoned”
I grew up with games that ended.
I don’t just mean that they had credits. Development ended too.
You bought a game, put the disc in your Wii and that was broadly the game you had. There might be an expansion or some DLC later. If something was horribly broken, you could reasonably expect it to be fixed. What you didn’t generally expect was a permanent stream of new maps, mechanics and events.
Nobody was wondering about Wii Sports’ content roadmap in 2012.
Eventually, the people who made a game went and made another game. This was normal.
Live-service games have changed those expectations, and not without reason. When a game actually sells itself as an ongoing service, continued development is part of the deal. If a publisher is selling battle passes and subscriptions, advertising seasonal content or promising a multi-year roadmap, players can reasonably expect those promises to mean something.
PEAK isn’t doing that.
Landfall’s FAQ is unusually direct about it:
PEAK isn’t a live-service-game.
The developers say PEAK was intended to be finished when it launched. Everything added afterwards was extra.
Its success meant they spent much longer on the game than expected, but it didn’t fundamentally change what PEAK was supposed to be. Landfall has also said it doesn’t want the game to become a “forever game”, continually accumulating things simply because there needs to be another reason for players to return.
I think that’s a perfectly reasonable creative decision.
PEAK can just be a game.
Those expectations came from somewhere
I don’t really blame individual players for getting used to games behaving this way.
The industry has spent years teaching us that successful games should keep changing.
I personally put a fair amount of the blame on enormous companies like Epic Games and the model they’ve built around games like Fortnite.
Fortnite is supposed to keep moving. That is part of what it is. There are seasons, events, collaborations, map changes, new modes, balance changes and regular releases. You can leave for a few months and come back to something noticeably different.
There’s nothing inherently wrong with that model. Fortnite is an ongoing service backed by one of the largest companies in the games industry, and Epic has built both the organisation and the business model needed to sustain it.
The problem is what happens when that starts to feel like the normal rhythm of videogame development.
If one of the biggest games in the world is constantly producing something new, it’s easy to internalise a simple rule: games receiving content are alive, and games which aren’t receiving content are dying.
Then you look at something like PEAK.
Fortnite keeps getting things. PEAK is eventually going to stop getting major things. On the surface, one looks “active” and the other doesn’t.
But they are completely different projects.
Epic maintaining Fortnite and a small group of indie developers maintaining PEAK are not comparable operations. They don’t have the same number of people, the same finances, the same goals or even the same idea of what their games are supposed to be.
That’s where I think the model built by huge live-service games starts distorting expectations of independent developers.
Not because somebody wanting another PEAK biome has done anything wrong. Wanting more of a game you love is about the nicest problem its developers could have.
It only becomes awkward when the development cadence of a company like Epic quietly becomes the benchmark everybody else is measured against.
An indie studio shouldn’t have to imitate Fortnite’s content cycle to prove that it still cares about its game.
Sometimes a small studio has simply finished making it.
PEAK was never supposed to last forever
PEAK’s history makes this especially interesting.
The game came out of a collaboration between Aggro Crab and Landfall. Three people from Aggro Crab joined four from Landfall’s side in an Airbnb in Seoul for what started as a month-long game jam.
They bought some desks, worked on a game together, talked about it over dinner and spent their evenings playtesting it.
Seven people made the thing that became PEAK.
It wasn’t conceived as the next decade-long entertainment ecosystem. It was a weird collaborative project between two indie studios which happened to become enormously more successful than anybody involved expected.
And PEAK is exactly the sort of stupid multiplayer game where that can happen. I mean “stupid” affectionately.
You climb a mountain. Your friend falls off it. You attempt a heroic rescue and fall off as well. Somebody eats something obviously poisonous. Somebody finds a brilliant route upwards and immediately walks backwards off a cliff.
Excellent videogame.
Its success kept the developers working on PEAK for much longer than planned. It got patches, new mechanics, new biomes and all sorts of other additions.
By December 2025, though, the team was openly saying the pace was taking a toll. Frequent updates from “just a few people” were exhausting them. They said they’d spent months running on “pure adrenaline and hype”, while Aggro Crab and Landfall still had their own studios and other projects to think about.
They warned players that updates would become less frequent in 2026.
Eight months later, on 11 August, The Final Ascent arrived. It brought Gloom, the Citadel, new items, mechanics, achievements, fixes and other additions.
That’s the last major content update.
To me, that doesn’t look like a game being abandoned. It looks like a development team deciding that the thing they made has reached the shape they want it to have.
I think about this from the other side too
Part of why this interests me is that I’m a developer myself.
I actively develop and maintain a frankly silly number of projects at once. Most are public on my GitHub, and my various projects and sites are tied together through croft.click.
There’s Inkwell, my standard.site app; Malachite, which works with listening data on AT Protocol; Wolfram; MetalBear; Clay; svelte-standard-site; my website; and a pile of smaller AT Protocol tools, bots and experiments.
Some are substantial applications. Some exist because I needed a particular tool one afternoon and thought other people might find it useful too. Some will eventually be archived.
Most of it is open-source and free at the point of use.
I also don’t currently have a job or a stable income.
Most of the public tools I deploy are affordable for me to run because Vercel provides infrastructure I can use for free. I’d prefer to use more of my own infrastructure, but recurring infrastructure costs aren’t something I can sensibly take on without reliable income.
My personal AT Protocol PDS is an exception. That runs on my own Mac, although I want to move it off the Mac eventually.
At the moment, my domains are basically the only recurring cash cost of my projects.
I rely on gifts from my family at birthdays and holidays, as well as voluntary support through Ko-fi and GitHub Sponsors.
I’m genuinely grateful for that support.
It makes what I do easier, and there’s something lovely about another person deciding that something I made is worth supporting.
But it also doesn’t mean that either side has entered into an indefinite maintenance agreement.
People star repositories, open issues, send me a few pounds or use things I’ve written because they find them useful. I maintain those projects because I enjoy programming, because I like solving problems, and because I like contributing things to the communities I’m part of.
Ideally, that remains a positive relationship.
If I eventually lose interest in a project, I can archive it. Someone might fork it, maintain their own version, find an alternative or simply be disappointed that it isn’t continuing.
None of those reactions would be unreasonable.
I just think there has to be room for the person making something to eventually say that they’re done with it.
Commercial software obviously involves different economics, but I think some version of that principle still applies there too.
Buying software comes with reasonable expectations
Paying for a game does create obligations.
If a game barely works, it should be fixed. If multiplayer is fundamentally broken, that needs attention. Serious security problems need fixing. If features were advertised and never delivered, customers have every right to complain.
And if a publisher explicitly sells people a five-year roadmap, taking money on the promise of five years of content, then continued development really is part of what was sold.
Saying that games can end shouldn’t become an excuse for releasing unfinished products and disappearing.
I write software. I know how optimistic the word “finished” can be once a sufficiently complicated program meets the real world.
There will be bugs. Someone will have a cursed combination of hardware and drivers you never tested. Someone else will somehow wedge an object through a wall in a way your physics engine clearly regards as a personal attack.
Fixing that stuff is maintenance.
What I think is worth separating is maintenance from expansion.
Keeping a game working is one thing. Continually making it bigger is another.
PEAK demonstrates that distinction nicely. The Final Ascent came out on 11 August, followed by patches on the 13th, 18th and 27th. You can see the continuing updates on Steam.
As I write this on 2 September, the game has already received its “final major update” and its developers are still working on it.
Landfall says it intends to continue maintenance until it’s happy with PEAK’s stability. There are also console versions and cross-platform play in the works.
The game is still being supported. It just isn’t going to receive major new content indefinitely.
I think that distinction matters.
An update is work
As players, we mostly experience development from the other end.
A Steam notification appears, we download a few gigabytes, and now there’s a new biome.
The work required to get it there is mostly invisible.
Someone decided what that biome should actually be. It had to be designed, modelled, textured, programmed and tested. New mechanics had to interact with everything already in the game. Multiplayer created its own collection of things that could go wrong.
Then somebody discovers that an item falls through the world when Player 3 reconnects while facing north-east on a Tuesday.
Fix that. Test it again. Find out the fix has broken something seemingly unrelated.
This is what making software is like.
There’s no ADD CONTENT button in an office somewhere.
Even a fairly small update represents working hours. A major update represents a lot of them.
It’s easy to forget that because the work gets compressed into patch notes. We see a list of features and fixes rather than all the people who spent time making them happen.
PEAK’s developers have already told players that their update schedule was exhausting them, and that they have other work they’d like to do.
I think it’s reasonable to take that into account when talking about what the game should receive next.
Free doesn’t mean costless
Open-source software has made me especially aware of the gap between what something costs the user and what it costs to make.
You can open a website, use a tool or install an application without ever being asked for money. From your side of the screen, it was free.
Somebody still had to make it.
Maybe a company is covering the hosting. Maybe a free cloud tier makes the project viable. Someone might be paying for domains or running hardware at home. Plenty of maintainers are simply spending their own spare time on it.
“Free at the point of use” only describes the transaction the user sees.
Commercial games obviously have a different funding model, but they’re still made by people.
There are programmers, artists, designers, writers, musicians, testers and producers on the other side. They get tired. They change jobs. Their circumstances change. Sometimes they simply want to work on something new.
That doesn’t mean players shouldn’t have expectations of a commercial product. It just means those expectations exist alongside the fact that the people making it have finite time.
Success should create room for something new
One of the things I find exciting about PEAK’s success is what Aggro Crab and Landfall might be able to do with it.
Maybe it funds another weird experiment. Maybe they make something completely unlike PEAK which none of us know we want yet.
There’s always a tension there, because success naturally creates demand for more of the successful thing. From a purely commercial perspective, continuing to make the thing people already like is often much safer than starting again.
But creative studios also need room to move.
Landfall has made Totally Accurate Battle Simulator, Stick Fight, Content Warning, HASTE and all sorts of other strange games. Aggro Crab made Going Under and Another Crab’s Treasure before PEAK.
I like those studios partly because they make different things.
Every month spent producing another PEAK biome is a month that can’t be spent on whatever comes next. That isn’t an argument against updates; PEAK has benefited enormously from the updates it received.
It’s just the trade-off involved.
If every successful game has to become a ten-year project, developers inevitably have less time to make new games.
More isn’t always better
It’s also worth questioning the assumption that adding more stuff necessarily improves a game.
Software developers know where that road can end.
A program gets popular because it does something well. People ask for features. Years later, something that once had a clear purpose can become difficult to recognise under everything added to it.
Games aren’t immune to that.
Every new mechanic has to coexist with the old ones. Items affect balance. New areas create more things to test. More systems mean more for a new player to learn, and every addition becomes another thing that somebody somewhere would miss if it disappeared.
There’s nothing inherently wrong with a large, constantly evolving game. Some games are designed around exactly that.
PEAK apparently isn’t.
Its developers have said they don’t want to keep adding things simply to retain people’s attention. They have an idea of what they want the finished game to be.
I think developers should be allowed to make that call.
Sometimes a game already contains enough.
Mario Kart Wii is “dead”, and that’s fine
Mario Kart Wii is probably the clearest example I can think of.
Nintendo released it in 2008.
Development ended. Nintendo moved on. The official Nintendo Wi-Fi Connection service eventually shut down on 20 May 2014, taking the game’s official online multiplayer with it.
By the standards we sometimes apply to modern games, Mario Kart Wii is extremely dead.
And yet people are still playing it.
The difference is that the community picked it up.
Wiimmfi provides replacement online services for games which used Nintendo Wi-Fi Connection, including Mario Kart Wii. CTGP Revolution has spent years extending the game with custom tracks, new modes, online time-trial leaderboards and other features.
The game has effectively moved into a different phase of its life.
Nintendo doesn’t need to keep assigning developers to Mario Kart Wii for people to continue enjoying it. The original team finished its work years ago, while the community has chosen to preserve and extend what they made.
I think that distinction is useful.
Nintendo made Mario Kart Wii.
Nintendo finished Mario Kart Wii.
Players kept Mario Kart Wii.
The modding community can add tracks, maintain online play, organise competitions and generally keep experimenting with a game that is approaching two decades old.
None of that means Nintendo left the original game unfinished.
It just means a community loved a finished game enough to keep doing things with it.
That feels like a much healthier way of thinking about longevity.
The original developers can move on.
If people still love what they made, the work can continue to have a life of its own.
Things end
There’s a wider discomfort here with the idea that things have lifespans.
Software changes. Websites disappear. Services shut down. Communities move. Developers change jobs. Sometimes something you love reaches its final version.
That can be disappointing without somebody necessarily having done something wrong.
Modern software makes permanence look deceptively easy because software can always be changed. There can always be another release.
But somebody has to make that release.
Dependencies need updating. Platforms change underneath applications. New operating-system releases break old assumptions. Bugs need investigating and issues need triaging.
Most importantly, someone has to still want to do it.
There isn’t really a technical state called “forever”. There are just people continuing to spend time maintaining something.
At some point, they may decide that time is better spent elsewhere.
Preservation is different from continued development
I’m very sympathetic to preservation.
I want old games to remain playable. I want local saves, documented protocols and useful export tools. I want old releases archived. Where multiplayer depends on central infrastructure, I’d much rather communities could run their own servers than watch a game become useless when a publisher loses interest.
But preservation and continued development are different goals.
Mario Kart Wii demonstrates that nicely. Its original developers are long finished with it, but the game itself hasn’t disappeared. Communities stepped in where they wanted more.
PEAK can eventually work the same way.
If it stops receiving official patches one day, preserve it. Keep the builds. Document it. Mod it. Play it.
Landfall already encourages PEAK’s modding community and says it has shared some source code with modders where useful. That gives people who love the game somewhere to take it without requiring its original developers to keep producing major updates themselves.
The work can outlive the period in which its creators actively work on it.
That’s normal for almost every other kind of creative work. I don’t see why games need to be completely different.
PEAK isn’t going anywhere
The funny thing about calling a finished game “dead” is that the game is still there.
You can play PEAK today. You’ll still be able to play it after Landfall announces another project. The Final Ascent eventually disappearing from the top of the Steam news feed doesn’t remove Gloom or the Citadel from your copy of the game.
People have played finished games for decades.
A game doesn’t need to be current to be worth your time. It doesn’t need a roadmap or a battle pass ticking down in the corner. It doesn’t need another update announcement every few weeks to remain enjoyable.
Most people asking for more PEAK are doing so because they like PEAK.
That’s a nice problem for a developer to have.
I just think we should leave room for an answer of “we’re happy with it now”.
PEAK launched as a finished game, became far more successful than expected, and received more than a year of additional development afterwards.
Now its developers want to stop adding major content and eventually move onto other things.
I’m fine with that.
I like PEAK because of the game they made, not because of the possibility that it might keep growing forever.
And when Aggro Crab and Landfall eventually show us whatever comes next, I’d quite like them to have had the time and energy to make it.