Well, I was mostly joking when I took credit for Ace Combat's director finally getting to fly in a real jet. Turns out our interview really was the catalyst. If anyone else has a dream waiting to be realized, give me a shout. Apparently I have the magic touch.

Kazutoki Kono: You're welcome.

Back in June, I asked the Ace Combat brand director, who's been working on the series since 1997's Ace Combat 2, if he'd ever gotten to strap into an F-16 and experience the real thrills of doing barrel rolls at 30,000 feet. "Unfortunately over the last 30 years of my career I've been appealing, saying 'hey, we want to ride in one of these jets!' but we have not yet had the chance," Kono said at the time. "Neither myself, or anyone on the team."

Mere months later, it's a whole new world. In a promotional video released today alongside Ace Combat 8, Bandai Namco revealed that Kono finally got to climb into the cockpit of a jet and live the real life version of the series he's been overseeing for years and years. Well, minus the combat parts.

"I was scared before getting in. But once we were airborne, it turned into one of the most beautiful experiences," he said in the video. "The G-force during turns was noticeable, but what stood out the most was the force pushing me downward into the seat. Being up in the sky inside such a compact cockpit felt incredible. It was a surprisingly comfortable and enjoyable space to be in. Compared to a large aircraft, it really made me appreciate the appeal of fighter jets."

One way to look at this sequence of events is to see the ad as a logical and inevitable marketing step for a series that Bandai Namco has clearly put more money behind than ever before, including a live-action YouTube miniseries. The other way to look at it—as I'm choosing to—is that the public outcry after my groundbreaking interview was published reached a level of fervor that simply could not be ignored. It may have only gotten one comment on pcgamer.com, but the ripples must've been felt all the way over in Tokyo.

Clearly the people demanded Kazutoki Kono be given the chance to soar that big blue sky, and the money men had no choice but to pay for our guy to hitch a ride in a MiG-21. Whatever it cost, it was clearly worth it. Look at Kono in his post-flight interview. He's glowing!

That's the look of a man who's just had his dream fulfilled. And even after having just experienced the wonder of flight at a speed and intimacy few on this earth will ever know themselves, he remained steadfast in his duty as brand director.

"Looking out at the clouds and the sky while flying, I couldn't help but think how Ace Combat is incredibly well-made," he said. No argument here.

Of all the games on the horizon, the one I may be most looking forward to Let's Build a Dungeon. It's a sim from Springloaded, makers of Let's Build a Zoo and the upcoming RollerCoaster Tycoon Wonderworks, and the premise is one we've all mulled at one point or another: what if I could make my own videogame?

And in Let's Build a Dungeon, you can. Dig dungeons, populate them with monsters and traps and loot, design quests, and watch simulated players play your game. You'll even be able to let real players play the game you make.

But there's also the simulation surrounding the game you're building. The development studio you've founded. The devs you hire. You'll have to deal with investors, finances, the public, the press (like me!), crunch, gaming conventions, hackers, and even text messages from your mom.

There's a lot going on in this sim. Like, a lot, a lot. During a recent preview and interview with Springloaded founder and Let's Build a Dungeon lead developer James Barnard, here's what leapt out at me as being the most a lot among the lot.

"We have, like, 700 [fictional] games or something, and the list keeps on getting longer."

Let's Build a Dungeon simulates Steam's backend and sites like SteamSpy where you can see your game alongside the competition. Lots of competition: Springloaded has invented hundreds of fictional (and unplayable, sadly) games to populate your reality.

"We have all these games, which obviously our concept artist, poor guy, he spent his his whole life drawing these things."

And look, there's even an advert for a gaming show. A computer gaming show. That's a great idea for an event! Someone should do that for real.

"You can see sort of how your game is doing with the wishlists and things like that. The objective is to get into events, and if you do, you get massive number of impressions. So we really use the impressions, click through rate, conversions, wishlists, and all of that stuff feeds into the chart."

"When you go from CGA to VGA, you have to upgrade some of your computers, and then the size of everything expands and becomes bigger, so suddenly you need more hard drive space because all your assets are now high resolution."

The fictional game you're making in Let's Build a Dungeon runs on fictional hardware, and as your game gets bigger, it takes up more space. Time is passing, too, meaning there are advances in PC hardware as well as your game's required specs, and that's reflected in the sim.

"We drew every single component of every computer, so there's about 150 components that we drew from, and it goes through the history of computing."

You won't just see a difference, you'll hear one. The soundtrack of the game will change based on the audio card your simulated computer is using.

Buying plane tickets and booking hotels for your devs to attend game showcases is a "ridiculous minigame" that even simulates layovers

"Once [your employees] head off to the [showcase], you get to pick who's going to go to the event, and then they aren't working, so they won't appear in your design board. If I send these two people to an event, they're going to disappear for three days and and, you know, fuck up my schedule.

"The flights kind of curve up in price as you get closer to the day, but then sometimes there'll be a bargain right at the end if they've got too many seats to sell, and the same for the hotel. So if you wait till the last minute, you sort of take this risk of: Am I going to spend too much?

"We calculate where you started your company and where the event is. We actually calculate the distance and how long it takes and we do the stopovers in the right cities, so where your office is, is gonna impact you. So if you get EGX Rezzed, it's just down the road for you, if you pick London."

Remember Cyberleek? That could happen to you

"This is some super nerd who's basically hacked your computer. Oh no, he's been following your investors' flights around the world, and he's decided that you are making an MMO, and he's looked at your financials and the other guy's financials, and then he holds you to ransom.

"He's either going to leak the game, or he's or he's going to convince you to put him in the game. There's lots of different outcomes."

Your mum sends you pie recipes and cat videos

"Your mum calls you. Have you eaten? And then she tries to make you eat different food, and gives you desktop screenshots and things for rewards.

"And this conversation, for example, you can get a song, you can get like different desktop images to cheer you up because she wants you to be encouraged in your daily life.

"One of the other things she gives you is a pie, and then she sends you the recipe for the pie, and then you get inspired, and you add a the ability for your designers to come up with a pie recipe, and then you get a pie that you can put in the game, and the players can eat, and you can adjust all that kind of stuff."

By the way, if all of this game dev studio simulation stuff isn't your thing and you just want to use Let's Build a Dungeon to design your own game, you can do that too! There's a campaign mode, build mode, creative mode, or just play the games created by other players.

Let's Build a Dungeon is coming to early access in February. There's a demo you can try in the meantime.

I am a disastrously untalented individual, but my biggest failure of all? My inability to sew. Holes in clothes? Unpatched. Long trousers? Unhemmed. Thrifts? Unflipped. It is both a pure skill and creativity issue, and I forever look on longingly at those far more talented than myself.

Turns out that extends to one of the hottest new cosy games of 2026, too. Dressmaker has absolutely taken over my life—and the life of many new virtual sewists—with its relaxing yet devilishly moreish garment creation. It's genuinely brilliant, a testament to how good games can be when you zoom in on a niche and go all-in on it.

Sketch dresses with different bodies, skirts, sleeves, collars; doodle in rough colour schemes and lay fabric samples over each other to find the perfect combination. Sift through bolts of silk, lace, linen, cotton, and velvet in an absurd number of colours, before laying out all the parts of your bespoke pattern and sewing them all together. It's brilliant, but I am naff at it. Literally, look at this dress I tried to design with the brief "what if Geralt of Rivia was my client?" in mind.

Where's the pizzazz, the drama? Thankfully, social media has been blessing me with people who actually understand how to make the most out of Dressmaker's gorgeous adornments and fabrics—for both good and evil. Here are some of my absolute favourites for you to feast your eyes upon:

This incredible armour-inspired piece

A stunning Van Gogh ballgown

This recreation of an upcoming Genshin Impact character, Tsaritsa

Of course, I had to include this amazing Emet-Selch dress from Final Fantasy 14

And finally, this deeply cursed but also peak artistic Excel minidress that I would 100% wear in real life

Imagine for a moment: It's the year 2007. All software looks like glossy, rounded plastic, dewy glass, and photographs of verdant fields. Every electronic device makes ethereal chimes when it starts up. You've never heard the phrases 'live-service,' 'large language model,' or 'prediction market.' It's safer here, amidst the dreamy lime greens and ocean blues. It's better.

And now, thanks to a new house-builder that launched on Steam this week, we can live the definitely-healthy fantasy of retreating into the security of yesterday's software aesthetics. Frutiger Home is a home decorating sim where you can build an impregnable psychic cocoon out of the finest design elements of an era where technology could still feel optimistic, and not like perpetual cognitive radiation poisoning. Hooray!

All its architectural themes, available color swatches, and interior accent pieces are inspired by the Frutiger Aero aesthetic—a term retroactively coined by Consumer Aesthetics Research Institute co-founder Sofi Xian to describe the glossy, glassy, gleaming visual language that dominated tech and UI design from the mid-2000s to the early 2010s. Here, sharp corners are an aberration. Every surface that can be polished, will be polished.

When I loaded into an empty home to begin decorating, one wall of the living room was dominated by a screen showing clownfish and dandelions. I could change it to show a collection of citrus slices, because of course I could. Once, our devices held these images in reverence. A computer was a window to an idealized version of nature: one that was mostly seagulls and water floating through the air for some reason.

It was the Zune's time.

After browsing the selection of furniture that looked like it was extruded straight from the dimension where sans-serif typefaces live, I stepped outside of my fully-rounded home and into a field where the impossibly green hillsides of that Windows wallpaper—you know the one—stretch into the distance, forever. You can join me here. I've placed two chairs with loops for legs. We can pretend it's 2007 for as long as we need to.

Frutiger Home is available now, on Steam, in 2026, for $5.

Ace Combat 8: Wings of Theve is technically out on Thursday, but as is commonplace now folks who pre-order the $90 deluxe edition get to hop into the cockpit a few days early. Anyone who pre-orders the $70 standard edition, meanwhile, gets a pretty sweet bonus: Ace Combat Zero, a 2006 entry in the series that has never been on PC before now. But how Bandai Namco has packaged Zero is now threatening to drag the new game's Steam review average down into Mixed territory.

Of the nearly 1,000 negative reviews as of this writing, there's a spread of complaints—performance, multiplayer connectivity, weightier flight physics, lacking HOTAS support—but one of the most common refrains is that the Zero bonus has been something of a bait-and-switch. As summarized by reviewer Akion:

"Before I bought the game the description said this about the Belkan War DLC Half the reason I got the game was because that is my all time fav. The integrity of this should matter to you.

"'*ACE COMBAT ZERO: THE BELKAN WAR is a standalone application separate from the ACE COMBAT 8: WINGS OF THEVE full game.'

"And now it says this:

"*ACE COMBAT ZERO: THE BELKAN WAR (Steam Version) is available as a DLC executable launched separately from the base game."

So why does this matter?

As another reviewer points out, "You need to install 140gb of data to play a 3gb 20 years old game." That's not an exaggeration: Ace Combat 8 is currently using up 140.24 gigs on my SSD, with the Zero DLC making up only 3.2GB of that total.

There's also a clear strain of dissatisfaction with the fact that on consoles, Zero comes along with its own achievements/trophies list, which is absent for the Steam version. Since it's tied to AC8, Zero also lacks separate cloud saves, something that would be pretty nice for playing it on the Steam Deck, where it should run very well given its age.

The lack of parity with the console version clearly has quite a few players annoyed, but the principle of the thing seems to be the biggest sticking point. The store page said Zero would be a separate app, and then it wasn't.

Some folks have countered that Bandai Namco couldn't release Zero as its own package due to the licensing terms of the planes featured in the game, seemingly based on this livestream featuring the developers on September 28 saying the licenses only permit giving the game away for free. But clearly it's possible to package the game separately on consoles even without selling it, and I'm skeptical it's impossible to find some kind of workaround on Steam. Hitman 3 still has a listing page, for example, despite not being purchasable by itself—it's listed as downloadable content for Hitman World of Assassination.

"Notice: At the request of the publisher, HITMAN 3 Access Pass: HITMAN 1 Complete First Season is unlisted on the Steam store and will not appear in search," reads a warning message at the top of the store page.

This criticism will soon show up in a smaller portion of Ace Combat 8's reviews as players who didn't bother to pre-order begin to filter into the reaction pool, but for now the pre-order bonus isn't quite the clean publicity win it otherwise could've been.

What happened:

• I Am Your Beast and Clickolding developer Strange Scaffold has announced Something is Wrong with the Library, a tidying game set in "a library haunted by blood and meat."
• Players will both reshelve the library's scattered books with a real Dewey Decimal System implementation while facing the accursed energies emanating from the corpses its billionaire patron entombed in the walls.
• A "Modular Chaos System" ensures each playthrough has its own set of books to reorganize.
• There's at least one terrifying, inscrutable obelisk.
• Something is Wrong with the Library is set to launch before the end of 2026.

It's nice to see some aura in the tidying space

In recent months, Steam has been awash in organize-em-ups and haystack searchers—games capitalizing on the simple pleasure of arranging a chaotic melange into an orderly assemblage. But while I'd never think to deny the meditative appeal of putting a thing in a place, I'm not going to pretend that the emergent craze isn't undergoing the same gold rush of low-effort iteration that we saw during the rash of retail sims. Visit the page for the "Organizing" tag, and you'll find as many generic variations on "Tidy up the [X]" with AI logos as the Unity and Unreal asset stores can sustain.

With Strange Scaffold at the helm, however, I feel safe betting that Something is Wrong with the Library's haunted meat labyrinth will be able to offer a sense of identity and direction that's been hard to find among the available tidying sims. After all, this is a developer that was able to use clicking as a medium for a harrowing test of psychosexual fortitude. And then made a match-three metroidvania-slash-survival-sim-parody set in a weirdo dinosaur mansion. And then an isekai-enabling driving sim.

If nothing else, we can trust the studio to commit to the bit. It seems incapable—possibly pathologically—of doing otherwise.

I picked up sewing as a hobby in college and, in my youthful cockiness, skipped the pillowcase-and-curtain phase. Instead, I dove right into extravagant cosplay and Renaissance dresses, and most of my first attempts were not fit for the human body. I'm sure some of that hubris came from watching my grandmother make sewing look so easy when, in reality, it requires patience, precision, and a blood sacrifice.

But that's what leads me to Dressmaker—a cute and cozy hardcore seamstress sim that respects the technical skill the craft demands while making it accessible to the inexperienced. It's genuinely a good place to start if you're mildly interested in the craft but have never even picked up a needle.

You'll learn some seamstress lingo, make tons of cute dresses, and spend significantly less money than I did ruining countless bolts of expensive fabric. I've played a little over six hours, and I've already got a growing list of finely stitched details I adore.

• Papers, Please: Your decisions as a shopkeeper have an effect. If you spot someone on the wanted list coming in to your store, you can report them to security for a reward—but keep in mind they may be members of the resistance fighting against the cruel dystopia.
• Dredd: With customers coming in from the poverty-stricken lower levels and the wealthier upper levels, it feels like I'm working a shop in one of those massive towers from the Dredd movie, even though in the game's fiction I'm actually working in a cyberpunk space-station.
• Strange Horticulture: My favorite shopkeeping sim is a fascinating detective game in disguise, and there's a hint of it in Probably Stolen. You can use a magnifying glass to examine items to see if they're counterfeit and as an appraisal tool to judge their value, a bit like how you search for clues to ID plants in Strange Horticulture. Fun!
• Fallout: You can farm rats for their meat, set up water purification systems, scavenge through trash heaps to find scrap to turn into sellable goods... I know I'm on a space station, but it feels a little like I'm trying to earn a buck in the wasteland.
• Recettear: An Item Shop's Tale: You're not just selling, you're buying, and you can haggle over prices, sometimes at the expense of your reputation. Demand in the station can change: if a mining expedition docks, for example, you can expect the prices of metals to plunge as supply explodes. You'll have to be nimble and keep your eye on what's valuable from day to day.
• Cyberpunk 2077: I probably don't need to elaborate on this one, choombatta.

Just a quick alert to let you know about the excellent demo for Probably Stolen, where you're a cyberpunk shopkeeper who can deal in rat meat, counterfeit cigarettes, and contraband weapons. Check it out on Steam. The sim is due out next month but the demo is surprisingly hefty!

What happened

• On September 22, Studio Minus—the developer of ragdoll torture sandbox People Playground—posted a Steam news update saying it had disabled the game's Steam Workshop support "due to yet another malicious mod," which the developer said was "especially bad"
• The developer has since released a new build of People Playground that prevents mods from running, saying on its Discord that Workshop and mod support will only return if "mods are fundamentally safe to run," which "might never happen"
• A redditor who inspected the malicious Workshop mod claims the mod hijacks the user's Steam account to republish itself on the Workshop along with the user's personal information, destroys files for People Playground and other installed Steam games, and sends slurs to the user's Steam friends
• Despite initially claiming that the infected mod "does not steal passwords" or other credentials, Studio Minus is urging People Playground users to change their Discord passwords as soon as possible, as well as "the passwords of all your important accounts"
• This is People Playground's second Workshop malware incident this year: Studio Minus previously disabled the game's Workshop after a similar outbreak of self-replicating malware in February

At this point, you're playing People Playground at your own risk

And Studio Minus seems to agree. In a blog post from last month, zooi—People Playground's primary developer—called the game a "technical hellscape," which makes it "technically impossible to properly sandbox mods in their current state." While the studio claims that the software has measures in place to prevent malicious mod activity, zooi called it a "hopeless arms race."

When a studio's developer doesn't seem confident about the viability of their project's security measures, you shouldn't either. But while People Playground seems like a particularly intense dumpster fire, 2026 hasn't inspired much confidence in the safety of Steam Workshop overall.

Since the year started, we've seen at least three other cases of Workshop-related malware incidents: first malicious Project Zomboid mods surfaced in April, then virus-laden anime ladies emerged in Wallpaper Engine in June, and then malware-infested Meccha Chameleon maps led to its Discord getting hacked in July.

It's the double-edged sword of Steam's relatively open publishing platform. We get to enjoy a bounty of games from small developers—but they can't always offer the same expectation of software security as their more established counterparts.

There aren't many things I'm looking forward to more than Nightdive's remaster of the original Thief. Is this because I'm old? Yes, but it's also because Thief is an all-timer videogame and I can't wait to see Nightdive work its magic on it.

At the PC Gaming Show Tokyo Direct, we got a hefty helping of gameplay in a new trailer for the remaster, showing Garrett stuffing his pockets in all manner of poorly-lit, erm, manors, plus sewers, crypts, tombs, dungeons—you know, Thief places.

It looks great. I've said it before and I'll say it again: Nightdive has gotten 'making games look how you remember them looking' down to an art. No ugly uprezzed textures or weirdly 3D models, just a lot of deft touches that keep the game's original art style intact.

I'm also pretty sure we got a glimpse of the game's controller support, and it looks like the remaster is taking its cues from the (very good) already-existing controller mods for the un-remastered version of the game.

You can select items using a handy-dandy inventory wheel, which is just how the mods do it. Supposing Nightdive is taking its cues from them, then I imagine we're going to get something that feels almost weirdly well-suited to a PC game originally released in 1998. The mods make flicking between your various arrow types (remember to stock up on water and moss, hard-mode players) very easy indeed—I imagine Nightdive's version might make it even easier.

I'll say this, though: this trailer, with its driving soundtrack and ample blackjacking and arrow-loosing, doesn't quite match my memories of Thief, which mostly consist of a lot of standing in dark corners and memorising guard patrol routes. Where's the 10-hour 'Hanging out with Garrett behind a door' ASMR YouTube video, Nightdive? You're missing a trick there.

Check out our Steam page with all the games, updates, and DLC featured in the show. You can also read our roundup article which includes every announcement from the show in the order it aired.

It felt like the moment I turned 30, all my friends got really into climbing. Pretty perfect timing then that Cairn came out when it did. We scored it 91% in our review, which is high praise indeed. For all of you impatient climbers who have been awaiting the delayed DLC, I bring good news: its new release date is only one month away.

As revealed at the PC Gaming Show Tokyo Direct, the first of Cairn's free On the Trail DLC, Deep Water, will be launching October 29, 2026. This is the first of a series of free planned expansions that'll add more to the climbing game.

Included in Deep Water are three new climbing locations. There are the cliffs of Haken and Ruka, where you'll be ascending overhangs above water. The wet stuff isn't just pretty to look at, though, it adds a bit of a safety net. Now, you can fall in without any consequences, so don't worry if your free solo adventure doesn't go quite as planned.

The Lamu climbing gym is also being added, so you can test your skills in a safer environment. I know that's not as thrilling for some of you, but others like to keep their bones in one piece.

As well as the new locations, you can also play as Marco to see how his climbing style differs from Aava's. You can also take either of them out in a new free roam mode. This will allow you to reload saves on the mountain and practise, or try out new routes.

If you've not given Cairn a go yet, it's currently 20% off on Steam until September 26, so now is a good chance to grab it before the DLC drops next month. If you've already got it, then you can wishlist the DLC right here.

Check out our Steam page with all the games, updates, and DLC featured in the show. You can also read our roundup article which includes every announcement from the show in the order it aired.

I'll tell you this right now: I'm already into what Lemonade Apocalypse is putting down. Let me count the reasons. First, its PC Gaming Show Tokyo Direct reveal trailer is a very fun satire of both billionaires and the billionaire-mindset podcast industrial complex. Two, I think it will tickle the parts of my brain that just loves turning numbers into bigger numbers. Three, a lemonade apocalypse sounds much preferable to the various other apocalypses that feel like they're on table for us right now.

Lemonade Apocalypse is actually a sequel to, uh, a game also called Lemonade Apocalypse. Stay with me. It's an economic simulator and incremental game where the task is to sell your soul for the dollar as quickly and dramatically as you can. "Build recipes from 70+ ingredients with evolving properties," the game's Steam page implores you, right before it also implores you to "Fuel climate change" and "Manipulate voters."

You will go from flogging your lemonade at a street-side stand to running an all-conquering, lemon-based capitalist empire. Morals have gone out the window. Friendships have been sacrificed on the altar of profit. All human feeling has been suspended. There is only you, the lemon, and the dollar.

So think something a little Cookie Clicker-esque, perhaps, but with more savage disdain for the ruling class and a bit more active involvement on your part. On top of "a genuine supply and demand engine" for your lemonade empire, the game also boasts "a presidential campaign," more than 70 ingredients to experiment with, and "a political opponent who has been watching everything you do." Which is assuredly fine.

This is hitting my fatigue with actual reality and my love of big numbers in all the right ways, folks. I'm pretty eager to get my hands on my competitors and grind them to dust under my boot. Also, to get my hands on some lemons and grind them into lemonade. Do you grind lemons? Probably. Check it out on Steam to find out.

Check out our Steam page with all the games, updates, and DLC featured in the show. You can also read our roundup article which includes every announcement from the show in the order it aired.

The best thing about machines is that they do exactly what you program them to do. The worst thing about machines is that they do exactly what you program them to do. In Brave New Wonders, releasing tomorrow, you'll build and automate a set of production lines with robots that you can write commands for.

In the launch trailer revealed at the PC Gaming Show Tokyo Direct, we hear a narrator say, "Give the command, and the machines will answer." The commands you give can be written in plain English—or any language the game supports—and the game will ensure the machines obey your whims.

Rather than setting up a supply chain manually, simply tell your robots to pick up ammo from the warehouse and deliver it to the turret. The game seems to detect which words in your command are the keywords, such as locations and conditions. So, if a little lost robot wanders off and simply circles an area instead of going into it and grabbing what you need, you'll have no one but yourself to blame.

Brave New Wonders touts itself as "part factory builder, part city builder rising from the ruins." It's set in a post-apocalyptic far future, so it'll be up to you to mine and refine all the resources you'll need to rebuild society.

That's no small feat, so you won't just be building one factory or base. You'll be able to link factories and logistics routes, and even copy and paste huge layouts. Even the robots keep their instructions when you paste more, so you won't have to micromanage too much—unless you want to change their commands, of course.

This isn't a peaceful factory sim, however. Old machines roam the decimated world, so you'll need to place turrets to defend your bases and use dynamite to clear the way forward if you hope to expand into new territories.

Brave New Wonders launches tomorrow, September 21, and you can pick it up on Steam.

Check out our Steam page with all the games, updates, and DLC featured in the show. You can also read our roundup article which includes every announcement from the show in the order it aired.

It's finally time, everyone. It's time to figure out the answer to the world's oldest question: what's wrong with Miami? Quarantine Zone—the game about manning a checkpoint and making sure plague-bearers don't sneak their way through—just showed off its new, Miami-based expansion at the PC Gaming Show Tokyo Direct.

Quarantine Zone: Miami Zero will arrive as a free update for the original game and it's coming on October 15, bringing you to the titular, and entirely new, location. It's sun-drenched, absurd, and comes with a whole gamut of new ailments and symptoms for Florida's unlucky inhabitants to suffer from and smuggle past you. Or attempt to smuggle past you, anyway. Your job is making sure they don't.

I must say, "making sure they don't" involves more hitting people with hammers than I'd expect for a job manning a quarantine checkpoint, but I guess that's why I do this for a living and not that.

You don't just hit people with hammers. That's one tool in your diagnostic arsenal. In the trailer, we see our brave quarantine officer catching poorly people using X-rays, some kind of scanner stick, and more. They also just start tasing people at one point, which feels like it must violate at least a couple constitutional rights. Do they have those in Miami?

Perhaps the most tantalising part of the trailer—and hats off to the marketing genius who came up with it—is its giant, bold, green-text exhortation to "Experiment with drugs." In addition to rooting out Miami's secretly sick at the checkpoint, it looks like you'll also be working long hours in the lab to… do something?

It all looked very involved, and featured a number of different scientific instruments. The end result seemed to be turning a normal-looking guy into the Incredible Hulk, though, which I think can't have been the original intention. Could it?

Check out our Steam page with all the games, updates, and DLC featured in the show. You can also read our roundup article which includes every announcement from the show in the order it aired.

Since you're reading this, chances are you've thought to yourself, "I bet I could make a better game than that." I know I have. Well, the chance for us to prove it will soon be upon us: MMORPG-maker Let's Build a Dungeon is launching into early access February 11, 2027.

In Let's Build a Dungeon, you take control of a videogame studio making an MMORPG. It's part management simulator, part RPG maker. You'll be in charge of hiring devs, assigning work, even deciding if you'll make your workers crunch or not.

But on top of the management side of things, you also have to actually make the game. You'll be in charge of what goes where, from enemies to roads to shops, everything, the world is your oyster. You can play the game yourself, but it doesn't stop there.

The problem is, if you make a stinker of an MMO, your in-game players are going to make sure you know. They'll leave reviews, criticising features and asking for others. In the trailer shown at the PC Gaming Show Tokyo Direct 2026, we see people clamouring for more environments, bosses, quests, better servers. God, do us players really sound that needy and ungrateful? I guess we do.

We'll get to find out what it's like on the other side of the aisle February 11, 2027, when Let's Build a Dungeon launches in early access. Maybe it'll make you think twice before posting a mean review. Do I find it kind of ironic that a game all about running a videogame studio is launching in early access? Yes. Yes, I do. It's all very meta. Wishlist the game on Steam.

Check out our Steam page with all the games, updates, and DLC featured in the show. You can also read our roundup article which includes every announcement from the show in the order it aired.

Thief: The Dark Project is one of my favourite games of all time, so naturally I am very excited for Nightdive Studios' upcoming remaster. But it's worth remembering that the community has done an excellent job supporting the game up until this point. Unofficial patches like NewDark have ensured that Looking Glass' landmark stealth game can run in high resolutions and with support for widescreen and multi-monitor functionalities for years.

And if you thought an official remaster might put such efforts to an end, think again. Not to be outdone by Nightdive, NewDark's creator just released a new version of the overhaul which adds a radical new feature to Thief—dynamic shadows.

Yep, NewDark 1.29 introduces hardware transform & lighting to the Dark Engine, opening the door to GPU-driven effects in Thief's gothic steampunk world. Chief among these is shadows that are dynamically cast by moving objects—guards, doors, even master thief Garrett himself.

The update's effects were shown off in a YouTube video by Thief fan Bentraxx. What's particularly interesting about the dynamic shadows is not simply how they look, but the mechanical implications they have. For example, opening a door between a dark corridor and a well-lit room will flood the former with light, while opened doors create shadows behind them, meaning Garrett can now use doors to create hiding spaces.

At the same time, shadows can be a hindrance to Garrett too. According to Bentraxx, NPCs now have awareness of Garrett's shadow, meaning they might spot him skulking round a corner if he happens to be backlit. In the comments of his video, Bentraxx further notes that Garrett's shadow only appears humanoid in Thief 2. In The Dark Project, it is "a simple box shape", due to the fact that Garrett's character model in the first game is a six-foot tall, poo-coloured cuboid.

Alongside dynamic shadows, Newdark 1.29 also adds dynamic reflections, normal mapping and parallax occlusion mapping, options for SSAO, volumetric light and fog, underwater effects, various post-processing effects, light luminance adaptation, and light streaks. As noted by several replies to Bentraxx's video, these features bring Thief much closer in line to the lighting seen in Deadly shadows and the Doom 3 total conversion The Dark Mod.

The download link for NewDark 1.29 is available via the TTLG forums, though for some reason, it's currently only available for premium members of storage.to. And although this is all impressive, I should make it clear that none of it makes Nightdive's remaster redundant.

Nightdive's overhauls focus on preserving the original without introducing radical changes, adding things like improved models and animations, but nothing that would fundamentally alter how the game plays (which NewDark's dynamic lighting does). Nightdive also recently revealed that the remaster will bring a new campaign called The Tarnished Mirror, which is undoubtedly what I am most excited about. Nightdive hasn't said anything further on the new campaign yet, but no doubt we'll find out more in the coming weeks and months.

When Dishonored director Harvey Smith announced his new studio, co-founded with fellow Arkane alumnus Ben Horne, there was no doubt about the kind of games the team would be working on. Black Pony Immersive makes it pretty damn clear. Putting "immersive" right there in the name is a statement of intent, one born from his experiences at Arkane.

"About the name, it was important to us to say to people, 'This is who we are,'" he tells us. "When Arkane was independent, very often people would come in and they would try to get us to work on some game. And it was not in our wheelhouse. People would say, 'You're going to go under by turning your nose up at this project, this is money. People are offering you money to do this DLC for this third-person action shooter.' And we'd be like, 'Yeah, but we don't like those, we don't know how to do them well, and we think that game is dumb.' So we're generally going to say no, and gamble the company on doing what we are known for."

Of course, Arkane would go on to become the best known immersive sim studio. And that was because it stood its ground and kept working in that genre, despite the offers to work on projects outside of its wheelhouse.

"There was a moment when we were pretty desperate," he recalls, "and a bunch of people working on BioShock 2 said, 'God, we're in a tight spot. We need somebody to take on a bunch of levels. You know who's out there? Arkane is out there.' So we took Christophe Carrier, our level design director, who is really the wellspring from which a lot of that thinking at Arkane came from. We took him and a team, and we put them on BioShock 2. And that only happened because people associated us with that kind of game."

Ultimately, though, after spending decades building this reputation, Arkane would eventually take a detour—the vampire-themed multiplayer shooter Redfall. "That was a game that did not come together," says Smith, "and we never should have been working on a games-as-a-service game, to be honest."

Redfall led to the closure of Arkane Austin, and thanks to the Xbox reset Arkane Lyon is now in a difficult position. French labour laws meant that Xbox couldn't just massacre it like it's already done with so many other studios, but it feels like it's hanging on by a thread.

As an independent studio with such a clear creative direction, Black Pony Immersive will hopefully avoid this fate. "There are many ways to fail, many ways to succeed," says Smith. "But specialising in what you love is not a bad path."

Should immersive sims rein in their budgets? Thief designer Randy Smith Evergreen thinks so. We chatted to him last month, where he said a lot of immersive sims are in "an arms race to be bigger and better", but that the size of the audience and the competition from other big-budget games makes that a mistake.

His friend and fellow immersive sim designer, Harvey Smith, has a different take. Smith recently left Arkane, where he was the director of Dishonored and co-director of Redfall, to set up his own studio, Black Pony Immersive.

"I go to get coffee with Randy every now and then, and I'm overdue for that," Smith tells us. "He's a vegetarian, so sometimes we hit one of these hippy places where you can get a black bean burger or whatever. So next time I'm together with him, I'm just gonna berate him for this opinion."

He doesn't really disagree, though. Smith Evergreen's main point was that big budgets can become a noose, one which inflates publisher expectations and forces niche immersive sims to compete with games boasting mass market appeal. And Smith agrees, you have to "right-size your budget".

Where they differ is that Smith thinks you can make an immersive sim at any scale. "If you wanted it to be a really short, two-hour experience that just models your mom's house, you could do that. On the other hand, I feel like when it really comes together and it has the budget behind it and the team, you end up with Deus Ex or Dishonored or BioShock."

But he acknowledges the risks that come with designing a money guzzler. "The thing you can't do is get so conservative with your features—which often happens the more money you're spending—[that] you don't do the fun, creative, interesting parts of immersive sims."

The blockbuster game space is undeniably conservative at times. Just look at the disappointing Wolverine, which won't be coming to PC and is no great loss. The whole experience leads you by the nose, and all you get to do is bash a single button. This simplicity is bad enough in an action game; for an immersive sim, it would be fatal.

Big imsims need to have a foundation that lets players go nuts. Smith offers an example from Deus Ex, where he was lead designer. Players used magnetic mines to climb walls, hopping onto one, then grabbing the one beneath them and placing it higher. This wasn't something the team had planned for.

"That's just one example," he adds, "but tactically speaking, as you're playing, emergent stuff happens all the time because the AI is very analogue and fully featured, and the physics is very analogue and fully featured. And then you combine that with creative tools and a player who can often hide and observe, and it just becomes this explosion of possible permutations."

For smaller projects, the same risks exist, however. "The person making their own attempt at an immersive sim might not have the sort of people-power, budget or time to make a system that has as much range and as much depth," Smith says.

"If you just end up with some cool art and music and nice notes here and there, and you can walk through it, it's kind of adjacent to an immersive sim, but it's missing that sweet spot. So I think you could equally argue that going the other direction, and I know Randy knows this, you could lack some of the things that make immersive sims special."

Ultimately, the two designers end up with similar conclusions. That there is a sweet spot. "If you go too small or too large, then yeah, you probably lose something," says Smith. "Maybe it's just a nuanced way of saying what Randy Smith Evergreen, my friend, said. I love Randy. I've known him for a long time."

Certainly, Dishonored benefited from having a big budget, and Dishonored 2 remains the greatest immersive sim ever made. But the industry has changed. A slew of huge acquisitions have given a tiny handful of publishers all the power, and while they will gladly throw vast sums of money at a project, that comes with daunting expectations—when they aren't met, the studios suffer.

This is what happened to Smith and Arkane. Dishonored 2 was a critical darling but failed to meet Bethesda's expectations. That happened again with Prey. And then Arkane Austin ended up making Redfall, a multiplayer shooter, that resulted in the studio's death. Arkane Lyon, meanwhile, faces an uncertain future following the "Xbox reset".

So while I do get excited when I see an immersive sim that's clearly been given a lot of resources, I'm always looking for the strings.

Smith's new venture, Black Pony Immersive, hasn't announced its first game yet, but I'm excited to see what he and his team are capable of when they don't have a publisher holding an axe to their necks.

Randy Smith Evergreen had been pitching the idea to publishers for a while. A nonviolent RPG, in tune with the values of the Looking Glass games he worked on in the '90s, about a young American woman searching for her missing uncle in Eastern Europe. But the publishers weren't biting. Often, they'd tell Smith Evergreen how much they loved the Thief games he'd helped create, and he'd wonder whether he ought to just be pitching a spiritual sequel to that seminal stealth series instead. Then, Microsoft came calling.

"I had an opportunity to work for a studio that's funded by Xbox, and this was at the lowest moment of our funding search when we were like, 'This might not happen, we might need to think about how to feed our families in some other way,'" he recalls. "I held on—and ultimately it worked out, and I got a funded project—in part because I felt hesitancy to work for a Microsoft-backed studio. I think the recent layoffs at Microsoft showed that there was some wisdom to that decision."

I ask Smith Evergreen where that hesitance came from—positing that it might have something to do with the fact that, over the decades, Microsoft's track record for keeping its studios open has been terrible. "You just answered your own question," he says.

By the time Smith Evergreen had to make a decision about whether or not to join Xbox, the company had already shut down Arkane Austin—the Dishonored and Prey studio that, more than any other, had stuck fast to the design principles, worldbuilding and atmosphere of the Thief series.

"I have countless friends who worked there or were the founders of that studio, and I know the high quality work that they're capable of. And I know the tortured history of their last project, Redfall, and that its shortcomings weren't entirely attributable to the talent in that studio," Smith Evergreen says. "For Xbox to buy them and quickly turn around in a few short years and dissolve the studio was just a pretty dire warning. To take a job with them at that point would be like walking past some stakes with human skulls on them and being like, 'It's probably fine in this direction'. And I did consider it, but I'm glad I ultimately went a different way."

Smith Evergreen has "snatched success out of the jaws of defeat", starting a new indie studio to work on an unannounced game that evolved out of the original nonviolent RPG pitch. "It's quite different [from] that, but some of the mechanics of that larger indie RPG were prototyped, and I took what I thought would be most commercially successful and reduced its scope and changed quite a few other things as well," he says. "And that was the new pitch that ultimately got accepted. With some additional modifications that were proposed by the party who ultimately funded us."

Smith Evergreen's team jumped through a lot of hoops in order to loosen those purse-strings. "But where it is today is great," he says. "In our case, it did not torture the concept out of desirability. It's still a great game that we're excited to work on."

A common trait among game developers is that they thrive under limitations. Many of the medium's innovators have found ways to squeeze magic through a tight funnel of technical and financial constraint, all while navigating the demands of their paymasters. But Smith Evergreen points out that not all interference is welcome or productive.

"There have definitely been places in my past where I have felt unwanted pressure from people who are backing the game," he says. "I think a notable version would be the Thief series, especially Thief: Deadly Shadows, where Eidos pushed us to make the game a little bit more violent than we were interested in. And you can debate whether that's good feedback or not.

"My own inclinations might have been to rely so heavily on stealth when I was working on the Thief series that it wasn't accessible to a larger market, and their feedback maybe made it more accessible. But at the same time, I think there was a little bit of, at that time, Eidos wresting some creative control away from us."

It was Eidos who decided the game should be given the slightly daft and bloodythirsty subtitle of Deadly Shadows. "They didn't want to call it Thief 3, and they did want to emphasise murder," Smith Evergreen says. "The team wasn't stoked about that."

Thief: Deadly Shadows was a AAA game at the time of its release in 2004. But since then, the costs involved in making a AAA game that meets modern standards have increased exponentially. So much so that Smith Evergreen believes the immersive sim genre no longer belongs in big-budget development.

"I think that might be a misread for the audience size of immersive sims," he says. "It's hard to know how an immersive sim wins the arms race to be AAA. And so there's quite a few of us who think the way to do it is to be triple-I or double-A. You're talking about a three to $10 million budget, a passionate team led by some of us folks who have experience with immersive sims or newcomers who just played a lot of immersive sims and really get it, making something that's a little bit smaller, more passionate, and can be made more quickly."

The original Thief cost $3 million to make, Smith Evergreen points out. "If you did it today, it'd be cheaper because the engine already exists, and you can get a lot of the functionality you need by just buying it off the shelf in the asset store. It's not clear to me that the $3 million dollar Thief or Deus Ex-style immersive sim that you want to make in the year 2026 is going to sell disproportionately different amounts of copies than the $250 million one that comes out of a big studio."

There's only so much fidelity a game needs, Smith Evergreen suggests. "I remember back when Alan Wake came out I was like, 'OK, this is relatively photoreal. I know 100% that I'm on a ferry. I see that I'm going to an island in the Pacific Northwest. Why do we need more graphics?' Because they just cost a ton."

So long as the graphics are good enough to establish an imsim's vibe, says Smith Evergreen, that's enough. "Would fans prefer an immersive sim that's got insane spectacle and gorgeous, Arkane Lyon quality vistas? Of course they would like that. But is it necessary for what you come to an immersive sim for? I'm not as sure that it is."

Smith Evergreen is not the only respected game director I've interviewed recently who has struggled to get a new game off the ground. Dragon Age's David Gaider is fighting to fund a heist game. And Wyrdsong, the new RPG from Fallout 76 lead Jeff Gardiner, appears to have suffered death by a thousand cuts during the publisher pitching process.

The games market has been flooded with content. As a consequence, Smith Evergreen says, "There are games that people work on that just never get touched." Venture capitalists hunt for AI-driven projects and games-as-a-service to throw their money behind, while indies jostle to break through with dazzling new ideas.

"It's kind of like the same evolutionary path that the dinosaurs went through after the asteroid hit the Earth," Smith Evergreen says. "They went from an arms race to be bigger, to an arms race to be smaller. And if you think about it, a lot of the most beloved indie games of the past 30 years—FTL is one example—were cheap, fast, scrappy games that just had amazing gameplay and great creativity. And that's the direction I think a lot of things need to go right now, especially in the indie space. They need to be high quality content, but not enormous production budgets."

Meanwhile, Thief is going through something of a belated renaissance. A Nightdive-enabled remaster of The Dark Project is coming, alongside a graphic novel set between the first two games in the series. "I used to say that when I played Dishonored, that was the first time I ever played a Thief game," Smith Evergreen says. "Because when you make a game, it's just a totally different experience than having played it. But I feel like enough time has passed since I played Thief: The Dark Project that playing a remaster will probably feel very fresh for me. So I'm excited about that."

If Smith Evergreen were ever to be tempted back to the Thief series in the future, it would be to work on a low-budget entry that bypassed "spending infinite amounts of time on fancy art and rendering problems".

"I would not be super excited to work on a huge budget immersive sim Thief reboot," he says. "Working on these giant budget games, there's a little bit of torture to it. It takes a very long time. It's a messy process, especially for immersive sims—it can be very hard to find their centre before the moment when you need to build tonnes of content. There are just all these reasons that, as a creator, it's a challenging thing. And challenges are awesome, but not all challenges are equally fun."

Thief designer Randy Smith Evergreen, whose credits also include System Shock 2 and Dark Messiah: Might and Magic, thinks we need to chill out a bit when it comes to the graphics arms race. Specifically, he reckons the race should have stopped 16 years ago with Remedy's original Alan Wake, which was quite a looker back in the day.

"The fidelity curve has become outrageous," he tells us. "It's become a spectacle. I remember it was back when Alan Wake came out that I was like, 'OK, this is relatively photoreal. I know 100% that I'm on a ferry. I see that I'm going to an island in the Pacific Northwest. Why do we need more graphics? Because they just cost a ton.' Like, I 100% get it. I've got the vibe. It feels right. What else are we doing with this graphics arms race, right?"

Quite a lot, it turns out. We had high-fidelity graphics that started to look photorealistic, but then it was decided that we needed lighting and reflections that looked better than real. None of these things serve any practical purpose, but now most games that fall into the AA or AAA categories feature ray tracing and a level of fidelity that's not possible to enjoy—for most people—without DLSS trickery.

"I think Alan Wake is the fidelity everything needs to hit," Smith Evergreen says. "Where you're like, I get it. I get your vibes. I can tell what's going on, and everything beyond that is gravy. Sure, would they prefer an immersive sim that's got insane spectacle and gorgeous, Arkane Lyon quality vistas? Of course they would like that. But is it necessary for what you came to an immersive sim for? I'm not sure that it is."

This ties into his belief that immersive sims shouldn't be aiming for AAA—it misunderstands the audience, specifically its size. It's just not enough to justify the cost of development.

"Like, you just have to put more money into them," he said, "make a bigger and bigger immersive sim. It's got to be AAA, and it's got to sit right alongside Ghost of Tsushima and GTA 6, and it's got to be a game that's big enough to capture the public's attention. But I think that might be a misread for the audience size of immersive sims."

This isn't something anybody is worrying about when it comes to Nightdive's Thief remaster, which won't be yassifying The City. Judging by what we've seen so far, it looks a lot like OG Thief.

"I'm super excited to see what they do," says Smith Evergreen. "Nightdive seems very trustworthy, very passionate. I like the projects they're choosing to remaster. So that's going to be awesome to see.

"In a way, I'll be able to play Thief again for the first time ever. I used to say that when I played Dishonored, that was the first time I ever played a Thief game. Because when you make a game, it's just a totally different experience than having played it. But I feel like enough time has passed since I played Thief: The Dark Project that playing a remaster will probably feel very fresh for me. So I'm excited about that."

We're still waiting for a Thief: The Dark Project Remastered release date, and Smith Evergreen is currently working on a project that "you'll hear about before too long".

I stand knee deep in a pool of blood. My coworkers, comrades, and friends reduced to food for soulless seagulls. Give a man a fish and he may throw it into the mouth of a weary fisherman, teach a man to fish and he shall bring forth the wrath of a spider crab who knows how to stab you with your own knife.

This was my first experience of the new multiplayer fishing sim How to Fish, and now I can see why it's become such a smash-hit among bored and borderline-violent friendship groups.

It has a simple premise: you and up to four friends learn how to fish. The cycle of work involves players buying a basic fishing rod at first, using this to fish for, well, fish and crustaceans, and then beating them to death with your bare knuckles, brass knuckles, or a sick-ass knife. Then you can either eat your catch to replenish your health or hunger or feed it to the lonely fisherman who sits on the dock and gives you coins for your hard work. With the coins gained you can buy more tools to help you fish, like the knife, or supplies which you can use as bait, such as half a hotdog or an empty beer can.

Some things can be so simple that they become boring, and perhaps How to Fish would fall into this trap, if it were just you and the fisherman on the island. But when you add three other people with nothing to lose and access to friendly-fire things get out of hand in the best way possible.

When I first woke up on the island I was met with a friendly punch in the face. From there on out my friends and I worked together, with the occasional fist fight, to get our cash stash up and resources ready to defeat our first boss. Oh, remember when I mentioned health? Yeah, this is why you need it.

It turns out that using a beer can to fish will lure in a monstrous spider crab who emerges from the ocean with the rage that only the pinnacle of evolution could muster. The music changes and a massive health bar appears—you must defeat this beast to continue to learn how to fish.

Our first encounter happened almost by accident and we were not prepared for it. We were cut down in no time at all and I was even carried off by a hungry seagull. The second fight went a little better, however.

I decided to take a back foot, as I was the only person armed with brass knuckles and not a knife. Instead of fighting I acted as medic, slapping my deceased comrades back to life while screaming "You don't get to give up on me yet!" Their poor dazed bodies would then stumble back into the fight against the spider crab, powered by my inspirational words and some clams.

I did get a couple of punches in while the crab was dazed. We even managed to get its health bar down past the final third. But alas, we failed and my body was once again flown off into the sunset by a hungry seagull.

We unfortunately had to cut our session short, but that hasn't stopped us from planning our next hunt. I certainly can't wait to get stuck back in to beat some fish and maybe learn how to actually fish. I've also got my eye on that sniper rifle and other guns which you can buy later on—that'll show that damn spider crab.

Dwarf Fortress is a bit like caviar—not everyone's into it, but the people who are describe their experiences with it like a religious awakening. But the game's legendary co-creator, Tarn Adams, didn't even know it had a big following for many years, as he told PC Gamer's Joshua Wolens in a recent interview at Gamescom 2026.

"It was something like 2013, 14 going to PAX in Seattle, and there was an indie mega booth at the time," he recalled. "Of the 50 or 60 games showing—I'm just ballparking the numbers—I'd say 90% of them heard of Dwarf Fortress and also had a very amazed fan-ish response when we were introduced." It was the first time Adams had a sense of the reach his game had.

"That was crazy. I had no idea that that was even a thing," he said. "I didn't know that anyone knew what we were doing … I enjoyed it. I try to have good moral qualities, but I am apparently subject to all forms of flattery and vanity and such."

If he could go back in time, he wouldn't so as to not "disrupt the timeline." But if he could go back in time without the risk of doing that, Adams said might encourage himself to evangelize the game more in its early days.

"I'd probably be more confident about getting it in front of more people sooner. I didn't know the people to do that, though, so I'd probably be like 'go meet some more people,'" he said. "From the release of the game in 2006 up until around 2013, I didn't meet another game developer really, except maybe one or two at a Dwarf Fortress-specific meet."

He acknowledges things would have felt different then—he was "not a very social person" at the time and DF was "outsider art" for a number of years—but he'd still give his old self "a little pep talk, maybe."

Back in 2024, Dwarf Fortress surpassed 800,000 copies sold, more than enough to make co-creators Tarn and Zach Adams millionaires. At the time, they seemed to be taking a low-key approach to the whole mo' money thing, so when our intrepid interviewer Joshua Wolens caught up with Tarn at the recently-concluded Gamescom, he asked if, in the two years since then, he'd more fully adjusted to his moneybags lifestyle.

"I guess, a bit, somewhat. I mean, I live in a house now. I have two cats and they love me. That's great," Adams said. But, he added, it wasn't so much a matter of adjusting to something new as just going back to what used to be.

"I know what [having a house] is like because I grew up in one. It's basically like reattaining what my parents could do with regular jobs. It's not much different from that.

"And I don't want to make light of financial security, right? I mean, yeah, that's great, and I wish it for everybody."

It makes me think of Adams' comments in the same interview about enabling a more equitable, better-functioning society in Dwarf Fortress by walling up dwarf nobles and abolishing the dwarf cops: He said there's perhaps a "lesson to be had there." Maybe there's a lesson to be had here, too.

Anyway, lest anyone worry about the impact Adams' growing wealth—Dwarf Fortress moved past one million copies sold in 2025—will have on the game's future, that too is moving ahead pretty much unchanged. "We're back to work," Adams said. "I mean, work is the same."

That work is presumably Dwarf Fortress' first Myth and Magic update, slated to go live later this year, which aims to bring magic to the game. And not just the occasional fireball or invisibility cloak: Adams said in August that the goal is to create "procedural magic systems that go beyond combining spell effects and go straight to the fundamental cosmological makeup of the universe." Sounds very Dwarf Fortress indeed.

While the legendarily absurd complexity of fantasy simulation in Dwarf Fortress provides countless perils for dwarven health and happiness, the most insidious threat to any mountain hall emerges from within: Nobility. As a fortress flourishes, it inevitably draws the attention of dwarven nobles who'll needlessly burrow themselves into industries they're not qualified to direct, issuing demands and mandates without any care for whether they're feasible or sustainable for the rest of dwarven society.

Sounds unbelievable, I know. But there are ways to manage the menace of nobility—like simply walling them off in an unpleasant corner and letting the rest of society live and labor as need, preference, urgency, and capacity allow. In an interview with PC Gamer at Gamescom, Dwarf Fortress co-creator Tarn Adams said there's probably something to be learned there.

While Dwarf Fortress players are capable of pursuing an egalitarian dwarven utopia, its simulation doesn't yet allow for dwarves to engage with their own political ideologies—but Adams said he's had distant plans to tinker with systems for simulating civilizational law codes, and maybe even some light philosophizing.

"I think there's room for that kind of thing. I read a lot of law codes. I read the Tang law code; Romans had crazy case law, and I read a bunch of commentary about that," Adams said. "And it's funny: I'm not sure about the modern systems, but in the Chinese one from 1500 years ago, it says 'if you know it's wrong, but it doesn't fit any of these 2000 things, you just hit the person 10 times and call it a day.'"

It might sound like a crude solution, but for Adams, it helped conceptualize how a simulated law code could essentially function like a series of conditional statements in computer logic.

"That feels like an if-else-if-else-if-else-if-else-if—and then you have your final else in your law code. I'm like, I can do that. I can model a civil law code," Adams said. "When it comes to case law, it's more like a scripting language that interprets itself. I can also get into that now, especially now that [Dwarf Fortress shifted to Lua scripting.] That's more complicated, but manageable—as long as you've got safety valves to escape from a process completely breaking down."

As for what that process breaking down might look like, Adams said you might have "cases being argued by dwarven barristers or something, but then there'd be a dog wandering into the room and biting somebody. And then the goblins come."

I suspect surviving Roman law commentaries don't offer any immediate guidance there, but in the meantime, Adams reckons that Dwarf Fortress can offer some political wisdom of its own. Being able to literally wall off the vampiric influence of a mountain hall's noble class, he said, is a model that might offer some applicable lessons in our own lives.

"Has been tried, could be tried again. It sounds like a fine thing. Where'd EA go? It's been purchased. So we didn't have a wall there," Adams said. "Could have had a wall. Didn't have a wall. Many such cases. It seems simple to me, but it's complicated somehow."

Adams did note, however—as I have in my own experiments with ushering in dwarven utopia by subjecting the upper classes to a living entombment—that the influence of dwarven nobility persists as long as there's an armed entity enforcing its demands. If your fortress maintains a fortress guard, they'll continue punishing dwarves who violate mandates issued by nobles under wall-arrest.

"If you put a wall around your nobles, their demands still somehow get into the aether, and your little justice system will still punish dwarves," Adams said. "So you have to not assign a captain of the guard either. Another lesson to be had there."

Even as it achieved fractal complexity by modeling feline alcohol poisoning, procedural bardic traditions, and delayed-action thermonuclear dragonfire, the creators of Dwarf Fortress—brothers Tarn and Zach Adams—have maintained still loftier ambitions for their legendary fantasy sim. In recent years, that effort has been guided by one goal: Creating magic. Procedural, universe-specific magic, interwoven with the unique cosmology and mythology modeled each and every time the player clicks the "create new world" button.

For a game already setting a high watermark for simulational complexity, it's an incredibly ambitious endeavor. It's taken a decade. It's demanded a complete scripting overhaul, aided by the heroic effort of Bay 12 Games' first non-Adams programmer, Putnam. Frankly, it's almost strained believability. And it'll start arriving later this year.

Tarn Adams confirmed that Dwarf Fortress' first Myth and Magic update will land before the end of the year in a short announcement video, embedded above. His goal, he said, was to create "procedural magic systems that go beyond combining spell effects and go straight to the fundamental cosmological makeup of the universe," enabling "magical situations that you cannot get any other way."

It's a notion Adams has shared as we've spoken with him about his Myth and Magic ambitions for years, going back to 2017: By creating a procedural magic system that's anchored in each world's simulated creation myths and cosmology, Dwarf Fortress wouldn't be generating world variations to play within—it would become an entire fantasy setting generator, conjuring universes with their own emergent forms of magic that are as fundamental as the laws of physics.

"It allows your fortresses to pursue magical research that really feels like discovery," Adams said.

The video's only a minute long, but it offers some brief glimpses of magic-wielding Dwarf Fortress figures in action. A goblin sorcerer teleports a cat. A dwarven wizard repeatedly force-blasts a goblin against a wall. A six-limbed fly-man holding four magic staves macerates a hallway full of enemy swordsmen with a barrage of magic bolts. A necromancer curses an innocent dwarf with undeath, and they immediately start ripping into a nearby comrade.

"Now, there will be magic effects that blow things up," Adams said over footage of magic projectiles rapidly demolishing a dwarven outpost. "There will also be magic materials, enchantments, rituals, ancient ruins, everything that makes fantasy what it is—with every piece tied into how that particular universe works, down to its bones."

The first Myth and Magic update doesn't have a release date yet, but Dwarf Fortress is available on Steam and at the Bay 12 Games site now.

In retrospect I'm not sure why it's taken Lego so long to get here: it's made Star Wars games, Batman games... heck, it even made a Fortnite game. The first Lego videogame came out way back in 1997! But at least it's finally fulfilling its destiny: The blocks that were made for building are finally being used for a city builder.

Lego Skylines was announced today at Gamescom, and yeah, it looks like the perfect fit. Have a peek at the announcement below.

Looks pretty darn sweet, and looks pretty darn big. But how big?

"In this cozy, customizable city-builder, your choices bring a living, breathing metropolis to life. Go big or obsess over the details. No instructions. Unlimited space," says Paradox Interactive.

Unlimited space? For real? That's... quite a lot of space.

"Remember hospitals to keep everyone in one piece. Wheat for the donuts. Garbage collection, sewage, and all the other things that make a city click. How it all comes together is up to you. So stick your hard hat on and greenlight your weirdest ideas," Paradox says.

Sounds good to me: it's got that Lego cuteness but also looks like it doesn't skimp on the complexity.

Interestingly, Lego Skylines is being developed by Iceflake Studios, who were brought in to take over for Colossal Order on the trouble-plagued sim Cities: Skylines 2 in late 2025.

There's no release date yet, but you can visit the Lego Skylines store page on Steam.

Last we saw RollerCoaster Tycoon it was in pretty rickety shape. More than a decade after RollerCoaster Tycoon 3, the follow-up—RollerCoaster Tycoon World—was released to nearly universally poor reviews. Even worse, in one of the all-time dick moves in gaming industry history, Atari launched it exactly one day before rival sim Planet Coaster was due to release. Wow.

That nakedly desperate move didn't help the situation: people liked Planet Coaster and didn't like RCTW, and the Tycoon series has been silent since 2016.

But here we are another decade later, and Atari is bringing RollerCoaster Tycoon back yet again—though I'm already feeling more hopeful about it.

RollerCoaster Tycoon Wonderworks was announced at Gamescom today, and the new entry in the series is being developed by Springloaded, who made the excellent Let's Build a Zoo and is releasing promising gamedev sim Let's Build a Dungeon later this year. That's a good sign, at least!

There's no date on the books yet, and RollerCoaster Tycoon Wonderworks won't be feature complete when it launches: it's "coming soon" to early access. Here's its page on Steam.

You've boiled the Spider Crab and turned the Giant Piranha into fish sticks, but How to Fish amps up the difficulty with the Pufferfish boss battle. The bloated, buoy-like fish took me quite a few tries in both solo and co-op saves, but I finally skewered it with a few SMG upgrades and plenty of patience. Lots of patience.

Find endangered fish to exchange for Pufferfish bait

The third How to Fish island gives you a few quest options, but start by speaking to the swimmer in yellow floaties to begin the quest for Pufferfish bait. The swimmer asks for help slaying a scary fish and has spare snacks you can use as bait, though he won't hand them over until you promise something in return.

Why do we need to pay him back for tools to complete a task he asked us to do? Who knows, but you'll have to find an endangered creature before he'll agree to hand over the snacks.

1. Buy the new Fishing Rod ($3) and Standard Lure ($15)
2. Use the Standard Lure to fish for endangered creatures
3. Do not sell your fish before examining them with the F key
4. Feed the fish labeled 'endangered creature' to the swimmer
5. He'll give you the Carrot Bait. Use it to reel in the Pufferfish

Pufferfish boss

The Pufferfish is tough, and it took me quite a few runs before I was finally able to fillet this guy. I never managed to kill it with the shotgun, and had to splurge on the SMG (with a few upgrades) to get the job done.

Tips for beating the Pufferfish

• Eat enough seafood to top off your health and hunger
• Buy the SMG ($650) and as many bullet upgrades as you can afford
• Don't stop moving, and keep putting as many trees between yourself and the boss as possible
• Be careful not to run into its purple poison; the damage tick turns lethal fast
• The Pufferfish grows in size as you hit it, but I found it easier to watch and dodge when it lost a little under half its HP
• Try luring the Pufferfish around the merchant's shop; it sometimes gets stuck cutting corners when it's bigger

Because I can't stress it enough: Do not stop to shoot the Pufferfish. I only managed to stand still a few times when the boss got stuck on trees while trying to chase me around the shop, but you'll want to keep running and gunning until it's dead. Lucky players can skirt by the previous two boss fights, but if you haven't learned to make the most of the terrain in How to Fish, you will now.

How to Fish is the name of the game, but defeating its first crabby encounter is less about your rod-and-reel technique and more about your skills with a pair of brass knuckles. I'll spare you the play-by-play, but I ran at the Spider Crab boss multiple times with my bare fists, and it didn't end well. To spare you from the same trial and error, here's how to beat the Spider Crab in How to Fish.

Finish the tutorial quest for the sailor

Your first goal in How to Fish is to complete the sailor's tutorial quest for a quick and easy onboarding, but being the goody-two-shoes student I am, I'm recommending you follow a few extra steps to prepare for the Spider Crab boss:

1. Feed clams (~$3 per clam) to the sailor; don't buy his beer immediately
2. Buy the Fishing Rod ($3) first
3. Feed the sailor more clams or fish with the rod for more profitable meals
4. Make a Knife ($45) or Brass Knuckles ($24) your second purchase
5. Buy the Beer ($12), and have the sailor drink it
6. Take the empty can and use it as bait to reel in the Spider Crab boss

The Spider Crab boss

So there I was, fistfighting a giant, somersaulting crab with my bare hands, as you do, before it knocked me out twice. It's a bit of a pain, since every time you lose, you'll have to gather enough cash to buy another beer. I'm too familiar with that grind, but I graduated from university almost a decade ago, and I'm over that life.

The Spider Crab is easy to take down using only your fists in co-op play, but my first How to Fish run was solo, so that's why I recommend buying a melee weapon—preferably the knife. The added damage is a game-changer, so with that in mind, heed my few extra tips before reeling the critter in again:

Tips for beating the Spider Crab

• Eat enough seafood before the fight to top off your hunger and health
• The Spider Crab hurts; 3-4 quick hits from its claws are enough to kill you
• If it's giving you trouble, grab a clam or two for quick healing
• After the crab attacks, it's temporarily dazed. Run in and hit it while it's incapacitated
• When you're done, I recommend buying the radar ($10) before leaving the island

Even the easiest difficulty takes a few rounds of running in and out for a few hits while the Spider Crab is dizzy, so be patient. How to Fish is all about using the terrain and physics to your advantage, so make the boss work harder to reach you and dodge its attacks while standing on higher ground.

When you're done, pick up what remains of the Spider Crab and deliver it to the old fisherman. He'll give you the keys to his boat as a thank you. Now it's time to load up and head to the next island, where piranha is on the menu. After that, you're off to face the Pufferfish boss.

It's great that the Nightdive's remasters of various classic games give new players a chance to experience them without having to download a fixpack or remap the controls or whatever, but it's also a wonderful opportunity to learn new things about games we already loved. Each remaster is like an archaeological dig and you never know what it'll uncover.

Digital Foundry's Retro Super Show recently spoke to Nightdive about their current project, a remaster of Looking Glass Studio's stealth classic Thief: The Dark Project. If you're super into the minutiae of Thief, as you should be, it's a cool insight into the making of the game and in particular, the Dark Engine that powers it.

"There's nothing else like it even to this day," Nightdive software engineer Josh Dowell says. As he explains, Nightdive got to see "a very very out of date version control repository for the code" and that made the origins of its oddities more obvious. "I think most of the team that started working on it ended up moving to different projects and then a whole new team took over for the second half of its development," he says.

Back when it was just called The Dark Project, Thief was shown at E3 in 1997 and scheduled to release later that year. It was soon delayed to 1998, and the comments Dowell found attached to the code make it seem like it was still being worked on as it made its way out the door. "My favorite part of looking through the Dark Engine codebase has been seeing all of the evil comments left by the Looking Glass programmers, all the Irrational Games programmers," he says. "They have the wildest things to say about the code. They say, 'Giant hack! Never replicate this! We must fix this before '98!' 'E3 hack! We must delete this after E3!' and it's still there in Thief 2."

In spite of the rushed and hacky nature of its creation, the Dark Engine is capable of incredible things—feats of audio, AI, and lighting that were cutting edge at the time and still able to impress. Part of why it works so well is what it shares with another boundary-pushing first-person game of the era. "Thief's rendering and the technology that powers the levels is, shockingly, it's just a clone of Quake," Dowell says. "The way they do the lighting is exactly the same as Quake, the way that lights are turned on and off is Quake."

While he calls the two "eerily similar" the differences are "really strange" as he puts it. "So Quake uses precomputed visibility, which means when you're standing in a specific part of the level, the game knows exactly what you can see without having to think about it. But the Dark Engine does what Quake does during its precomputing phase, where every single doorway in the world is a portal and it sees the rectangle that you can see through that portal and it sees the other portals to other doors and then calculates which ones you can see through that portal. It's super complicated and I can't believe they got away with it in '98, because it's doing the thing that Quake did that was slow, but it's doing it in real-time."

Thief: The Dark Project Remastered doesn't have a release date yet, but you can follow its progress on Steam, GOG, and Epic.

As someone who grew up in the '90s and early '00s when cooperative games were scarce, I'm a big fan of this current era of cheap and cheerful experiences designed to play with your pals. I enjoy the knockabout vibe of games like Peak and R.E.P.O, which acknowledge their disposable nature and focus on being daft fun, rather than trying to engage you for weeks and months at a time as more traditional live service games do.

As such, I'm always interested to see what coop nonsense bubbles up to the surface of Steam's algorithm next, and it appears the catch of the day is How to Fish, a 1-4 player, physics- based fishing simulator where you subdue marine life in the most preposterous ways possible.

The premise sees you and your pals stranded on a barren, rocky islet after drunkenly crashing onto its shores. You start out surviving by picking clams off the rock, which offer meagre sustenance. But you can also sell the clams for cash (which you do in the traditional manner of shoving them into an old seadog's mouth) thus enabling you to buy a fishing rod and begin learning how to fish.

The fishing element of How to Fish is relatively simple. You cast your line, wait for something to take the bait, and reel it in. Where things get more interesting is in what happens after you've landed a catch. Fish and other sea-life don't just appear in your inventory: they will flop, scuttle or crawl around land until you've subdued them. Initially, you'll have naught but your fists to do this with, but over time, you can purchase knives, brass knuckles and even guns you can use to exact bloody vengeance on the ocean.

Needless to say, How to Fish is an extremely silly game. Not only can you launch fish into the air and then blast them with firearms Ridiculous Fishing-style, but you can also barbecue seagulls and gamble your catch on slot machines and roulette wheels for a chance to gain even more cash. That said, How to Fish isn't wholly devoid of structure. There are NPCs dotted about who will assign you quests, and there's a chain of islands to explore that yield bigger, more dangerous catches like sharks and king crabs.

Although How to Fish is clearly built for cooperative shenanigans, it can also be played single player. It may be worth bearing this in mind for the moment, as while most Steam reviews are positive, some players are reporting lax security at the backend, with random players able to drop into the game and cause a ruckus.

Nonetheless, if sniping fish from the sky sounds like your idea of a good time, you can currently pick up How to Fish at an introductory discount that brings the price down from $8 (£6.40) to $5 (£4). That discount ends on August 27.

Of all the other professions I never quite saw myself as a hairdresser, I'm simply not creative enough and my small talk leaves a lot to be desired. But I was still willing to give it a go (within the safe confines of a game) because Beastfolk Barber just looked too cute to pass up on.

The premise is simple: you play as Freddie, the owner of a small hair salon in Starfish City, where beastfolk now outnumber humans. So to keep up with the changing clientele Freddie rebrands her shop for beastfolk and snags herself a magical mirror in hopes to beat out the competition who have more money and employees at their disposal.

Any time a client sits down ready to get their haircut you can use the mirror to show them up to three hairstyles they might like. This way they can get a feel for a hairdo before any hair is chopped off. But the catch is that they won't always tell you out right which one they prefer.

It's up to you to keep track of the options you present them with and their reactions to them and come to a decision of which one is best for them all on your own. You are a professional hairdresser after all.

Luckily, it's not too hard to suss people out, I managed to please every customer who walked through my doors, except for the owl. But he was really grumpy to begin with, and looking back on our interaction I think he may have been a really shady criminal. You see Beastfolk Barber has more going on than just cutting people's hair with little minigames, there's actually more to these beastfolk than just looks.

Everyone who sits in your chair has something going on in their personal life and they really want to talk to you about it. Whether it be chasing the guy of their dreams, getting ready for a tough family reunion, or trying to push legislation through the senate to try and adjust age limits for various beastfolk.

It was the two headed dragon, Luca and Varo, who brought this issue to my attention. Apparently every beastfolk has to go by human rules and regulations. For example, bees don't live long enough to access their government pension (which is 65 in this world; we should all be so lucky) and mice have to wait over half their lifespan for a chance to vote. But Luca and Varo are trying to change this, and apparently needed the help of a good haircut to get their confidence up to do so.

On top of all the civil unrest and shady business Freddie also has to deal with a demonic presence called Edi who's also in need of a haircut. But being a demon, their request is not exactly straight forward. First off, they won't pay in cash, instead opting to grant you one re-do, magically erasing or changing whatever thing you wish for. Secondly, in order to get a chance to cut your hair you need to prove your skill by acquiring 10 regulars.

Sadly, the demo ended before I could make any real headway with the task, but I think I did a good enough job for at least three clients to return to my salon. I guess the proof will be in the pudding once the full game releases in the future, hopefully we won't have to wait too long.

Who knew that all I needed to be interested in maths was the command of a five-thousand-tonne artillery battery? If only my secondary school teacher had known. Turns out that the ticking clock of exam pressure is a hell of a lot less motivating than the boom of artillery shells growing ever closer, as enemy gunners zero in on your position, trying to blast your colossal cannon to kingdom come before you blast them.

But Iron Nest is a similar test of mental endurance. You hustle and bustle around your one-man gun emplacement, triangulating degrees and distances, praying that your next shell won't sail clear over the target, leaving you scratching your head as to which vital calculation you somehow messed up. It's an addictive challenge, bolstered by an array of wonderfully tactile gadgets and gizmos.

Sure, it's an artillery simulator of sorts, but it's more like one of the best versions of Battleship you'll ever play. It's about employing what you know and learn through each shot, to track down enemy positions and destroy them, often before they destroy you. And it's hard to overstate how satisfying it is when you land a shell squarely on-target, especially when it's the payoff for 10 minutes spent messing about with machines and the map table.

So who are you shooting? Iron Nest is set in an alternate-history version of 1920s Spain in which the Kingdom of Castille somehow got their mitts on a gigantic, city-levelling artillery piece, and is using it to maintain their grip on power. That's your job: shoot what they tell you, or don't. As the lone denizen of the Iron Nest—save for one presumably very deaf cat—your job is operating the battery by yourself.

You have to read the teleprinter and note vital info on your clipboard. Use the map table and its elaborate array of markers, coloured pencils, and measuring devices to triangulate enemy positions. You're inputting distance into a machine to calculate barrel elevation—which prints off a cute little firing solution ticket—and wheeling the gun on its axis to the correct bearing. And when all is prepared, you're flipping switches as if arming a nuke, and pulling the handle to take your shot.

There's just so much gravity to Iron Nest, in fiddling with its tactile machines and gadgets, but also in the methodical step-by-step that plays out before each firing. Despite its initial learning curve, before long the pre-shot preparation becomes a bit of a ritual, a process you refine through repetition and gain mastery with. It's strangely relaxing for a game about obliterating enemies with explosive ordinance.

And as you gain mastery, Iron Nest unfurls even more. You unlock a requisition station that lets you satisfyingly stamp cards to buy recon plane flyovers and different shell types, but it adds a new challenge element as you try to spend the bare minimum possible, relying on your smarts to bridge the gap.

You unlock a second barrel, letting you hit two targets in quick succession. That is, if you're able to juggle the calculations and tweaks you have to make between each firing without your brain melting. And thanks to its procedurally-generated objectives, with enemy positions changing even when you replay a mission, plus medals for performance, Iron Nest has a lot of replayability.

The only thing Iron Nest is lacking is co-op. It would be ridiculously fun to run around the gun with a friend, but the game is already a pretty incredible achievement considering it was built by one guy. It'd also be a great candidate for a Battleship-style head-to-head mode, where two Iron Nests try to triangulate and destroy each other, but multiplayer is listed on Iron Nest's roadmap, so maybe we can look forward to that.

While I'm still very excited for next year's PVKK: Planetenverteidigungskanonenkommandant (*gasps for breath*) which sees you commanding an orbital defense cannon, Iron Nest is the best artillery game I've tried. Its playground of tactile tools and machines makes it appealing to simulation fans of all stripes, while its smart yet addictive gun-prep formula feels almost endlessly repeatable, as you strive for that elusive yet satisfying sense of mastery.

They called me mad. They called me a fool. They said it would never work—my plan to build a business empire on the back of a single department store that sold three products and a silica mine. 'You'll bankrupt us!' yelled red-faced advisors. They saw no wisdom in my schemes at all.

And, frankly, they were correct. Fair play to those guys. We are very bankrupt now.

In 1995, PC Gamer—a monthly journal about computer videogames—issued a challenge to me directly. In the back-most pages of its 22nd issue, it tucked away an ad for a then-new game called Capitalism. "Good enough to make a convert out of Karl Marx himself," declared an included testimonial from, yes, PC Gamer.

I was too busy being two years old to issue a response at the time, but I'm here now. As one of surely very few people who have reviewed The Communist Manifesto for a major gaming publication, I want to know: can it convert me? And more than that, can it offer a way forward? Can I find, in Capitalism's thicket of spreadsheets and input/output graphs, a solution to our current economic and social malaise? I accept your challenge, PC Gamer from 31 years ago.

Business 101

Capitalism—the game—does not work how you expect it to. For one thing, it is not on Earth. It happens on a procedurally generated pseudo-region of an Earth-like planet that you can spin up with every new game. Want to build the free market on a vast archipelago? By all means. A single, contiguous mega-landmass? Absolutely. The only guarantee is that, wherever you end, it will look nothing like anywhere that exists on our actual planet. Also, cities like Lisbon and Chicago will end up a short walk from each other, much like they are in real life.

I admit it, my first forays into the world of business were jazz capitalism. I was hitting buttons and trying things out without much regard for structure or logic. I built a department store in central Jakarta (next to Tehran) and filled it literally, exclusively with laundry detergent.

This was not just an idle whim. Capitalism is big on the supply chain, strangely, which means all the goods you sell have to come from somewhere. The wares I flogged in my department store were sourced from Jakarta's nearest seaport, and the things Jakarta's nearest seaport had to offer were: cookies, leather shoes, and detergent.

Were the people of Jakarta buying my suds? Absolutely the hell not. They could not have been less interested, even as the industrious worker-bees in the department store's purchasing unit accumulated more and more boxes of the stuff, gradually entombing my sales team like soapy pharaohs. Supply outstripped demand by a gazillion to one. I was making negative $80,000 a year.

I asked myself what Warren Buffett would do and decided the answer was probably 'keep digging'. I added the cookies and leather shoes to my department store's roster of goods, turning it into one of those malls—you know the ones—that occupies an entire city block and sells three unrelated products. If this didn't work, the game was broken.

There was, happily, more demand for shoes and cookies than laundry detergent, which is to say I was selling literally any of the former at all. Nevertheless, my expansion into the footwear and baked goods sectors had taken its toll. In spite of the new products' relative success, I was now making negative $300,000 a year. I added an advertising unit to my department store and partnered with a TV station.

This either helped or, more likely, coincided with an entirely unrelated change in my fortunes, because my annual profit suddenly skyrocketed to an incredible negative $3,000 a year. I went mad. I built an R&D Centre, a silica mine, and a new department store—this one selling CPUs, silica (a lot of silica about) and raw tobacco. Had I maybe started trying to sell directly to consumers raw materials I was meant to use in my own factories? Perhaps! Was I nevertheless now making an utterly inexplicable $400,000 a year? Yes!

I could only ride the high for so long. Eventually, my objectively boneheaded business decisions ricocheted back on me. The people of Bombay (where my second store was) were not interested in bulk-purchasing silica or computer chips, the fools, and even adding another advertising department didn't help. Though I did partner that advertising department with a newspaper rather than a TV station, and we all know traditional media is dead.

It wasn't long before the game told me I was out of cash in a Nickelodeon font.

Visions of the past

Here is the serious part of the article: Capitalism is genuinely rather fascinating. It is the crystallised end of history—that moment of triumphalism that marked the end of the Cold War and the seemingly final victory of liberal capitalism over all its foes. Its pregenerated scenarios, which you can play if you don't want to generate a random world for yourself, have names like The Fall Of The Berlin Wall and The Emerging Dragon (yes, it's about China).

It presents a vision of the world where consciously-held ideology is dead. In its place, ranks of men with big phones and suits chasing ever-greater numbers on ever-bigger spreadsheets—the "dog-eat-dog" world of capitalism. Was it good? No. Was it bad? No.

Capitalism, the game, does not sing the praises of capitalism the concept, nor does it criticise it. In this way, it reflects its era: good and bad were outdated—relics of a past where people deluded themselves by believing in things rather than chasing stuff that is concrete and consumable. Humanity had outgrown the distractions of faith and ideology. Now it could focus on reality, and the accumulation of CPUs and silica.

And air conditioners.

The aircon king of Guangzhou

My attempts to grow fat from detergent sales had run aground, but I did learn some things in that first run. Mostly, what I learnt—right there at the end—was 'hey, there's a button here to change the price of the things I sell'.

So I tried again, and a little bit slower this time. In Guangzhou (near Adelaide), I erected a department store, sourcing a thick gush of shampoo from the nearby port and selling it to anyone with hair. Crucially, this time I paid attention to things like 'product quality' and 'price'. The shampoo I was sourcing was a fair bit better than what Guangzhou already had access to, and by fidgeting with the price-per-bottle I could establish a neat equilibrium between supply and demand. Before long, I was making a stable $500,000 in annual profit.

What else did the port have to offer? Air conditioners and jumpers ('sweaters,' to our American readers). Guangzhou was hot; I suspected aircon demand would be high. I funnelled them into my department store.

It turned out I could charge essentially whatever the hell I wanted and get away with it. Demand was too strong to deter. I was charging people double what I was paying for them and raking it in. Before long, I got the jumpers on the shelves too, which inscrutably sold like hot cakes at the same time as the aircon units did—though, in credit to Capitalism's simulation, I am fairly sure the aircons sold better in summer, and vice versa for jumpers.

I was making $5,000,000 a year. I built factories, invented the video camera, purchased my competitors' jeans and sold them at a mark-up. I even started mining silica again, just for old times' sake. Pretty quickly, I hit $12 million a year. Which is kind of where I left it, since it rapidly became apparent I was simply too good at Capitalism for anything funny to happen ever again.

I'll say this, though: I like Capitalism (the game) a lot. The depth of its simulation is impressive, and not even 'for a game from 1995'. It just is. Here is a toy free market that consistently surprised me with the things it took into account and the things it asked me to take into account, from climate and environment to the nitty-gritty of the supply chain to the precise number and arrangement of business units in each of my outlets. It's a good time, if you are also the kind of person who gets excited by a nice spreadsheet.

Total conversion

Am I converted? Well, apparently I could be making $12 million a year in shampoo sales, which is only slightly less than I make writing about videogames, so maybe. I'm certainly into the game itself, which is probably the most I can reasonably ask of an entertainment product from three decades ago.

Does Capitalism, with its sepia snapshot of a period of post-Cold War triumphalism, represent a way forward for our current period of deep societal stagnation? Well, not quite. Capitalism captures not so much actual things that happened so much as it captures a vibe that existed in a select few places during a vanishingly small period of time.

You couldn't return to it if you wanted to—it was a product of a historical juncture, and no amount of nostalgia for an era when things felt better (whether they actually were is perhaps debatable) will change that.

Unless, of course, you start selling air conditioners.

Immersive sims like Dishonored and Deus Ex are typically solo experiences, but there have been a few attempts to bring multiplayer into the mix over the years. System Shock 2 featured an infamously wonky cooperative mode that was overhauled in last year's Nightdive remaster, while Arkane's Deathloop successfully folded an invasion mechanic into its mix of roguelike running and emergent sneaking.

Now, you can also play Arkane's excellent System Shock successor Prey with a bunch of pals, albeit via unofficial means. The functionally named Prey Coop enables up to 16 players to explore the Talos 1 space station together. Prey Coop is the creation of modder null234, who says that the mod "syncs global campaign progression, story choices, enemy AI, inventory, doors, hazards and world physics across all connected clients."

Although Prey Coop is still in Alpha, null234 says the whole campaign is playable from start to finish on "both new and existing saves". Nonetheless, there are still plenty of kinks to be worked out, such as "visual rough edges, animation quirks, and potential bugs."

Frankly, 16 players running around Talos 1 sound like the worst thing ever, but playing Prey as a 2-player cooperative game should work fairly well. Prey is by far the most methodical and cerebral of Arkane's immersive sims, one that encourages ingenious use of gadgets and powers to get the drop on the Typhon menace. Working together with a partner to set up traps, fight enemies, and bypass obstacles seems like a viable way to play, especially if you compensate for the extra player by amping up the difficulty.

As far as I'm concerned though, you can play Prey with as many friends as you like, so long as you play it. Arkane's dark sci-fi adventure has gone up in my estimations every time I've played it, to the point where I believe it might just be the best thing Arkane has produced to date.

Sure, Dishonored 2 has the more galaxy-brained levels and flashier powers. But I think Prey's slower-burn has the greater satisfaction, especially over repeated playthroughs. It has the superior story, by some margin and the way Talos 1 unfolds as you explore is marvellous. I also find that treating the Typhon as puzzles to be solved rather than enemies to be fought also elevates the experience. There's tons of fun to be had in manipulating those inky aliens from the shadows using all the weird gizmos Prey furnishes you with.

Anyway, I'll step off the soapbox now. If you want to try Prey Coop for yourself, you can download it via Nexus Mods. Alongside the mod itself, you'll need to install Chairloader, forward the appropriate ports, and ensure all players have a legitimate copy of the game and the same DLL build of the mod. Full instructions are available here.

Randy Smith Evergreen knows his immersive sims. As a designer on Thief: The Dark Project, he was there in the early days of the genre, and his credits include System Shock 2 and Dark Messiah: Might and Magic, as well as Thief 2 and Thief: Deadly Shadows. And he's pretty sure the genre is a bad fit for the AAA arms race.

"Myself and other folks who have worked extensively on immersive sims have a belief that the immersive sims that are being created today and are slated for the future are, by and large, thinking of themselves as [existing] in an arms race to be bigger and better," he tells us.

And this has been happening for a while. Just think about how Deus Ex: Mankind Divided pushed things in terms of fidelity. Or how big and expensive Arkane's modern offerings have been. They aren't competing with other immersive sims, they're competing with blockbuster shooters and vast open-world games from the biggest studios out there.

"Like, you just have to put more money into them, make a bigger and bigger immersive sim. It's got to be AAA, and it's got to sit right alongside Ghost of Tsushima and GTA 6, and it's got to be a game that's big enough to capture the public's attention. But I think that might be a misread for the audience size of immersive sims."

It certainly hasn't worked out for Arkane. Dishonored 2 and Prey are landmark immersive sims, but they were backed by a big publisher, Bethesda, who has since been swallowed up by Microsoft. Both games were considered commercial disappointments because Bethesda's expectations were not realistic for what has always been a niche genre.

Arkane Austin was shut down by Microsoft in 2024, following the poor reception to co-op vampire shooter Redfall. Arkane Lyon, meanwhile, is still hanging on, but only because of French labour laws. Microsoft reportedly wants rid of it. If a studio of this pedigree, with this amount of talent and experience, can't fit the square peg into the round hole, what chance does anyone else have?

Smith Evergreen believes the solution is to not participate in the arms race, unless you make something aimed at the mass market rather than the comparatively tiny immersive sim audience. He uses the Arkham games as an example.

"I think that the company that does this most closely is [Rocksteady with] the Batman Arkham games. Those superhero games have a lot of variability and a lot of player expression in them now, right? But they are still button-mashing combat fighter games with a huge IP."

If you want to make a proper immersive sim, though, it shouldn't be AAA.

"There's quite a few of us who think the way to do it is to be III or AA," he says. "So you're talking about a $3-10 million budget, a passionate team led by some of us folks who have experience with immersive sims or newcomers who just played a lot of immersive sims and really get it, making something that's a little bit smaller, more passionate, and can be made more quickly."

As he notes, the original Thief had a $3 million budget—but that was back in the '90s and would be $6 million in today's bucks. Still, there are ways to make immersive sims on the cheap, using existing engines, for instance, or evoking the simpler but still very potent aesthetic and fidelity of the immersive sim OGs.

"So it's not clear to me that the $3 million dollar Thief or Deus Ex-style immersive sim that you want to make in the year 2026 is going to sell disproportionately different amounts of copies than the $250 million one that comes out of a big studio. So I think that's an example of where the market is going, or [where] I think it should go at least."

I think the sales disparity is probably larger. Arkane's Deathloop had shifted five million copies by 2023, while indie immersive sims—Heat Signature and Gloomwood, for example—rarely reach 500k. But they don't need to!

Heat Signature was considered a success, and designer (and former PC Gamer editor) Tom Francis went on to make the also-well-received Tactical Breach Wizards. With lower costs and without a big publisher's inflated expectations, immersive sims can reach profitability a lot faster, allowing the studio to survive.

Given how vibrant the indie side of the genre is, and how close to death the other side is looking, I think Randy Smith Evergreen might be onto something here.

Among the heavy hitters and viral sensations that make up Steam's most-wishlisted games, Over the Hill exudes "little engine that could" energy. It's a chill, vibey offroad driving game that puts you in an old school 4x4 and sets you off among some low-poly hills to get lost. You've got a grappling hook and the option to drive in a co-op pack with a few friends, which may be the key ingredient that's seen Over the Hill build up such a strong pre-release following.

It's accrued more than a million wishlists in the couple years since it was announced, and all those folks are about to be very happy. Over the Hill is out in just a couple months, developer Funselektor announced today.

October 14th, to be exact, is when the co-op driving game is venturing out on Steam. Console releases won't arrive until next year.

PC Gamer's Andy Chalk played a bit of the game earlier this year and came away very pleased with how it captured "the broader experience of being 'out there,' going slow, taking your time, and thinking about how you're going to get where you want rather than just hammering on it. You can hammer on it if you want, and it'll carry you for a while, and also likely put you over onto your roof when you're not expecting it."

Where racing games offer the thrill of speed, Over the Hill "lands closer to real love of the wheel: Funselektor understands it in a way that an awful lot of studios miss," Andy wrote.

You'll be able to see what so entranced him soon. Funselektor says the launch will include "several areas in the Canada and Alergia regions" with "more content and surprises than what we had in the demo. If you want to play it sooner, you'll have to get yourself over to Gamescom later this month.

North Dakota may not be the most dazzling US state, but it is one of them, and so it's set to be absorbed into American Truck Simulator's expansive recreation of life on the road. The game's next DLC will take players through the state's "sweeping prairies, fertile farmland, winding rivers, and endless skies."

I don't know that its skies are any more endless than, say, South Dakota's, but it's still a looker in the trailer (the advertising kind, not the big rig kind) that went up today. Predictably, it shows a bunch of semi trucks hauling supplies past herds of buffalo, small towns, and giant statues of fish. It brings the game up to a whopping 61 DLCs on Steam—though notably, a number of those are free and others are minor additions that come cheaper than the addition of a full-on state.

North Dakota will join the likes of Texas, New Mexico, Idaho, and so on when it launches, though no release date has been named yet. Folks on social media are getting their yuks in: "Easiest update ever," said Reddit user Fun_Assignment_269. "Do you even need to design any assets to replicate North Dakota?" A reply from logicalguest chimed in, "Lol! Empty field, CTRL-C, CTRL-V. DONE!"

Okay, maybe the new locale isn't that exciting. But this isn't some key-jangling, action-packed, high-octane thrill ride we're talking about, this is a game where you drive the speed limit and deliver goods while saying "10-4, good buddy" and blasting Jerry Reed so loud you get a noise citation. Perhaps the placid sights of one of America's most boring states is merely a blank canvas for the determined trucking roleplayer, the kind of player who sees a giant fiberglass catfish statue and says, 'Yes, this is what America is all about.'

Or maybe it's just boring anyway. I mean, this is North Dakota we're talking about.

I am lucky enough to live close to some of the best plant shops I've ever had the pleasure to stumble into. I love taking my sweet time strolling around each one, looking at all the plants, succulents, seeds, and pots—all of which I could never dream of fitting into my small apartment. The best part of all of them is just how jumbled they all are, brimming with all sorts of strange plants for customers to discover. So, with that in mind you'd think I'd also create a maze-oasis of foliage right? Wrong.

The plant shop I created in Leafy Corner, the management sim game which just released, is regimented and ordered. Pots are colour coded and plants are separated into strict categories. You see, the whimsy will be better accessible if everything is ordered clearly.

I only actually noticed just how precise my layout was after seeing some examples of what other players made. These actually remind me of my local shops, with huge plants branching out over the store, little pots tucked away in dark corners, and fairy lights and all sorts of wall hangings adorning any shred of free space. I'd like to make something resembling that, but unfortunately my brain seems to crave order more than cosiness.

Thankfully there's no right or wrong way to go about creating your shop in Leafy Corner. My store may not look quite as fanciful, but that didn't stop customers from arriving in droves and picking up plants faster than I could restock them. I ended up earning over 500 coins in just my first session—not a bad day's work if you ask me.

I then used this cash to buy more shelves to create more order for the new plants I got in stock such as orchids, cacti, and peperomia rosso, of course. The more customers you successfully serve the better your store rating gets. This means you'll have more money to buy decorations and ship in new kinds of plants.

There's also a bit of customer interaction involved—no don't run away, these guys are nice and fairly normal. Every now and then someone will walk into your shop and know they want to buy a plant, they just won't know which one (we've all been there). This is where you come in, to help them decide.

Usually they'll have some sort of criteria such as something hardy, small, or geometric—I told you there were some oddballs. If you manage to find them the correct plant they'll buy it straight away, if not then they'll leave your store an unhappy customer, so it's worth taking a look at your plant journal if you're unsure of the demands.

Leafy Corner really is a calm and cosy sim which you can enjoy at your own pace, as organised or chaotic as you want. There's loads of customisation and plants to unlock, but if you're still unsure whether it'll be for you then you can check out its short demo.

As a former delivery guy, I have a fondness in my heart for videogame trucking. It lets me reclaim the simple satisfaction of moving things where they need to go without needing to endure all the things that make people glad they aren't a delivery guy. There's just one problem: Videogame truckin' isn't weird enough.

My six years of delivery work regularly involved surreal events, uncanny individuals, inexplicable decisions, and plenty of other inscrutable encounters. Videogame trucking, however, tends to be extremely scrutable—until this week, when developer Code Wakers launched an open playtest for Prospice: Anomalous Logistics, an upcoming driving and delivery sim that's properly unfathomable.

In Prospice, you play as an employee of an organization that would be right at home alongside the SCP Foundation, the Federal Bureau of Control, and any number of groups poking around the Backrooms. Its mission is researching an island that is, in the Steam description's words, "cursed by the corpses of ever rotting gods" and "a place where the world itself unravels."

Unfortunately, the open wilderness isn't the best environment for researching the remains of a massive, mouldering entity—which is where you come in. As a logistics driver, you'll undertake delivery tasks to support and supply the expedition, all while managing fuel, navigating unpaved terrain, and ensuring whatever still-twitching godflesh might be weighing down your flatbed gets to its destination unspoiled. Ideally while surviving, which—in my experience—isn't guaranteed.

When I started the playtest, I was just yards away from a massive, partially unearthed corpse, its hand already removed and waiting in my truck for transport—cargo that was, I'll admit, somewhat eerier than anything I've delivered in person. But still, despite how much that severed hell-hand continually quivered throughout the drive, I completed my inaugural delivery with only some passing clumsiness as I juggled gears to get my truck over and through some swamp mud and rocky inclines.

When my next run went just as smoothly, it was almost enough to convince me that this Anomalous Logistics business might not be all that bad. Sure, the island's ambient energies mean my PDA's map can't always triangulate my position with full precision, and I did have to stop and wonder whether the hovering, bleeding orb I found in the woods would do anything untowards once I strapped it down for transport. But my first day of work was going great!

Until ten minutes later, when a huge, twitching monstrosity crawled out of the trees, and "dissolved" my "dreambody" in "the labyrinth of collective death," according to the text that briefly appeared on-screen following my immediate and grisly undoing. Which, I can confirm, is an accurate description of the actual deliveryman experience.

Prospice: Anomalous Logistics doesn't have a release date yet, but Code Wakers says its playtest on Steam "will remain open for approximately one week."

I've taken the train from Bath to London hundreds of times in my life, usually to visit game developers in the capital. And while I enjoy that part of the job, the journey itself is a dispiriting chore. Rail travel in the United Kingdom is a chaos of delays, overcrowding and overpriced sandwiches. Yet when I saw Great Western Express on Steam—an expansion for Train Sim World that recreates part of that route—I felt compelled to play it.

Videogames are a powerful form of escapism for me, but what if I played one to achieve the complete opposite of that?

Train Sim World is primarily a game about driving trains, but it's unique in that it also lets you be a passenger. Bath isn't in the game yet, so I start my journey at Reading, and I'm immediately struck by how well they've recreated the station. I've spent many hours here over the years waiting for delayed connections, and the layout is perfect—although the lack of a pasty shop is jarring. The pasty is a station staple, and I will forever associate the smell of hot pastry with waiting for trains.

On Rails

I get on the train and it's a strange experience seeing something so mundane, so tediously everyday, replicated in 3D. I walk through the buffet cart and can almost smell the reheated bacon rolls. I go through a vestibule and think of all the times I've sat there on my suitcase because the seats were all reserved. It's a bit too clean, though. It needs a layer of grime to capture the experience, and perhaps some men drinking tins of Carlsberg and shouting.

I find an empty seat, which knocks some serious points off for realism, and the train pulls out of Reading. Next stop, London Paddington. I watch the city give way to the countryside through the rain-spattered window, and I do what I always do when I'm on a train: I listen to a podcast and eat a Twirl.

And, honestly, it's just like the real thing, which I guess is the point of any simulator.

The grey, misty gloom streaking past me is authentically British, although it's far too quiet. No one is talking loudly and no children are screeching. I think about how much more atmospheric it would be if you heard the train manager piping up over the intercom. But these minor criticisms aside, it's the closest any game has ever come to replicating being a passenger on a train in the UK. So well done for that.

Knowing that there's no manager on duty, I relocate to first class. But it feels wrong. I don't belong here. I should be in standard with the "real" people. But before I head back down the train pulls into Paddington, which is as accurate as Reading. This isn't an experience I'll be hurrying back to, but it was an interesting one.

I'm fascinated by games that recreate real life, especially if they capture the rubbishness of the UK. And I'm glad sims like Train Sim World exist to fulfil this very specific desire.

In theory, I get the appeal of an NPC powered by AI. What if Lydia from Skyrim could talk endlessly about topics other than your shared inventory and could offer more than the handful of phrases she shouts at enemy bandits? I like the idea of characters whose interactions don't end just because their scripts do. It's an intriguing concept.

The reality is that, at least for now, the AI solutions I've seen just don't work all that well. AI-powered NPCs can talk and talk and talk, that's for damn sure. But they're not saying anything worth listening to.

A while back I took a modded, ChatGPT-powered companion with me to see if the AI that can supposedly pass college entrance exams could solve the easiest puzzle in Skyrim. Spoiler: it absolutely couldn't, and the AI character I spent all that time with wasn't a convincing conversationalist, either.

More recently I've been playing Seed, an MMO colony sim from Klang Games that just arrived in early access after about a decade of development. In Seed you create a seedling, a Sims-like avatar, who shares the world with other players' seedlings, and they can converse with you and each other using Seed's AI.

My first AI interaction was actually a good one. During character creation I noticed seedlings had very long legs and short torsos, making it look like my avatar was wearing extremely high-waisted jeans. I decided to ask him about it, and this is a game where you can absolutely ask whatever the heck you want. It's even kind of cute because when I type a question, he'll take a little phone out and text me back.

"Hi!! I'm so happy to finally be talking to you!" my seedling said.

"Hey why are you so high-waisted" I typed.

"Everyone's wearing high-waisted right now! It's like, totally the trend… Is it… not looking good?"

I mean, that's a great response. I couldn't ask a character in Skyrim or The Sims or any other game I can think of a question about the length of their legs or the equator of their jeans. It's genuinely fun that you can do that in Seed.

But it mostly went downhill from there, and after a few hours it became hard not to see this first amusing interaction as just dumb luck that Seed's AI stumbled on an entertaining response. The rest of my conversations with both my doppelganger, and between my seedling and other player-made seedlings, didn't go well at all.

Here's a conversation (seedlings use text bubbles, not spoken words) between my seedling and the seedling of another player. Note that neither I nor the other player are typing these messages, it's all Seed's AI. The other player wasn't even online: seedlings are persistent in the world even when the player isn't around.

"So, how old are you actually? Just curious!" my seedling asked another player's seedling.

"I am eighteen!! It feels like such a heavy, beautiful number," said the other seedling.

"Eighteen is such a classic age! Totally iconic," my seedling agreed.

"And what about you?" the seedling said. "How many years have you lived?"

"I am eighteen too!" my seedling gushed. "We are literally the same age, how cool is that?"

A few things immediately make me aware that this is AI, especially the practice of filling up space with completely unnecessary words. "I'm eighteen!" would have been an adequate response: you don't need to then give the age of eighteen a bunch of nonsensical attributes, such as that it's a "heavy, beautiful number" because, y'know, that doesn't make any damn sense.

"How many years have you lived?" is also another indication of an AI that's trying way too hard. You don't have to rephrase simple questions to sound alive! I'm alive, and I don't do that. "How old are you?" still does the trick.

Here's another typical conversation, in which I choose an AI-generated topic that is related to another seedling not having facial hair:

My seedling: "Wait, look at that person over there! Their skin looks so smooth and perfect!!" (A weird thing to say.)

Other seedling: "Smooth? No way. They look completely washed out to me." (A weird response.) "Actually, look at how clean that person's face is! No hair at all!" (Yeah, that's what my guy said originally.)

My seedling: "Haha, my bad! I guess I was just seeing what I wanted to see?" (What?) "Wait, I don't know… I actually think that look is super trendy right now!! Like the minimal vibe is totally in everywhere."

As conversations go, the topic (no facial hair) is odd, the observations don't sound natural, and the responses barely make sense. The player's seedling and mine have added some points to our relationship meter, but it came at the cost of a pointless conversation.

I'll say this much: the Seedlings do at least have detectable personality types that come from players' choices during character creation, and those come through in how the AI talks. Some speak very curtly, indicating they're not really interested in a conversation at all, others gush with enthusiasm about just about everything, some like to throw shade while others find it unappealing to gossip. The AI does make it easy to get a quick read on what type of character you're talking to, which is good!

But the conversations themselves aren't compelling and the immersion grows weaker the longer they go on. As you've seen, "heavy" and "beautiful" and "iconic" and "vibes" are words that come up constantly in Seed: not only is my seedling's age "heavy" and "beautiful" but so is my seedling's name and the color of someone's shoes (brown) in two other conversations. Everything, everywhere, is a vibe of some sort, even when it's clearly not. I find this sort of thing as unconvincing as any other AI chatbot I've ever spoken with. It's just filler, something that may briefly appear genuine but doesn't hold up to even a bit of scrutiny.

I do see some real benefits of AI here in Seed. I like that I can ask the tutorial robot questions about aspects of the game, like how to perform certain tasks or locate specific resources, and I'll usually (not always) get a helpful answer in plain English. But it can be vague and unhelpful at times, too, such as when it misunderstands the question or delivers unrelated information. Seed's AI doesn't appear to make up or hallucinate stuff the way most chatbots do, but while it's faster than consulting a wiki or asking another player for help, it's not always better.

Plus, Seed's tutorial bot will sometimes mispronounce a common word or stress the wrong syllable, and for me that's immediately immersion-breaking because these are the things I've learned to instinctively check for when trying to determine if something is AI or a real person, like in a narrated YouTube video. It's an uncanny valley thing, an instant reminder that you're seeing or hearing something fake.

Like I said, there are some good examples of the AI in Seed working as intended, and it may improve over time. But I'm still left wondering if these types of AI interactions are needed or even wanted in games. Are two avatars mindlessly blathering about how beautiful and heavy a number is really an improvement over The Sims, where two avatars can produce a word bubble containing a picture of a guitar, and we all immediately understand that they're talking about music? It's shallow, sure, but not as shallow as a full conversation that contains no actual meaning.

Even though I'm not loving this aspect of Seed, I can imagine the possibilities of talking to characters in a life sim or MMO or RPG and them never running out of things to say. But at the moment, I'll still take limited but compelling dialogue, or even simple word bubbles with emojis, over unconvincing AI wordslop.

The best games like Stardew Valley

Fields of Mistria

Why: The level of heart and detail in the character dialogue is unmatched.
Fields of Mistria review (89%): "Fields of Mistria is the first farm sim in a lot of years that's made me want to spend time in town visiting with its cast even after I've gotten rich off shipping berries to the capitol."

Now that it's finally hit 1.0 launch, Fields of Mistria has become my absolute favorite in the post-Stardew farm sim scene. It solves so many of my little farm simming gripes with little quality-of-life touches and just feels like it was made with a deep love of the genre.

Mistria commits so hard to the '90s anime and Game Boy Advance aesthetic with pastel colors and a bunch of loveable cuties wearing capes and shoulder pads. I can't stress enough that it's got so much substance underneath that anime trope crust, though. Yes March is a textbook tsundere and Juniper is a witch who says "OH HO HO" with that arrogant hand pose when she laughs, but it leaves room for characters who defy stereotypes too. Mistria has so much detailed, character-building dialogue that I truly love getting to know its characters and their quirks.

Roots of Pacha

Why: It's the best exactly-like-Stardew thing out there.
Roots of Pacha review (88%): "It delights in genre staples while tuning the formula just enough to create something engaging and fresh."

Roots of Pacha is by far the closest thing to just playing more Stardew Valley, only set in the stone age. It's got a very similar pixel art style, character types, and soundtrack, but unlike so many things chasing Stardew's coattails, Pacha doesn't suffer for the comparison. It's just downright solidly good with lots of little extras like a fun musical minigame to tame animals and a tech tree for unlocking farm inventions that your community members come up with. This is the one that I managed to get my group of friends to play together for several weeks despite never doing co-op Stardew together.

Coral Island

Why: It's one of the biggest farm sims around.

Coral Island has 30 dateable characters. That's wild, when the standard in this genre is more like a dozen. Plus there's all the other characters on Coral Island who are equally personable and unique. After a year of early access and several post-launch updates with massive feature lists, Coral Island become one of the biggest farm sims around. It's now got multiplayer, a (toggleable) jealousy system if you're neglecting your spouse for other datable characters, and even more big updates on the way. I came to this one later than most, but honestly I'm glad I'm seeing it after it's been jam-packed with goodies.

Moonstone Island

Why: Creature-catching will always be catnip to me.

Moonstone Island leans closer to a crafting sim than a life sim, so some of the community and social elements like dating are shallower than other Stardew-like games. I cannot deny that adding creature-catching to a genre I already like is pretty much always a winner, though. Moonstone brings my favorite bits of base-building and procedural generated world exploring into a cute farming and crafting experience the way that almost no other post-Stardew life sims do. It flips the usual farm sim priorities on its head by making my creatures the focus instead of my rows of crops.

Kynseed

Why: The world feels so enticingly creepy and interesting.
Kynseed review (89%): "A hearty stew made from elements of Fable, Stardew Valley and Crusader Kings."

Kynseed is an odd one, in general and also, admittedly, as a pick for this list. It's built on creepy fairytales with an almost overwhelmingly dense pixel art style. Having roots in the older Fable games means that its characters are more like simulated little pawns in your world and don't really have that Stardew-style depth of character. It has gotten more Stardew-like with all of its years of updates though, adding in more house customization and other personal bits. Playing through family generations means you can sink a lot of hours into this one. I'm always fascinated by the crusty old RPG feeling of exploring its world and chasing down quest hints between coming home to farm. I just think it's very niche and neat and I'll never shut up about how its super tactile cooking minigame is amazing.

Staff Picks

I may be the tender of this list, but I'm just one of many farm sim lovers at PC Gamer, so here are the latest favorites from my fellows.

Readers' choice

Upcoming games like Stardew Valley

Steam's Train Fest has begun, meaning there are deals and discounts on lots of locomotive games, even a few brand new ones like Denshattack, which just came out this month.

But when it comes to train sales there's really only one train game to talk about: Train Simulator Classic, a sim notable for having a whole heck of a lot of DLC: a whopping 790 pieces of it. And almost all of them are on sale this week.

The base game itself has a hefty discount of 80% off, making it a steal at only $6 during Train Fest. And just about all of that tasty DLC is also on sale for anywhere between 20-80% off as well.

I won't lie, it was trickier than I expected to figure out what the total discount on all that DLC adds up to. Perhaps for obvious reasons, Steam won't let me add all 790 pieces of DLC to my cart at once, but it also won't just display all of the pieces of DLC in a nice tallyable column, truncating the list to a mere 200 entries. Even the ever-useful Augmented Steam extension won't let me fill my cart with every piece of DLC in one fell swoop. Come on, storefronts, let me accidentally purchase thousands of dollars of software!

But after a lot of clicking and scrolling and digging and copy-pasting raw data into a spreadsheet, and then doing a whole bunch of organizing and sorting and calculating, I think I've worked it all out.

Typically, in order to buy every piece of DLC for Train Simulator Classic at full price, you'd have to spend about $15,621.11. With today's Train Fest Sale where most of it is heavily discounted, it'll only run you about $6,933.80. That's about 55% off! Not bad if you have thousands of dollars to blow on train DLC—or, far more likely, if you've been looking for one or two bits of DLC and were waiting for a sale.

If you're looking for some train games that won't cost as much as a used car, there's quite a few to choose from, even newer games. As with Denshattack, train survival game EverRail also came out this month and is 10% off, and Fogpiercer, a strategy game also released this July, is 25% off.

No deals on Monster Train 1 or 2, however. Feels like a missed opportunity. Here's the Train Fest trailer with more sales to choo-choo-choose from:

The country I lead is in uproar. My approval ratings have to be measured on entirely new axes to capture popular upset. The streets throng with rebels and malcontents. The only reason I have not been assassinated is that the sheer number of people who want to murder me cannot fit through the door at the same time.

The election takes place tomorrow. The situation is excellent.

The United Kingdom is in a parlous state (I'm talking about real life now). The current Labour prime minister—who I will not name because that's simply not information any of us will ever need again—has resigned and a new one, former Manchester mayor Andy Burnham, is imminent. Based largely on vibes and the fact Burnham has a northern accent, the British elite is girding itself for hyper-Bolshevism. How will the nation fare under our imminent war communist regime?

I can answer that. Using cutting edge models (a copy of political sim Democracy 4), I have exhaustively plotted out the trajectory of a radical premiership in an exacting replica of the UK in 2026. Friends, I've cracked it. My advanced technologies have revealed exactly what course Britain's prime minister should take. Join me as I walk you through it.

What is to be done?

Of course, to accurately foresee the possible political trajectories of the UK, my simulation needs to reflect the country as it is. The British Isles are a peculiar beast, and capturing them in simulation means I need to be able to make a variety of fine-tuned adjustments.

Fortunately, Democracy 4:

1. Lets you simply choose the UK as a country to run from a list of options.
2. Lets you tweak it a little bit before you hit the road.

Per YouGov polling, the British public considers itself +15% liberal-leaning and +3% socialist-leaning, which I duly reflected in Democracy's settings. The national debt—a made-up number the Treasury invents to give the chancellor something to talk about in meetings—is around £3 trillion. Into the simulator it goes.

But it wasn't quite enough. This might be my home country numerically, but the spirit wasn't there—I had detail without soul. The sine qua non of British politics was missing. Some subtle, herbal ingredient was absent.

It came to me. I named all three main parties The Fascists and gave them different-coloured rosettes.

Nationalised sausages

My UK simulacrum—simUKlacrum—was mostly perfect. The skies were grey, the people dour, the spending high, the income low, the bourgeois revolution frustratingly incomplete. The only thing that seemed off was that my party—the Burnham crew—inexplicably had 10 million members and climbing. If I woke up tomorrow and heard that 10 million people had joined any UK political party I would immediately flee the country.

Apart from that? All verisimilitudinous: a true-to-life canvas on which to etch the social revolution. I decided my first move would be to try to bring that debt down—bringing expenditure below spending by waging a merciless Red Terror on the City of London and military-industrial complex.

We started as we meant to go on: boldly. Democracy 4 tasks you with balancing ministers' loyalty and their capability when selecting your cabinet. You might have troops who would go to war for you, but who would be useless at implementing your policies, or vice versa. Both are important.

I completely disregarded this and filled every post with the most obedient stooges I had to hand. Then we got to work.

The UK nuclear arsenal? Abolished. The military? Reduced to five of England's reddest dads. Corporation tax went up, capital gains had their pockets turned out; the nation's wealthy were a tube of delicious tomato paste and I was the chef de partie.

Shark-infested waters

International capital was not happy with this turn of events. The UK credit rating tumbled down a cliff face and smashed its head into every rock along the way: AAA became AA became BB became CCC became C—the ruling class was mashing the keyboard to register its displeasure, as meanwhile rivals took advantage of our newly hamstrung defences.

Argentina planted a flag on the Falklands, Russian ships drifted alarmingly close to the coast. At one point pirates manifested. I had implemented socialism in one country with such forceful enthusiasm I had somehow returned us to the Age of Sail.

None of this particularly bothered me. Let credit agencies toss entire cans of alphabetti spaghetti at me, let Argentina call them Las Malvinas. If Russia wants to paddle menacingly in our waters I'll just add it to the yearly tourism stats. Pirates are fun!

What did bother me was that the British public wasn't too psyched about how things were going either. I had come into office with unenviable approval polling and my moves to reshape the national finances along more egalitarian lines had only made things worse. Though, hilariously, I never polled as badly as the actual, real-life Labour Party, things were nonetheless on a downward trend. Even more galling: expenditure still outpaced income to the tune of £15 billion or so.

I consulted my advisers—a YouTube clip of Limmy saying don't back down, double down—and decided the answer was to become even more radically communist than I had been previously.

The five-turn plan

We needed new policies, not just tweaks to the old one. I began by implementing city farms: new kolkhozes to feed a hungry people as their fortress weathers capitalist assault.

I went further. We banned cryptocurrency, subsidised healthy food, handed out university grants to anyone who showed interest in buying a book, and taxed employers who used AI. Inflation spiralled and expenditure climbed, but my polling stabilised. Britain did not love me, but its hatred stopped growing.

Then two things happened which turned things around. First, I legislated the right to die. Britain loved this. They went absolutely nuts for it, as though they had been waiting around for someone to tell them they could die for decades. My polling shot through the roof. By which I mean it entered the general vicinity of 43% or so.

Shortly after, the whole credit rating and inflation scenario got so bad that parliament declared a state of emergency. Paradoxically, this meant I—the guy who had very much caused all this—now had way more power and political capital than I did before. I went hogwild with it. I created a state postal service and dragged its slider all the way to the right: maximum postal service. State airlines got the same treatment, as did the NHS, which now had sufficient annual income to cure death.

I also embarked on a bold policy of 'giving everyone free money,' which was popular. All of this was well-timed: the election was now imminent.

All power to the soviets

Here we come full circle, right back to where I started this piece. Britain was a nation cut in two. Socialists, trade unionists, and the poor (all of which are actual, modelled voter groups in Democracy 4) loved me. Meanwhile, in the blue corner, religious conservatives and capitalists had become so outraged at my leadership they had begun to create armed paramilitaries with names like "The Angels of Heavenly Justice."

The question was: who could better mobilise at the ballot box? Also, could I make it to the election without some modern-day Fanny Kaplan taking my head off with a light machine gun?

I hit the campaign trail with the aim of infuriating my enemies even further. There would be no rapprochement, no outreach, no olive branch to my foes. I had no interest in winning over floating voters or in persuading anyone who was not firmly in the tank already. Every speech I gave had the goal of inciting my followers to a spittle-flecked fever pitch. Meanwhile, I focused my campaign on slandering my opponents.

Somehow, it worked. 70% of my base turned out to put an X next to my name, while my nearest competitors barely managed 46%. The result was that, on a nationwide turnout of 72%, I came away with 30 million votes (or 44.7% overall), completely blowing my reactionary foes out of the water.

Britain's credit was in the tank, Russia was on the verge of seizing Kent, and buying a loaf of bread required one of those duffel bags full of cash they have in Heat, but the people were with me. I had been re-elected. The red dawn brings a bright new day.

Feel the burn

What can we take from this? In real life, Britain will soon get its new prime minister, and the compromises have already begun. Many of the faces and ideas that surrounded the outgoing, now-despised PM are already reappearing.

Clearly, this is doomed. As my rigorous testing has proved, what 21st century Britain needs is a sweeping cavalcade of courageous policies. We need collective farms, punitive taxes, the immediate execution of anyone with a business studies qualification. Bold, radical action—this is what science says this green and pleasant land requires.

That is, if you want to win the next election. I guess if not then, hey, just keep doing whatever.

Mr Burnham, I'm just a phone call away.

Mail delivery is a thankless job. No one notices all the hard effort that goes into sorting through mountains of boxes, parcels, and letters to get deliveries ready on time. Oh, but I drop a couple of dozen boxes in the sea and suddenly I'm bad at my job? Everyone's a critic.

But someone's gotta do it, and that's the fate which befalls you and your friends if you decide to play the new cozy management sim Cat Mail Co. Residents of Cat Island will turn up to your post office with mail to collect or send and it's your job to ensure everything gets to where it needs to go on time, and doesn't end up lost among a pile of old packages.

Cats will turn up to collect their mail, they'll describe said package, some in great detail and some in what feels more like a riddle than a description. Parcels will have first names and the initial of a last name on them alongside stamps that signify if they're heavy, fragile, or frozen, as well as some other visual signifiers like ribbons, stickers, or what I have deduced are scratch marks from alligators—I try not to ask too many questions.

Because I'm very much Type A, I created a thorough sorting system in which I separated boxes by weight, type, and crocodile scratches, making it pretty easy to locate mail and race through customer orders each day. But that's not to say I didn't have any trouble.

I did find myself rejecting customer requests a couple of times—when they gave me absolutely nothing to work with. One cat asked for a box, no name, no weight, no visual signifier, just a box. If I'd actually been on the other side of that counter I would've probably thrown a box at them. But I wasn't, and after giving them almost every box I had in stock only for them to reject each one, I gave up and told them to shove off. My customer service isn't great either, I suppose. But it's also not dreadful, because after the first couple of hours I managed to complete orders for 100 cats, an achievement I'm quite proud of.

But it's not all about handing off parcels. You also collect them from customers and sort them so they're ready to be handed off to the cheery Captain, who arrives at your post office by boat each morning and evening and then sets sail to deliver them to the surrounding towns.

So far I've unlocked Sunny Shores, Port Windy, and Crescent Bay. The Captain's delivery schedule will change each day, however, meaning certain areas may be missed out on, depending on which route he's taking. So you'll have to manage your stock quite closely, otherwise you can have packages sticking around the office too long, and before long you'll be flooded with piles of boxes strewn across the floor.

The Captain will also have a random amount of space on his boat, meaning you have to stack boxes like a jenga tower, with the heavy cargo on the bottom and the fragile stuff on the top. But there are other variations, like lovers boxes that have to be shipped together, or frozen items that need to be kept in the cool room until the very last second, making for a different kind of puzzle for you to tackle each dawn and dusk.

I've only been playing Cat Mail Co. for a few hours now, but there's clearly so much more to experience, with locked rooms, broken gadgets, and new places all being out of reach right now. And if you get tired of playing on your own and want to throw a live grenade into the mix, you can invite friends to play with you—warning, this will severely decrease the cosy and significantly increase the stress, but it's still a ton of fun and means you're essentially getting two games for the price of one.

Everything I know about crime scene investigations I learned from the TV series CSI: Miami, a show where sexy people get murdered and different sexy people—plus David Caruso—solve the crime by doing lab work.

My memory of the show is that Caruso always kicks off the episode by saying the first part of a sentence, then puts on his sunglasses and finishes the sentence with a devastating zinger related to the sexy murder victim. But that's sort of not true. His zingers in this 8-minute montage of Horation Caine one-liners are pretty darn underwhelming. Sometimes he just states the obvious, like when a body that had been thrown off a building was determined to have been killed elsewhere:

"Priority number one…" *sunglasses* "...is for us to find the murder scene."

Yep. That's… that's the thing everyone already knows you need to do. Wasn't a cool thing to say, and didn't need to be said at all.

Hoping to fill his shoes this week, I'm trying out a new career in Forensics: Crime Scene Detective, and unfortunately it seems I'm about as good at solving crimes as David Caruso is at delivering memorable zingers.

My first case is to collect evidence from a drug dealer's apartment. Not drugs, unfortunately, but electronic devices that include communications to his clients, perhaps text messages that read "hey if you would you like to buy some illegal drugs hmu 💉💊🌿🤪"

At the apartment I get to label, photograph, and collect evidence, which is rad. I've seen them do it on TV a million times: placing little yellow plastic numbered markers next to things like bullet holes and blood stains and then taking photos. In the alleged drug dealer's apartment there are just a couple smartphones and a password written on a post-it note, so I label 'em, snap 'em, and take them back to the lab.

That's where the problems begin. I use the password to access one of the phones, but the other is damaged so I have to open it up and "desolder" it: heat up the memory card so the glue melts and it can be removed—but without heating it up so much it's destroyed. I gingerly heat the card and try to remove it. Nope. I try again. Nope.

Ultimately, it goes the way things go anytime I toast a bagel:

• It's untoasted
• It's untoasted
• It's untoasted
• It's burnt beyond all recognition.

My first act of lab work and I nuke the evidence. And not only did I destroy this device, it turns out I also missed a third electronic device somewhere back at the apartment. I guess this drug dealer… *sunglasses* …is going to continue dealing drugs.

Embarrassingly, I flub my next two cases, too. Finding fingerprints is kinda fun: I shine a purple light around a room until I spot one—and it's always just one, because people only touch a surface in their home with a single finger then never touch it again. After labeling and snapping a pic, I sprinkle dust on the print and lift it with tape, which I bungle on the first two attempts by using almost no powder and then by using way too much. That makes it tricky, or some might say, impossible, to match prints back in the lab.

My second case: Unresolved! I'm glad this game doesn't simulate getting yelled at by the chief of police.

My third time out I'm hopeful because there's blood, and I'm good at collecting that: you just stick a Q-tip into it. But again, I wind up torpedoing the lab work. I use some sort of machine to extract DNA and then compare results on a readout to see if it matches. I know it must match: this is a crime-solving game, so I figure most of the cases will result in a "solved" scenario instead of a "nope, this one stumped us" ending, but no amount of fiddling with the knobs helps me ID a perp.

Why am I dealing with squiggly lines, anyway? Aren't we living in the grim future of no privacy and information harvesting? Surely I should just be able to feed both DNA samples into a computer and it gives me two pictures, and if the guy looks the same in both pictures, he's guilty. Why doesn't science work like that?

Finally, I get a case involving a gun. I'm an American and a former gun store owner, so this should be a slam dunk. No one was killed, this is just sort of a "please don't fire a gun in your apartment" type of case. The bullet is lodged in a wall, so I do that fun thing where you put a stick in the hole and it points where the gun was fired from, and then something more fun: shining a laser along the path of the bullet, and then spraying some kind of mist so the laser can be clearly seen and photographed. Awesome.

And for the first time, I don't bungle the evidence back in the lab. I take the gun, fire it into a water tank, and collect the bullet for comparison, which is done basically the same way as in The Great Mouse Detective: line 'em up and twist 'em to see if the scratches match. Finally, I score 100% on a case—and it only took four tries. What's that famous saying? Let justice be done though three hardened criminals go free so you can imprison one guy who shot his own living room wall.

It's about time I reveal the shocking twist in this story: those four cases were just the tutorial. That's how bad at this I am: I even fail at being given step-by-step instructions on how to do basic tasks. It's with real trepidation that I take my first real case, praying it involves someone else shooting a gun because I nailed that. C'mon, violent gun murder!

I open the folder and it's a coin burglary. Shit. Someone has been stealing antique coins. Filled with dread and disappointment, I go to the perp's apartment.

It's not so bad: at first. I dust for several fingerprints and do a decent job at lifting them. I find a whole crapload of evidence to label and photograph, including an actual pile of coins—which the game oddly does not consider relevant to this case.

But while I know I'm pretty good at working the crime scene, I'm just utter shit in the lab. And here's another twist—in my first real case I was also terrible at the scene! I missed several additional fingerprints, a second electronic device, and at least one password. Labwise, I failed to even complete the fingerprint analysis. My score for this case was a mere 16/100.

I hope someone can solve the case of my brief investigative career, because it's dead.

Performance Evaluation

I think moths are great. Their hardiness, fuzzy little bodies, and the fact that I relate to their obsession with bright lights; I too find them fascinating. Unfortunately, other than the Scalloped Oak and the Puss Moth, I've not been able to see many of the 2,500 species of moths recorded to be in the British Isles, which is why I've now turned to Mothkeep's demo to take the edge off my lack of moth finding.

Mothkeep is all about finding and documenting native species of moths, in woodlands, rolling green fields, overgrown pastures, and even inside homes—you know, moth places. The full game is releasing in just a few days and it's not expected to be very long, clocking in at just a couple of hours. But that doesn't detract from the peaceful and cozy practice of listening to calming music while searching around for some cool moths.

I spent most of my time sat in an overgrown mossy woodland waiting for the Four-Dotted Footman and the Garden Tiger moth to turn up. I waited so long for the latter that I actually had to go to the shop and spend some of the coins I'd collected from spotting moths on a bunch of leaf litter, which I could pile up in the woodland to attract the attention of the Garden Tiger moth. Luckily that one coin paid off and I soon added the Tiger to my list.

But it's not just about appreciating moths while in their Imago stage (when they get their colours and wings), you need to find instances of moths during every stage of their life: eggs, larva, and pupa as well. You can use their preferred native plants to attract these stages also.

It's a pretty basic game, but sometimes those are the best. It's nice to have something where the stakes and stress are low, it certainly balances out all the horror and competitive FPS games I play.

Mothkeep is whimsical, chill, and the perfect game to play if you want to wind down or just see some cool critters in the woods. It's also just a great way to find out a little bit more about the 160,000 species of moth we share the planet with.