There's a new subscription service on the block: Indie Pass, which promises access to "a curated catalog of games from indie developers" for $6.99/month.

Indie Pass is set to launch on April 13 with 70 games, most of which are unfamiliar to me—they come from the catalog of publisher indie.io, the company behind the service, which aims to recruit more developers to the platform over time. Those who do bring their games to Indie Pass will earn revenue based on how much time subscribers spend playing their games.

It was probably inevitable that someone would try a cheaper, indies-only version of Microsoft's Game Pass, but the reception to the idea has been less than stellar.

The prime complaints being aired on social media are:

• Compensation based on time played discourages games that aren't optimized to keep players in an endless loop
• Indie games are already relatively cheap, so why not support developers with a direct purchase?
• It's another way to not own the games we play
• Oh great, another launcher

Or, as 3D Realms co-founder George Broussard put it, "this sort of thing is poison for game developers and a net negative for gaming."

The reactions aren't entirely hostile. "If the split to the devs is solid and it helped me find titles I otherwise wouldn't or would be on the fence about... Maybe there is something there?" speculated one commenter, as an example.

Speaking to GamesIndustry.biz, Indie Pass director of growth Jess Mitchell emphasized discoverability as a reason for developers to join the service.

"Candidly, one of the biggest challenges indie developers face right now is the sheer number of games out there," Mitchell said. "That number seems to be growing by the minute. If you don't have a huge splash on your launch, then you might not have another discoverability moment for three, six, even 12 months after release."

The company is "confident that Indie Pass will not damage games' sales performance," according to GamesIndustry's report, but even if it could guarantee that, I wonder how many developers would bother. It struck me recently that Mega Crit could have launched Slay the Spire 2 on the Epic Games Store if it had wanted to, and is probably leaving money on the table by not getting in front of the Fortnite audience, but presumably launching only on Steam simplified things.

Publisher Devolver Digital doesn't appear to be taking the idea seriously, joking on X: "If you send us $100,000 we will give you an Indie Pass to play all of our games."

Given that the response I've seen has mostly been 'no thanks,' Indie Pass clearly has an uphill battle ahead as it tries to convince developers to sign up, and us to put yet another recurring charge on our credit cards.

It may also not be a great sign for Indie Pass's chances that a Google search for its name currently autocorrects to "indy pass," returning results for a ski pass usable at independent ski resorts around the world.

One might wonder, given the endless selection of good videogames out there: Why play a bad one?

But let me turn that idea around on you. Given the endless selection of good videogames out there, what's more enticing than one you've never heard of? Maybe it's bad, but has one fascinating idea. Maybe it has zero fascinating ideas but is so stupid it makes you laugh nonstop. Maybe it's completely irredeemable and just makes your next Slay the Spire 2 run seem that much better in contrast.

These are the sorts of games that ScummVM, a tool that makes old games play nice on modern machines, excels at supporting. Once primarily focused on adventure games, ScummVM has broadened its remit over the years to make all sorts of old PC games playable on all sorts of platforms, from modern Windows to Macs to the Nintendo Wii.

Some of the classic games preserved on GOG even run on ScummVM, and I always look forward to new ScummVM releases introducing compatibility for old game engines and their respective games I've never heard of. On March 28, the addition of two new (old) engines to ScummVM's innards brought support for eight games, at least one of which you've likely heard of: Mad Dog McCree. PC Gamer scored the western-set arcade shooter a generous 4%.

Now wouldn't you like to play Mad Dog McCree's less famous cousin? Then I give you Crime Patrol and Crime Patrol 2: Drug Wars. Both are '90s FMV shooters that will make you desperately wonder how many of the amateur actors on screen are actually on cocaine despite the drug bustin' theme.

The real treat is perhaps adventure game Necronomicon: The Dawning of Darkness, which you can buy on GOG and study to understand why point-and-click adventure games were not doing so hot around the year 2000. YouTuber Ermacgerd Longplays finished it in two hours, then gave it a "terribly low 2/10."

"I really despised it and thought it awful, I'd give it a Certified Crap title and what not," they wrote in the description. Sounds pretty entertaining to me! Without bad adventure games, we'd never get great adventure game Let's Plays like Nextlander's Full Motion Vinny.

ScummVM helpfully maintains a wiki page listing where most of the games it supports can be bought or otherwise found, and quite a detailed FAQ about how everything works. Use it to play an old Discworld adventure game, or Mad Dog 2: The Lost Gold. It's the weekend!

If you could type a series of sentences into a computer and have it generate a videogame for you, would you?

Well, you can—kind of. Tesana is one of a few companies promising that we can all make games simply by typing prompts into a generative AI system. In tandem with third-party AI services like Claude and asset generators, its proprietary game engine "translates descriptions of environments, mechanics, characters, and rules into structured game code."

Judging by the examples on its website, Tesana's users are mostly testing it out by generating garbled versions of other games. Valdenholt, for example, is like Skyrim if you took it out of the computer, stomped on it, and then smushed some of the pieces back in. But it is technically a game-shaped piece of software.

The service attracted around 10,000 paying customers in its first couple weeks, according to co-founder Johannes Vermandois, who has previously founded a skincare company and an AI marketing platform. Vermandois showed off Tesana at the Game Developers Conference in March.

Vermandois thinks Tesana will be useful for prototyping games, and imagines that the speed at which users can iterate on ideas will lead to the invention of entirely new kinds of games. He also foresees full, finished games being produced with prompts, and wants future versions of the software to be capable of creating games as good as, for example, Valheim.

"If we can have 100 million new people being able to make games, you know, that will be an exciting thing," Vermandois told PC Gamer.

For Vermandois, it's a given that more people making more games is good. He agrees that many of those games will be bad (lots of bad games are made without the help of AI, after all), but thinks the quality bar will rise as more people are able to turn their ideas into software.

"I feel like indie games are really the future of gaming," he said. "And I think that these kinds of tools and platforms that we're building at Tesana could probably help more people make games, and then the sheer volume of games will increase … And I think the quality bar of games will increase as well, because there's so many more games."

I'm not so sure about that myself. Game development is more accessible than ever—you don't need a PhD in computer science to make something in open-source engine Godot—but it has never been easy. It seems to me that generative AI offers ease in exchange for control over the details, which is no small thing if you believe that "art is all in the details."

If 999,999,999 other people can generate the same game as you, why would anyone care about yours?

"I think people will still pay for other people's creative vision," says Vermandois. "Like for the same reason people go to a movie: You want to experience someone else's story or vision. I think that's still true. For example, I've been playing around with [AI music generator] Suno. I can make some really cool music myself that I like to listen to, but then I also listen to, you know, Spotify. I like artists on Spotify."

It seems that where we differ is in how much we consider prompting generative AI systems to itself be an expression of creative vision. Would anyone bother with Spotify if it were exclusively full of other people's Suno-generated tracks?

Vermandois is obviously more optimistic about the technology than I am, but he doesn't claim to know what the future holds, either. Regarding worries that AI processing costs will go up once companies like Anthropic can no longer subsidize their users, he's hoping that model costs instead go down. And on how Tesana will compete with Roblox and other big companies slapping generative AI onto their products, he's holding onto a startup founder's necessary optimism that the small guy can beat the big guy.

Success for Vermandois will be creating a new market for new kinds of games, and "a new hobby for a lot of people." For now, the first test games made by Tesana users have nothing to offer the rest of us, but as we often hear when talking about generative AI, "it's very early days."

Slay the Spire 2 is getting badges—not the ones you get on Steam, but not entirely different, either. In the notes for a new beta patch, developer Mega Crit said badges are a bit like achievements, in that they let you know about the cool things you accomplished during your run.

"Badges may appear at the end of a run, noting various things that happened during a run like defeating a boss without losing HP, beating the game quickly, specific easter eggs, or pointing out that you forgot to spend your gold," the studio explained. "They're meant as little reminders to let you know what was unique about each adventure."

For now badges will only appear at the end of a run, but Mega Crit plans to have them also appear in your run history and stats screen at some point in the future. The intent, the studio said, is "to make runs feel distinct and improve scoring," something it will share more on later—although, notably, the patch notes further down also state that this update has "adjusted and added placeholder assets for score lines in run summary and game over screen."

The addition of badges is the top-line thing, but there's quite a bit more going. Map generation via Ascension, the Blade of Ink and Borrowed Time cards, and Doormaker's Grasp have all been adjusted, and numerous smaller balance changes are also in place.

There's also "new art additions for various cards and a few enemies, as well as some cool new VFX for afflictions!"

So far, the response seems fairly positive—or at least, not hugely negative. The previous beta balance patch sparked a massive backlash from players that resulted in nearly 9,800 negative reviews in one day. Mega Crit co-founder Casey Yano said he wasn't bothered by the review bombing because he understands that players are simply "trying to interact with the game and be heard" using the tools they have at hand, but even so, that kind of forcefully negative response has to be a little rough.

There's been a surge in negative reviews in the wake of this balance patch too, but not nearly as ferocious: 801 negative reviews posted so far today, compared to 817 positive. That's still not a great balance, but it's a lot better than the last time around.

As the name implies, this is a beta patch, so if you want it you'll have to be running the game's beta branch: To access that, right-click Slay the Spire 2 in your Steam library, select "Game version and betas" from the menu, and then the public beta branch. The full patch notes are below.

CONTENT

• General:
  
  • Reworked Ascension 6 from Gloom - "Less rest sites." -> Inflation - "Removing cards from your deck at the Merchant is more expensive."
    
• 
  As mentioned last week, affecting map generation through Ascension has increasingly seemed to be a miss. It's often too hard to tell what change if any is a result of Ascension, or if it was just an unlucky map generation. Instead Ascension 6 will now make card removal at the Merchant start at 100 Gold, and will increase in price by 50 gold per removal instead of 25. I expect this to be felt significantly more than the old Gloom modifier.
  
  Silent:
  
  • Reworked Blade of Ink card: "Rare - Cost 1 - Skill | This turn, whenever you play an Attack, gain 2(3) Strength this turn." -> "Rare - Cost 1 - Skill - Add 2(3) Inky Shivs into your Hand."
    
    • Added new Inky enchantment: "This card deals 2 additional damage and applies 1 Weak."
      
  • 
    
• 
  The Blade of Ink rework is an experiment with generating Enchanted cards mid combat. Definitely let us know how this feels or if Enchantments are best left to things like Events and Relics. I suspect this card will be high impact, and gives Silent another strong Weak applier.
  
  Necrobinder:
  
  • Reworked Borrowed Time card: "Uncommon - Cost 0 - Skill | Apply 3 Doom to yourself. Gain 1(2) Energy." -> "Uncommon - Cost 1 - Skill | Gain 4(6) energy. Cards cost an additional energy this turn."
    
• 
  This new Borrowed Time should go particularly well with high cost cards like Reap and Bury and help to flesh out other ways to play Necrobinder.
  
  Enemies:
  
  • Reworked Doormaker Grasp turn and power: now attacks for 20(23) and the power is reworked to "Whenever you play a card, lose 1 energy."
    
• 
  Doormaker's Grasp turn was too underwhelming, and overall lowered the difficulty of the fight too much. This new version should make the entire fight high stakes and actually give the player a difficult problem to puzzle over turn to turn instead of making the fight about just living until turn 4.

BALANCE

• Silent:
  
  • Buffed Leading Strike:
    
    • Shivs increased from 1 -> 2
      
    • Damage decreased from 7 -> 3
      
  • 
    
  • Nerfed Speedster: damage decreased from 2(3) -> 1(2)
    
  • Nerfed Untouchable: Block decreased from 7(9) -> 6(8)
    
• 
  Potions & Relics:
  
  • Changed History Course relic: when it plays a copy of an X-cost card, it now uses your current energy for its X-value instead of the energy spent the last time the card was played. Note: It still plays the copy for free
    
• 
  Enemies:
  
  • Buffed Skulking Colony:
    
    • Now attacks on all turns
      
    • Deals slightly more damage
      
  • 
    
• 
  Ancients:
  
  • Buffed Nonupeipe's Fur Coat relic: now affects restocked Axebots and other enemies that enter combat during a fight
    
• 
  Events:
  
  • Nerfed Hungry For Mushrooms event: Fragrant Mushroom option upgraded cards decreased from 3 -> 2
    
  • Changed Waterlogged Scriptorium event:
    
    • Prickly Sponge option gold cost lowered from 155 -> 99
      
    • Tentacle Quill gold cost lowered from 65 -> 55
      
  • 
    
  • Buffed Whispering Hollow event: Exchange option gold cost decreased from 50 -> 35+-9

ART

• Added portrait art for the following cards:
  
  • Hotfix
    
  • Modded
    
  • Synthesis
    
  • Sweeping Gaze
    
• Added new art for Ruby Raiders
  
• Added background art and VFX for Vantom boss fight
  
• Added new visuals for the following afflictions:
  
  • Bound
    
  • Entangled
    
  • Galvanized
    
  • Hexed
    
  • Ringing
    
  • Smog
    
• Updated empty chest VFX, smoke puffs

USER INTERFACE & EXPERIENCE

• General:
  
  • Adjusted and added placeholder assets for score lines in run summary and game over screen
    
  • Adjusted timings and pauses for dialogue for Gremlin Merc
    
  • Adjusted Cultist "Cacaw" dialogue timings
    
  • Adjusted speech bubble positions for Damp and Calcified Cultists
    
  • All monsters now have distinct speech bubble colors
    
  • Players can now navigate between cards/potions/relics in the Timeline unlock screen on controller
    
  • Players can now navigate to the top bar character portrait to see the Ascension hovertip on controller
    
  • Pressing the top bar shortcut while already focusing on a top bar element now moves your controller focus back into your hand
    
• 
  Multiplayer:
  
  • Better handling of intermittent Steam errors
    
  • Disabled skipping chests in multiplayer. Too many people thought it was the Proceed button.
    

WRITING

• General:
  
  • Foul Potion description now specifies that it deals damage to "ALL players and enemies" instead of "EVERYONE", making it clearer that it does not hit Osty
    
  • Fixed wording in the Statistics Total Playtime/Fastest win hovertip
    
• 
  Localization:
  
  • Updated translations for various languages
    
  • Fixed translation errors in the following languages:
    
    • German
      
    • Japanese
      
    • Italian
      
    • Spanish
      
    • Thai
      
    • Polish

BUG FIXES

• General:
  
  • Fixed broken Steam rich presence translations
    
  • Fixed a crash that could occur on startup after a Steam Cloud sync conflict
    
  • Fixed a crash that could occur when the game failed to start and tried to show an error dialog
    
  • Fixed an issue where numbers were improperly formatted on the victory screen
    
  • If there is an error while spawning a creature into combat, a placeholder is shown instead of softlocking the combat
    
  • Fixed softlock caused by being able to use your controller to confirm a hand card selection while you are peeking
    
  • Fixed being able to press down arrow to controller navigate the card grid while an upgrade/remove/transform preview is up
    
  • Vulkan is now defaulted to on Intel iGPUs with known D3D12 stability issues
    
• 
  Ironclad:
  
  • Fixed softlock when selecting a card to upgrade via Armaments card while Lethality power still applies
    
• 
  Silent:
  
  • Fixed Bullet Time card getting played by Mayhem power not setting the cost of the cards in your hand to 0
    
• 
  Potions & Relics:
  
  • Fixed Unsettling Lamp relic doubling the Strength you get from Dominate card
    
  • If a potion is discarded just before combat ends, it's no longer forever unusable
    
  • When clicking a potion that can be thrown at all enemies simultaneously, the potion popup now properly says "Throw" instead of "Drink"
    
• 
  Enemies:
  
  • Fixed an issue where a few enemies' damage intents were occasionally showing the wrong number
    
  • Fixed not being able to navigate from the relic inventory back to the combat in the Mysterious Knight fight on controller
    
  • Removed broken text from Thieving Hopper enemy's Escape Artist power hovertip when playing in Polish
    
  • Fixed Soul Nexus animation not playing correctly
    
• 
  Ancients:
  
  • Removed the peek button from the card selection via Neow's Hefty Tablet relic
    
  • Effects that occur before your turn now interact correctly with Vakuu's Whispering Earring relic
    
  • Fixed Vakuu's Whispering Earring relic continuing to play cards after Void Form power ended the turn
    
  • Fixed an issue where Vakuu's Lord's Parasol relic would occasionally only obtain some of the items that the Merchant sells
    
  • Fixed being able to end the Lord's Parasol purchase animation on controller
    
• 
  Events:
  
  • Fixed an issue where, if the player visited a combat-style event without actually starting a combat (i.e. choosing "Return the Key" at The Lantern Key event) and then visited the Merchant immediately after, throwing a Foul Potion at the Merchant would not grant the player any gold
    
• 
  Multiplayer:
  
  • Trash to Treasure power no longer procs from other player's generated statuses
    
  • Fixed softlock in Thieving Hopper fight if one person is dead when it tries to steal cards
    
  • Fixed state divergence related to Driftwood and Dream Catcher relics
    
  • Player name is now correctly displayed when hovering over your character
    
  • When re-entering the multiplayer host screen, the Ascension is now correctly synced
    
  • You can now reload after saving a game at a rewards screen with non-standard rewards

As was recently spotted by Kotaku, Esquire Singapore published a feature in March all about live-action One Piece actor Mackenyu Arata called "Mackenyu in Resonance." It includes what writer Joy Ling calls "an unprecedented interview for both parties," which isn't so much an interview as it is a transcript from an AI chat bot masquerading as Mackenyu.

"We were stoked to have some face time with the Japanese-American actor, but his schedule prevented it," the story reads. "We had the photospread, but nothing directly uttered by the 29-year-old. With a driving need for a feature, we had to be inventive. Harnessing our creative license, we pulled his [responses] verbatim from previous interviews and fed them through an AI programme to formulate new responses."

The "interview" is a bizarre read, and was produced using Claude and Copilot. Esquire asks about the actor's experience becoming a father, how he deals with "disillusionment" (not in regard to a particular part of his life, just… disillusionment in general, I guess?), and even gets into his memories of his late father, Sonny Chiba. The AI responds with the sort of unsubstantial meandering you might expect, though at one point it does type "(laughs)" so you know it's in a good mood.

Readers and writers are not pleased with the piece. "This makes me so angry," wrote New York Times-published journalist Nicole Clark on Bluesky. "In what world is any of this appropriate or justifiable? And even with 'disclaimers' everything else about the piece looks like a typical interview. It's the equivalent of using ad space to publish misinformation."

"That's not an interview; it's fan fic dialogue," posted Bluesky user Nona Jabiznez. A reply from user Maggie Sharp reads, "Fan ficcers write their own stuff and if they don't they're hunted for sport."

It feels like the golden age of the AI-operated sock puppet, considering this Esquire piece and the AI clones of real journalists recently in use by companies like Grammarly.

RTS games generally aren't my thing, mostly because I'm not very good at them. But I loved the original Homeworld: Enemy AI was too easily exploited (although for me, that's a feature, not a bug) but the setting and presentation were magnificent—more than enough to make it one of the most unforgettable games I've ever played.

I'm not the only one who thinks so: PC Gamer's Fraser Brown, who very much is a strategy guy, said a few years ago that even after decades, Homeworld is still his "dream RTS." That's a pretty powerful recommendation.

For those of you who haven't played it, this is your lucky day: The Homeworld Remastered Collection, with remastered and "classic" editions of that great sci-fi epic and its perfectly fine follow-up, is available for just $8 as part of the new 2K Tactics and Tycoons Humble Bundle. Astute online shoppers may point out that the Homeworld Remastered Collection can be had for just $3.49 on Steam right now—that's 90% off of the regular price—and that is true.

But wait! There's more! (You didn't think I'd point you toward a bad deal, did you?)

The 2K bundle also includes the outstanding prequel RTS Homeworld: Deserts of Kharak, which Rob Zacny—another serious strategy guy—said is "a terrific RTS" in his 90% review, and "also an excellent Homeworld game that reinvents the series while also recapturing its magic." Deserts of Kharak is also 90% off on Steam right now, dropping it to just a fiver—so you're already saving money with the Humble Bundle.

But wait! There's more!

I'm really just here for the Homeworld, but there's a bunch of other good stuff on tap too, if you're into the RTS scene. You'll also get:

• Railroad Tycoon 2 Platinum
• Army Men
• Army Men 2
• Army Men: Toys in Space
• Army Men: RTS
• Shattered Union
• Freedom Force
• Freedom Force vs. The Third Reich

These games are quite dated but generally well regarded (with the possible exception, I am informed, of the Army Men games), particularly Railroad Tycoon and Freedom Force vs. The Third Reich, and one of the nice things about RTS games, and strategy games in general, is that they don't rely on cutting-edge visuals to succeed. In fact, I look at this screen from Railroad Tycoon 2, released in 1998, and aspect ratio aside I think it looks pretty sweet.

Of course, Homeworld is just flat-out sexy.

So anyway, if you want just Homeworld Remastered, or Deserts of Kharak, Steam is probably the way to go. But if you want both (and why wouldn't you?), then get the bundle—you'll get a bunch of other good games along with them, and you'll save 50 cents, too! (Hey, every little bit counts these days.) The 2K Tactics and Tycoons Humble Bundle is available until April 24.

I wondered, last week, if the new call for aid put forth by the Skyblivion development team was a sign that the long-awaited mega mod—the one recreating The Elder Scrolls 4: Oblivion in the newer tech of The Elder Scrolls 5: Skyrim Special Edition—might be delayed again. It was supposed to be out in 2025 but was pushed into '26 at the last minute, and to me this new update made it sound like there's still an awful lot left to do for a game that's supposed to be out within the next eight months.

Anything's possible, but Skyblivion project lead Rebelzize told me in an email that he remains "pretty optimistic" about things, and that while the development team is "stretched thin," that's been the case for the past decade already.

"The call to action was an idea to try and get a few extra hands on the project to help out, mostly with polish," Rebelzize told me. "We listed all departments not because we are dead in the water and have nobody left to help out, but because we currently have good momentum and want to take every opportunity we can to make sure we can keep that up and maybe even find one or two more people to help speed things up further."

All quest locations have been finished and QA-checked, the Imperial City—which was cited as "our last major hurdle" in last year's delay announcement—has "gotten the breakthrough we needed," and of course bugs are being fixed "left and right." Rebelzize said "there's always the chance shit hits the fan and we suffer another major setback," but added that "from what I see internally, I am pretty optimistic."

Rebelzize also had some advice for followers who might be a little too overeager to criticize the Skyblivion team's efforts, or predict doom and gloom for the mod's future.

"My goal has never been to keep this game from people any longer than I have to, but I owe it to everyone who helped out over the years to make sure we have a functional game to deliver," he wrote.

"Just as the team and myself are very passionate about our work, so are members of the community. However you voice that excitement please take a step back and recognize the work and the people behind it. There are no ill intentions on our part and everyone is doing the best they can to get this out ASAP."

Thanks to the breathtaking power of artificial intelligence, we may now use natural language to beg the computer to perform a function that once came as naturally to its binary brain as breathing does to ours. Google—a company worth 3.5 trillion dollars, uncontested owner of both the almighty internet search algorithm and the web browser, has overengineered its latest Discover feed AI curation tool to such an extent that it cannot simply block a website from appearing as a source.

Google's 'Tailor your feed' feature, introduced a few months ago, adds a familiar chat prompt that encourages you to type something like "Keep me updated on what's happening in college basketball" to curate what shows up in your Discover feed. Discover, if you're not familiar, is what Google calls the list of recommended stories that shows up when you open a new tab in Chrome, and is also built into the Google app and Android.

'Tailor your feed' is an experimental "Lab" feature you have to enable manually. I can't recommend it, because it will make you feel mad, sad, and dumb, not necessarily in that order.

I'm no computer scientist, but my understanding is that all the stuff we're doing On Here ultimately distills down to, y'know, binary: on or off, yes or no, zero or one. On one hand, this is such an oversimplification of technology that it may have little bearing on the complexities of the large language models and mystery algorithms that power modern AI agents. On the other hand, I don't think that actually matters when the end result of that modern complexity is failing at a task as simple as yes or no. I enabled 'Tailor your feed' so that I could tell Discover to completely block X.com links from appearing in the list of curated stories, because I really don't need to see what the worst people in the world are saying when I'm scrolling for headlines about new brunch spots in my neighborhood. And here's what happened when I made that request:

You won't see as many posts from X going forward.

I can't guarantee you'll see none of something.

As I found with a quick search on Reddit, I'm not the only person who's tried this solution since Google started flooding Discover with posts from X in late 2025. (Given that Google could pull quality writing from a billion websites and blogs across the internet, I have to wonder why it instead decided the site that was plagued with AI-generated child sexual assault imagery was an ideal source.) After typing in something similar, one Redditor reported seeing only "one story from X as opposed to the usual dozen."

"It needs constant reminding but better than it was…" the Redditor wrote recently.

So is Google placing its thumb on the scale by refusing to remove a paying partner entirely from its feeds, or has it developed a piece of AI software so staggeringly stupid that it can't even create a blacklist (or 'blocklist' as they're more often called these days)? The entire damn internet is built around blocklists. Google Chrome has built-in blocklists of websites it deems unsafe. Every piece of antivirus software in existence? Blocklists. Internet Explorer had blocklists in 1996!

What are we doing here, man. "I can't guarantee you'll see none of something" should be a vastly more difficult operation for a computer than "zero." Google Labs is out here spending billions of dollars trying to unsolve problems. This new frontier for software is indeed revolutionary, because I've never simultaneously felt embarrassed for the computer and for myself when trying to coax it into the simplest possible action.

It's clearer than ever that big tech's vision for the algorithms choking the life from the internet springs from the idea that they know better than we do what we actually want. It really is time to turn off the recommendations and ditch this stuff once and for all. RSS awaits.

Norman Caruso, better known online as The Gaming Historian, has spent the last decade and a half making Ken Burns-style documentary videos for his YouTube audience; he currently has over 1 million subscribers. He's covered everything from a TV with a Super Famicom built into it to the Sega Mega Modem. But now he's calling it quits.

A video went up on Caruso's channel today titled "Thanks for Watching," and he promises there's "nothing dramatic" behind his decision—he just got burnt out after making his most ambitious video yet covering The Oregon Trail, and in the two years since it released, he hasn't wanted to make more.

"I assumed that after a few months, I'd get the itch again and make a new video," Caruso said in the video. "To my surprise, that itch to make a new video never really came back. I definitely tried to make something else. I even announced it on social media. But my heart just wasn't in it, and I knew that if I ever finished that video the quality would suffer. That's when it finally clicked: I was ready to move on from The Gaming Historian."

Moving forward, Caruso will focus on the history podcast he hosts alongside his wife Kristin. But even though he's putting his YouTube channel behind him, he unveiled one last treat for fans: a shedload of court documents he scanned for a video he no longer plans to make.

The documents cover a legal case from 1982 where Universal Studios alleged the game Donkey Kong was a trademark infringement of King Kong. Caruso discussed the case briefly in his video on how various Mario characters got their names, and in that he showed off a court document which lists several other names that were considered for Donkey Kong during development of the arcade game—among them are Bill Kong, Kong Holiday, and Kong Chase. I tell you that because I think we'll agree that Kong Holiday is a much better name.

For more trivia nuggets like that, you can view the documents Caruso uploaded to the Internet Archive. He said he also provided the files to the Video Game History Foundation in his video. "Making the Gaming Historian was truly a life-changing experience," he said. "I will always cherish that chapter of my life."

Samson, the open-world crime brawler being developed by Liquid Swords, looks to me like a Grand Theft Auto game leavened with a healthy dose of Condemned: Criminal Origins. It's a much smaller-scale project than (relatively) recent GTA games, of course—because what isn't?—but a tale of a low-level mook making his way in a city filled with crime and cars? Yeah, that sure sounds like GTA alright.

The studio has previously acknowledged those comparisons—in February, mission designer Donald Young told PC Gamer that the comparison to GTA "makes sense, just because it's an open world city game"—but also pushed back on them, pointing out that both the game and the development team are much smaller than Rockstar and the upcoming Grand Theft Auto 6.

In a new interview with PC Gamer, Liquid Swords co-founder Christofer Sundberg—also the co-founder of Avalanche Studios and creative director of the Just Cause games—went deeper on it, saying that Grand Theft Auto "has always been a phenomenon" that goes beyond just videogames: "It competes with everything in entertainment."

"I have a huge respect for Rockstar and what they've done to the industry," Sundberg said. "Because every time a GTA game releases, it's just Christmas for everyone. It's like when a new iPhone releases. It's something that so many want, that don't even play videogames. They buy it because, when it releases, it's the talk of the town."

A short video clip of Samson released in 2025 led to immediate comparisons with GTA 4, Sundberg continued, and he acknowledged that there are "similarities" between the two games. But rather than trying to duplicate Rockstar's magnum opus, "which is impossible," his goal is to make Samson a game people want to play alongside GTA 6, rather than instead of it.

"There are times when you want to put GTA down and pick up something else," Sundberg said. "And so I see Samson as being like back in the day when action movies were 90 minutes long, not over two hours. I keep on going back to watching Die Hard and Ronin and First Blood and Rambo. I think there's a space for us there, and that's where I want to end up."

I would first note, because I'm that kind of guy, that both Ronin and Die Hard are over two hours long, but even so the point is valid: Even the most compelling epics can be a bit much at times, and it's nice to have a smaller, simpler diversion when you need a break.

For me, that would be something more like Plants vs Zombies while I'm playing Elden Ring, and I do wonder if people who want a GTA 6 pause will immediately think that the best idea for respite is a smaller-scale GTA. But for folks who don't have the space or the temperament to take a Grand Theft Auto game head-on (and yes, I'm talking about myself here), Samson might prove to be a cars-and-tough-guys game that fits.

It's been something of a rough ride for Samson: It was originally intended to be a much bigger game, but Liquid Swords had to pare it back significantly midway through development because it laid off half its team in 2025. We'll see the results of those changes soon: Samson is set to launch on April 8.

You're spoiled for choice these days if you love games about doing chores, which I guess I can't fault you for given the thousands of hours I've spent making numbers go up in various MMOs. If cozy cleanup games like Unpacking or PowerWash Simulator aren't enough to sate your hunger for affectionately simulated mundanities, you may want to check out House Flipper—it is free for the next few days, after all.

You can claim a free-to-keep copy of the game on Steam through April 6, where it and its numerous DLC packs are discounted for the storefront's House and Home Fest. If you get hooked, the Flipper Bundle with all the game's paid DLC is also on sale for just over $36. The sequel might be out with a Scooby-Doo DLC and everything, but you can't beat free.

Another reason to claim this is that the remastered version is set to release later this year, and anyone who owns the game will get a discount based on how many DLCs they have. You can read the price breakdown here, but if you just get the base game that amounts to four dollars off the remaster's $50 price tag. It's not nothing!

If you aren't familiar with House Flipper, it's a home renovation sim where you bust down walls and clean up messes in grimy, devastated living spaces. Once they look spiffy enough, you sell them at a profit and buy new ones. The DLC expansions throw in extra furniture, pets, and new settings altogether, with one free DLC taking you to a cyberpunk city.

While it certainly appeals to a particular niche, it won't be for everyone. "There's a definite satisfaction in taking a gross room and making it look nice, and it's pretty cool that you can knock down (and rebuild) walls," said PC Gamer senior editor Christopher Livingston in his impressions of the game back in 2018. "But I just don't find the act of slowly and mechanically painting and cleaning much fun, especially with the knowledge that my actual house could do with a bit of that." But for free, it can't hurt to try.

Echo Point Nova is a wonder. Greylock Studio's indie FPS enchanted us in 2024 with its high-flying, hoverboarding, hyperchill gunplay, and—if ziplining onto a cloud with an energy grappling hook to shotgun blast an enemy soldier during a slow-motion backflip somehow left any room for doubt—it secured a permanent place in our hearts with a massive free expansion last August.

Now, the "2007-core" shooter is getting even bigger. In an X post yesterday, Greylock founder Matt Larrabee announced that Echo Point Nova will be receiving yet another meaty update later this month, bringing more replayability and remixed movement styles.

The main attraction of the update is the addition of a New Game+ mode, which Larrabee says will let players fine-tune their replays by deciding which unlocks they want to carry over between playthroughs. New Game+ will provide additional customization options like enabling randomized weapon pickup locations and enemy spawns.

While I appreciate the chance to embark on a fresh set of firefights across the sky islands without needing to relocate my favorite shotgun, I'm more interested in the new movement modes also being added in the April update. As alternatives to the default hoverboard and zipline, the update will offer a jetpack, wallrun-enabled sprinting for some Titanfall flair, and what Larrabee calls the "rope grapple" movement style.

That one's web-swinging. Like Spider-Man. And yes, just like the default grappling hook, it also works on clouds. That's videogames, baybee.

Echo Point Nova is a celebration of the simple videogame joy of moving and shooting at high speeds, and I could not be more thrilled to see it offer more ways to express and explore that satisfaction by shifting the player's relationship with momentum and gravity. How will the moving meditation of emptying a magazine change when a floating boulder is something to whip myself around rather than slingshot past? I'm eager to see.

Larrabee didn't put a date on the April update, but you can keep tabs on Echo Point Nova on Steam.

Six years after it was announced, and two years after it last came up for air, we have new proof of life for State of Decay 3, the zombie survival shooter in development at Undead Labs. And this time, it's not just a promise that it's still happening, but a call to the community to sign up for alpha playtests happening in May.

"I'm not going to spoil the surprise for you, but I can say that it will feature four-player co-op, some new base-building and resource strategies, and a whole lot of combat," State of Decay co-creator Brant Fitzgerald said in today's announcement video. "If scavenging supplies in the middle of a zombie outbreak sounds fun to you, then grab your ruck, pack some mags, and head over to our website to find more information and register for a chance to be included in the alpha."

It's been a long time coming. State of Decay 3 was announced with a short cinematic in July 2020, and that was it until another cinematic dropped in 2024 to remind everyone of its existence—but still without a release date. The only thing that's happened since was an overlong rebranding exercise for developer Undead Labs.

So this is kind of a big deal then, to the extent that—from the perspective of players, at least—State of Decay 3 is finally moving from talking about it to actually doing it, even if just in a limited sense at this point. But even in a rough, unfinished state, it'll be fun to see what Undead Labs has in mind for the new game.

If you want a shot at access, you can sign up at playtest.stateofdecay.com. If you get in, you'll be contacted (by email, I would assume) with further instructions; if you're not contacted, you didn't get in. But don't give up hope: Fitzgerald said names will be kept on file, and there will be "plenty of other opportunities for you throughout the year."

Not content with adding one kind of sea power to Rust in February's Naval Update, Facepunch studios has introduced a whole other way to exploit the oceans in its latest patch for the survival game. The Spring Clean update adds a bunch of small but significant changes to the game, but the one that immediately caught my eye was the addition of water wheels.

"You liked wind and solar? Now it's time for another classic, hydro!" wrote Facepunch in a Steam post, referring to Rust's previous additions of wind turbines and solar panels. Water wheels are the newest way to harness renewable energy in Rust, letting you generate current from currents by placing them in rivers or oceans.

Where things get interesting, though, is that you don't necessarily need water to make use of a water wheel. "The wheels can also be used as giant hamster wheels if you want to hop in and convert some calories into power", Facepunch points out. But as the capricious gods who govern Rust's world teach us, why waste your own energy when you can steal someone else's? "Before you ask, yes, you can force handcuffed players into them," Facepunch adds. Truly, Rust remains the cruellest game out there.

Water wheels and indentured servitude aren't the only new feature the update introduces. The headline addition is actually an armoured ladder hatch, designed to maximise protection of any vertical access points of your base. Others include snazzy coloured buttons to help you remember which is the right one to push, an overfishing mechanic so that players can't exclusively rely on the bounty of the sea, and a more efficient system for unlocking multiple technologies on tech trees.

Finally, this being Easter weekend, the patch also folds in some seasonal extras. Rust's Easter Egg Hunt returns, letting you scour the wilds for colourful painted eggs, which can be combined and upgraded to valuable bronze, silver and gold eggs. There's also some deeply unsettling bunny costumes and a knockoff Fabergé egg available for purchase in Rust's Steam item store, just in case you're really committed to showing off your love for Jesus and/or chocolate to your murderous virtual neighbours.

Outside of its ever-dependable updates, there's a small chance we might be seeing a Rust movie sometime in the future. Facepunch's director and COO Alistair McFarlane recently told PC Gamer that the developer had been "talking to people" about the idea, including "big-name companies", though we shouldn't expect it to happen anytime soon.

"When they approach us, they'll take it, they'll pitch it, they'll want writers to come and pitch it, and they'll want to nominate themselves," he elaborated. "They'll go away and create a script, and then they have to pitch it to another layer, and then that gets another layer, then they'll need to pitch it to the studios. So it's a whole pyramid to try and get anything off the ground."

If you're looking for something to do in about three weeks' time, HoYoVerse is now taking signups for a round of closed beta testing for its next game, Petit Planet, that's set to begin on April 21.

Petit Planet puts players in charge of an entire world, contained within the vast Starsea, where they can farm, fish, cook, hang out on the beach, make friends, form galaxies, and just generally live the cozy cosmic life.

"Spend relaxed and fulfilling days on your planet alongside your fuzzy Neighbors, and let your imagination run wild as you decorate your home or shape your own stylish landscapes," HoYoVerse explained. "You can also explore the Starsea and forge connections with people and things at the ever-bustling Galactic Bazaar.

"Each bond you form with Neighbors and the world around you weaves into an ever-growing web of connections that keeps the universe flourishing. As these bonds deepen and planets thrive, Luca, the vitality that flows through all things, shapes new landscapes and brings wondrous changes to your world."

Yeah, that sounds awfully cozy, alright. Lest there be any doubt, watch this:

The Stardrift Test, as HoYoVerse is calling it, will introduce new Neighbors, offer "enhanced Starsea Voyage experiences," and add additional activities at the Galactic Bazaar. If you'd like a shot at access, you can sign up at planet.hoyoverse.com—HoYoVerse said "a selection" of people who sign up will get in, but pre-registering will also unlock "exclusive rewards" that I will assume will become available when the game launches.

Petit Planet does not currently have a release date, but HoYoVerse leaker Ubatcha said in February that the game was "entering its final stages of development," so if this round of beta testing goes well, we may be hearing something on that front fairly soon.

Brian Raffel, who co-founded Raven Software in 1990 with his brother Steve, is retiring from game development after 36 years.

Brian has been Vice President of Raven since its inception, and Studio Head since 1997. In 2024, he began sharing this position with Dave Pellas, who joined Raven in 2011. Raffel has also served as Vice President of Activision since 1997, following Raven's acquisition by the publisher.

"It's hard to put into words what this journey has meant for me," Raffel posted on LinkedIn. "From those early days building Raven to becoming the first studio acquired by Activision, I'm grateful for the people, the culture and the games we created together. Most of all, I want to thank my brother Steve. Taking this path together and choosing Activision was one of the best decisions of our lives."

Steve Raffel retired from Raven in 2017.

Raven's first title was Black Crypt, a first-person RPG in the vein of Wizardry. This was followed a year later by ShadowCaster, a more advanced take on the idea released on DOS. But it was the studio's third game, Heretic, that would come to define its future. Published by id Software, Heretic infused Raven's dark fantasy worlds with the grisly, lightning-fast action of Doom. It was one of the first shooters to feature inventory items and the ability to look up and down, and still plays great today thanks to a recent remaster by Nightdive Studios.

Heretic's success defined Raven as an FPS specialist, further explored in games like HeXen and the gloriously gory Soldier of Fortune games. Perhaps the highlight of Raven's early period was its two Jedi Knight sequels, Jedi Outcast and Jedi Academy, which are still held in high regard thanks to their innovative lightsaber combat.

Raven's middle period brought 2009's Wolfenstein reboot and the underrated time-manipulating shooter Singularity. Since then, Raven has worked almost exclusively on Call of Duty, leading the design for Call of Duty: Warzone as well as working on the campaigns for Black Ops: Cold War, Black Ops 6, and Black Ops 7. The first two of these are among the better Call of Duty campaigns of recent years. Black Ops 7 was a mess, however, causing Activision to rethink its approach of back-to-back releases in the same series.

In a message to IGN, Doom cocreator and longtime Raven collaborator John Romero congratulated Raffel on his career and reminisced about their days working together. "Brian did a great job steering the company through those turbulent early days, transitioning from an Amiga-only studio to a PC studio that made great games. Their acquisition by Activision worked out well for Brian and Steve, and I was happy for them. It’s a rare team that can survive from 1990 to today—36 years! I definitely give credit to Brian for making that happen."

Recon is my favourite Shell to play as in Marathon, so I was surprised to learn that she is considered one of the weaker characters in Bungie's extraction shooter. This is true both for the PC Gamer team and, apparently, Bungie themselves. Yesterday, game director Joe Ziegler revealed that Marathon's intelligence-focussed Shell would be getting "more oomph", alongside other changes like similar buffs for Vandal and a nerf to Marathon's knife.

Ziegler didn't reveal the specifics behind the Shell tweaks, intending to save them for when their respective patches arrive. But in the spirit of Recon, players' desire for knowledge was simply too strong for Bungie to resist, as the developer has offered a sneak peek at the changes coming to the shooter's weakest shell.

"As part of our mid-season update planned for April 14, we've got some notable changes and updates on the way," the Marathon Development Team wrote on X. "But by popular demand we wanted to share just a few early—specifically, a preview of some of the Recon changes that are incoming."

Bungie revealed two changes coming to Recon, the first of which is a significant tweak to her Echo Pulse ability. This power lets Recon briefly scan the environment for hostiles, identifying their approximate positions via little red diamonds on the HUD. But with the changes coming on April 14, Recon's Echo Pulse will now identify whether that hostile is an AI controlled UESC robot or a human runner.

As part of this change, Bungie is making Echo Pulse less visible to enemies, making it tougher for enemy Runners "to gauge if they're in range." Depending on the extent of the adjustment, this could be a bigger deal than human/AI delineation. You can often tell the difference between UESC and human players as Recon anyway by watching how they move, but there's not much you can do about the big yellow-green wave that emanates from your position.

Enemy Runners do have a potential counter to this power boost, though. Using the Signal Jammer before being hit by Echo Pulse will make them show up as UESC, letting them potentially turn the tables on Recon. The other major adjustment is to Recon's tracker drone, which Bungie says is "about to get a bit more aggressive". Specifically, Bungie is changing it so that it can select a new target if it cannot reach its current target.

This should help alleviate frustrating moments when Recon's spider buddy gets stuck on a random piece of nearby geometry, resulting in it often blowing up uselessly and, occasionally, causing damage to yourself. I can see this change, having issues of its own if there are lots of potential targets in an area, though hopefully Bungie have already taken this into account.

All of this should put Recon on a more even standing with other Shells, and is obviously exciting for me personally. As for the changes Bungie has in store for Vandal and elsewhere, I'm sure we'll find out about those soon enough.

Take-Two Interactive has seemingly laid off its head of artificial intelligence, alongside an unspecified portion of its AI team, just two months after CEO Strauss Zelnick said the Grand Theft Auto 6 publisher was "actively embracing generative AI".

Luke Dicken joined Take-Two as its AI head in January last year following a decade at Zynga, which was acquired by the owner of Rockstar Games in 2022. Less than 18 months into the role, however, Dicken revealed that he and Take-Two were parting ways.

"It's truly disappointing that I have to share with you that my time with T2—and that of my team—has come to an end," Dicken wrote in a LinkedIn post (via Kotaku). Dicken stated that he would "make a more reflective post" upon his time at Take-Two at a later date, but for now was focused on helping support the members of his team also affected by the shakeup:

"We've been developing cutting edge technology to support game development now for 7 years. These folks know how to match innovation and novel problem-solving approaches with strong product design chops to create systems that empower people throughout the development workflow."

The nature of Dicken's role at Take-Two is not entirely clear. But according to his LinkedIn page, his position at Zynga shifted from data and strategic scientist to spearheading the company's generative AI strategy, first exploring "the intersection of AI and mobile games", before leading "various initiatives exploring and harnessing modern AI systems such as LLMs and Diffusion Models". This latter role apparently included "ensuring that all executive stakeholders across the business, including Take Two C-suite, are fully briefed on the actionable opportunities as well as the legitimate concerns".

Dicken didn't specify the exact number of people leaving Take-Two from his department, nor the reasons behind the change. But it's a further twist in Take-Two's curious relationship with generative AI, which has often appeared ambivalent and has become more confounding in recent months.

In March 2025, Strauss Zelnick stated he wasn't "worried about AI creating hits, because it's built on data that already exists", adding the technology was "backwards-looking", which doesn't sound particularly enthusiastic. Later that year, after Take-Two's stock price took a hit following the reveal of AI world model Google Genie, Zelnick further commented that there's "no evidence" generative AI tools can create great entertainment by themselves, specifically pointing out that "generative AI has zero part in what Rockstar Games is building" with Grand Theft Auto 6.

Zelnick has appeared more enthusiastic about the tech in 2026, emphasising the company's exploration of it in a recent investor call. "We're actively embracing generative AI," he said following the company's Q3 financial report, adding that Take Two has "hundreds of pilots and implementations across our company, including with our studios."

That doesn't sound like someone about to let a significant portion of his AI team go. Does this mean the pilot schemes have failed? It's difficult to say. But the broader AI industry has hit some bumps in the road of late. Last month, OpenAI discontinued its Sora video app, with Disney pulling out of a $1 billion investment deal at the same time. And earlier this week Oracle, one of the darlings of the AI boom, was reported to be laying off around 10,000 people.

Before a career second act at Nightdive Studios, the recently-retired Larry Kuperman's big project was Impulse. It was to be GameStop's answer to Steam, but it went the way of the dodo in 2014. Kuperman went into his personal history building up Impulse's catalogue when we spoke at this year's Game Developers Conference.

Kuperman came to the games industry about halfway through his professional career, with both his story and that of Impulse beginning at Stardock, a software company that was branching out into games. Stardock's management was thinking in terms of digital distribution early⁠—Kuperman started in 2001⁠—and the company was laying the groundwork for its own service.

"We reserved the rights to electronically sell the game," Kuperman recalled of his first game with Stardock, economics sim The Corporate Machine. "It was part of the contract negotiation that we could sell the game electronically. I'm sure the lawyer at Take Two [was thinking], 'It's electronic distribution. Who cares about that?' That moment was kind of pivotal."

The first iteration of this online store was a website called Drengin⁠—its archived form is a real hoot, with a throwback layout advertising the hottest games coming in 2004. "Back in those days, it was not the same game experience," recalled Kuperman. "You got this thing to download and the serial number that came in your email."

Around 2004 to 2005, when the Canadian publisher Strategy First (Jagged Alliance, O.R.B: Off-World Research Base) collapsed while working with Stardock, the software company walked away with electronic distribution rights to Strategy First's games. "That launched what would become Impulse, which was a Steam competitor from Stardock. It was a similar platform," said Kuperman.

Impulse first launched in 2008, and was then sold to GameStop in 2011. "I joined GameStop for two years as their head of electronic distribution on the PC side," recalled Kuperman."I thought that was going to be my forever job. Ironically, that didn't work out.

"I guess, back in that time, [it was] completely different management than is at GameStop now, but GameStop thought that electronic distribution was just a passing phase, and brick and mortar was going to come back strong: 'I've seen the future, it looks just like the 1950s.' But that really didn't happen."

The rest, as they say, is history. GameStop seems to have reached an equilibrium after years of turmoil: "Meme" stock manipulation, layoffs and store closures, even the untimely demise of GameInformer⁠—though that last bit does have a happy ending, at least.

It's a bit of brick-and-mortar hubris reminiscent of Blockbuster refusing to buy Netflix in 2000. Though Kuperman had no illusions of Impulse having true Steam-killer potential, the company giving up on digital distribution altogether was clearly the wrong move in hindsight.

Last month, Slay the Spire 2's controversial first beta balance patch inspired a massive backlash from its Steam reviewers, drawing over 9,000 negative reviews in the following 24 hours. In an interview with PC Gamer, Mega Crit co-founder Casey Yano said the intensity of the feedback rattled some of the studio's newer developers—but he wasn't shaken by it.

Instead, Yano said he understands the motivating impulse behind the flood of harsh Steam reviews, even if it was "unfortunate" that the tone of the feedback suffered from regional restrictions on online activity—and the nature of Steam reviews themselves.

"It's difficult for players to feel like they're heard by developers, so I understand why they choose to use Steam and do something that would impact visibility in the store to try to portray their feedback," Yano said. "I don't really get mad at players for doing bandwagoning stuff."

As the review bombardment intensified, reactions online were quick to judge its negative reviewers—the majority of which were writing their reviews in Simplified Chinese—for failing to use the Steam discussion forums or the feedback channel in Mega Crit's official Discord. In China, however, those alternate venues for offering feedback are hampered by restrictions on internet activity.

Despite the general assumption that users in China are forced to use the stripped-down Steam China platform unless they use a VPN, reports from players in China indicate that the global version of Steam is usable within the country—for some, at least. Its community features, however, are not. As a result, Steam reviews are the only direct method users in China have to offer their feedback without circumventing platform restrictions.

"I think it's really easy, if you're a player in America, to be like, 'Why don't they just use these systems?'" Yano said. "I don't really feel that way. It's unfortunate that they feel the only way to be heard is through Steam reviews."

Rather than trying to respond to the tone and intensity of the beta patch's feedback, Yano said Mega Crit is more interested in improving lines of communication with Slay the Spire 2 players in China.

"We're all just people, and we all have different restrictions. I don't like to say things like 'Chinese players.' In my mind, I think 'players in China,' because they're just in a different place," Yano said. "They're all just humans. They're all trying to interact with the game and be heard. Being heard is so important, and we want to improve that."

Yano said Slay the Spire 2's Chinese audience was, fundamentally, sharing the same feedback as its English-speaking players—they were just using a different mechanism, voluntarily or not, for communicating it. Ideally, he hopes Mega Crit can find a way to make those mechanisms more productive for all its players.

"Hopefully we can both react to the things that people like and make sure they're happy with the chances, and also try to improve the way we walk to each other," Yano said. "That's how I feel about it."

Ultimately, Yano said Steam reviews—and the culture around reviews more broadly—is an imperfect medium that encourages us to oversimplify how we feel about the games we're playing.

"It's just one blanket. It's one-to-five stars, or you like it or don't like it," he said. "There's many things you like and don't like about everything. And I think that nuance is extremely important."

This is April 2, which means April Fool's is over and we can put the whole thing behind us until 2027. Except, well, this was just now brought to my attention, and I must insist you bear witness to it. Because this is The Emperor Protects: A Warhammer 40,000 Musical.

"Experience Warhammer like you've never heard it before," the YouTube page states, and boy, it is not kidding. Seeing fully decked-out space marines and battle sisters belt out timeless hits like Suffer Not the Alien To Live, My Collection, and There Is Only the Emperor—"the ballad of the Dark Millennium"—is a stirring sight by any measure.

"Witness the spectacle of galactic cataclysm as it's never been seen before!" the announcer exclaims, and yeah, I think that's a fair statement. I've certainly never seen anything like this.

My first thought, sadly, was to wonder whether this was put together with generative AI, but fortunately that doesn't appear to be the case: A couple people in the comments say they recognize the players, including the battle sister and the necron—Trazyn the Infinite, apparently—from cons. So that's good to hear.

The thought that immediately followed was, briefly, tougher to answer: Is this real? The answer, it seems, is no—but also, well, maybe?

The link for ticket information leads to a video on the making of the musical, and as good as the trailer is, this is where the magic really happens because it's played completely straight as a behind-the-scenes look at a real musical production, including rehearsal footage and commentary from the producers, Adam and Eddie. (You'll probably need a VPN to see the video—when I tried to access it from Canada, I landed on a blank warhammertv.com page, and could proceed no further.)

"I think there's some parts of Warhammer 40,000 that can only be expressed through music," Eddie says at one point in the video.

"A lot of people thought that maybe an upbeat, very excited, poppy song wasn't appropriate for an alien of infinite evil. But we made it anyway."

The confusion—and look, I don't really believe this is happening, but I keep my heart open to all possibilities—comes at the end of the video, which says The Emperor Protects: A Warhammer 40,000 Musical opens on April 1, 2027. So even though this surely isn't happening, it seems clear that something is going to happen next April Fool's: Games Workshop's commitment to the bit is way too deep at this point to just let it all slide, right?

The other reason I leave open the possibility that this might be turned into a real thing is that, entirely predictably, the Warhammer community wants it. Comments on the YouTube video are packed with people doing their utmost to manifest this thing into reality. If there was someplace they could buy tickets, they'd already be standing in line.

And you know—why not? They've already got three banging songs, incredible costumes, the foundation of a cast that's clearly up to the task, and official grimdark lore and musicality consultants. So, why not?

As Eddie puts it: "Just because there is only war, doesn't mean there can't also be music." I can't believe I'm saying this, but I am really looking forward to April Fool's 2027.

Just in case you missed it yesterday, here's our roundup of the best April Fool's gags of 2026. None of them are as good as this, I'm afraid, but some are pretty good.

When I heard Call of Duty: Modern Warfare 2019 was back in the limelight thanks to an unusually steep $6 sale on Steam, I wasn't too surprised. Infinity Ward's Modern Warfare reboot was extremely popular in its day, reinvigorating a series that tread water through the mid-2010s and ushering in an era of followups that'd build on its reimagined progression and class systems.

MW2019 was the best these games had ever played, looked, or sounded at the time. Coming back to it now, it's obvious Call of Duty hasn't been meaningfully better since.

The first thing I noticed jumping into a match in 2026 (which only took a few seconds, as it's still popular) is how knockout gorgeous this thing still is. Like seriously, look at this seven-year-old videogame next to 2025's Black Ops 7:

Both look good, but Modern Warfare immediately wins on atmosphere. The cracks in the asphalt, the way those tree leaves are catching the light, the wildly detailed texture work on the aging buildings—what black magic was Infinity Ward up to making this work on an Xbox One?

Sound is still a standout, too. Assault rifles erupt with ear-splitting explosions followed by steel-on-steel cycling, and my kitted out revolver from launch sounds more powerful than its equivalent in BO7. Guns are appropriately loud in the sound mix—so loud that they can drown out the barks of the in-match announcer or interrupt the level-up guitar riff. Last night a teammate firing a Kar-98 rifle from behind caused me to violently flinch in my chair.

I'm reminded that this was Call of Duty redoubling down on grounded "realism" after years of wallrunning and jetpacks. Infinity Ward harnessed the immersive presentational qualities of a milsim within the confines of an arcadey shooter, and it's just as effective now as it was then. It's worth pointing out that this past streak of games helmed by Treyarch, Sledgehammer, and Raven just do not sound or look like this, even when it's clear they're trying to get close.

When folks say Infinity Ward's games are better made than Treyarch's, this is the stuff they're talking about. This game really checked all the boxes—excellent gunplay, memorable maps, sticky progression, and a campaign that doesn't suck (though that Highway of Death mission sure does).

Playing the last few nights, I'm taken by Modern Warfare's subdued Create-a-Class. You only get the basics—three perks, grenades, and two guns with five attachments. It's a bit limiting compared to the Swiss-army-knife-like monstrosities you can throw together in BO7, but that's appealing in its own way. When I can only modify a gun in so many aspects, I'm forced to embrace its intended strengths and weaknesses.

There's this observable pattern in competitive shooters with customizable guns, but especially Call of Duty, where everybody wants to create the same laserbeam. Whatever combination of grips, stocks, or barrels that lower recoil while juicing velocity or fire rate makes that gun objectively better, and therefore the most popular configuration. Modern Warfare's attachments are designed so that you can't boost those key stats without accepting tradeoffs in other areas like movement speed or ADS time.

Treyarch and Sledgehammer have loosened those restrictions on attachments in recent years, allowing up to 10 slots on each gun and mostly eliminating the downsides of making a laserbeam. The result is that every gun in Black Ops is pretty much the same. Modern Warfare isn't perfect in this regard either, but its launch lineup of weapons at least have standout qualities, like the M4A1's balanced damage and range offset by jittery visual recoil.

It's also nice to play Call of Duty from a time before it became a clown show of ugly skins. Modern Warfare still mostly looks like Modern Warfare, and it may be the last Call of Duty we'll ever be able to say that about. Modern Warfare 3 and Black Ops 6 will be forever besmirched by Nicki Minaj, American Dad, TMNT, and The Boys collaborations. Even after committing to chilling out on the tacky crap in Black Ops 7, Activision has proven this was a flat-out lie, with recent collaborations including Fallout and Dave Chappelle (yikes).

The irony is that MW2019 is both the last great Call of Duty and the one that kicked off its new worst era. Back in 2020, just a few months after it launched, Warzone hit the scene. A year later, the reboot was forcibly gobbled up by the battle royale behemoth that Infinity Ward co-created before giving the keys to Raven and saying "see ya."

We're still in this period where "premium" Call of Duty multiplayer is intertwined with Warzone—sharing weapons, classes, and roadmaps—in ways that are frequently a detriment to both games. The ultimate example of this is the modern Call of Duty launcher: a slow, grating layer of annoyware that inflicts extra steps upon those seeking to play it.

The only mercy today is that MW2019 was eventually decoupled from Warzone back into its own standalone product, though for some reason you still have to select it from a smaller launcher that also includes Modern Warfare 2 (2022), even if you don't own it or want it.

It speaks volumes of this particular chapter of Modern Warfare that it's worth the launcher hurdle. Seven years later, this is still the Call of Duty to beat (in retrospect, even better than the sequel I gave a higher score).

I imagine this is the sort of rekindled excitement Activision was angling for with the big discount, as 2026 is Infinity Ward's chance yet again to drag the series it created back to the forefront.

Linux is something that has been on our minds, and honestly on our gaming PCs quite extensively over the past few months. Since Valve dropped it into the Steam Deck, and Proton became the way forward for gaming on Linux distros the world over, there has been increased interest in the open source operating system. Is it a viable alternative to Windows, can it actually cope with gaming on high-end hardware, will you be stuck copy and pasting random code into a terminal from here until the end of time?

Yes, yes, and probably.

But there are other questions, the most pertinent is one that we put to the you, our dear readers over the past week: Would you, will you, are you making the switch to PC gaming on Linux?

For the hard of peeping among you, the results from that frustratingly ill-compressed image are as follows: 29% of the responders to the week's poll said they had already made the switch over to gaming on Linux; 26% said they actually don't mind Windows after all, and 14% said they planned to make the switch this year.

Then 10% of respondents each said they were either waiting for better multiplayer game support—Linux still suffers from being locked out of any game that has kernel-level anti-cheat support—or they were waiting for the Steam Machine to come out. Bon chance waiting on that.

Bringing up the rear, so to speak, we've got 6% saying that no, they don't like change, with 5% of respdonents saying they've been there already and got burned. Hey, me too, and I still went back.

It is obviously worth noting that the respondents add up to a grand total of 2,333 people, so the sample size is limited. Which means it behooves you to jump into the comments and make a noise about where you stand on this great debate.

We did have another quiz running last week, too, asking the question: Could Nvidia convince you to use DLSS 5? I'll give you the results to that soon enough, but I think you can probably guess how that one went already...

This week we want to know what resolution you game at. Are you a 4:3 Counter-Strike sicko, have you stuck resolutely to 1080p, or are you all in on ultrawides? Let us know.

Larry Kuperman made his mark on the industry as part of the team at Nightdive, the games preservation and remaster specialists, but before that, he was on the front line of the digital distribution wars from the early 2000s to 2013. He helped build up the online storefront, Impulse, which was later acquired (and ultimately shuttered) by GameStop.

Valve's unlikely victory against titans like EA and Microsoft has always fascinated me, and I asked Kuperman for his take on Steam's ultimate victory when we spoke at this year's Game Developers Conference.

"The idea was coming up to all of us," said Kuperman. "Let's also remember that Steam really began as a visual way of finding your Counter-Strike server." Steam had a number of early rivals in digital distribution⁠—Kuperman shouted out Paradox's GamersGate (don't say it)—but he thinks Steam was quicker on the draw to sell other companies' games on its own platform.

"The idea of selling games, and then selling third party games, didn't seem intuitive at the time," Kuperman said. "You're a game company, why are you selling other people's games? That was a hard thing to understand. But Gabe really had a great vision, coming out of Microsoft."

Another advantage, according to Kuperman, was Steam's "stickiness" and embrace of social elements. He noted that Steam and Impulse both let you redownload your games without restrictions, something that wasn't always guranteed⁠—I still have a GamesPlanet receipt from 2008 that would only let me redownload Fallout 1 and 2 through the service six times a piece.

But Steam also had friends lists, messaging, playtime stats, and those little popup notifications about what your buddies were playing. "What Steam did better than anybody else was to create a community," Kuperman argued. "They established a stickiness to it, that people came back because it was Steam."

There's one more decision by Valve that Kuperman cited as key to Steam's victory: Its relatively low barrier to entry. There were so many great games in the 2000s, it's easy to forget all the small to mid-sized studios that shuttered during that era, particularly PC specialists who couldn't make the jump to consoles. Shareware is dead, digital distribution is yet to be born, now is the time of monsters.

"If you did not get a retail buyer to pick up your game," recalled Kuperman, "If your game wasn't at Walmart, GameStop, three or four retailers⁠, you were done. You didn't make a game. Games that were kind of unusual and quirky, that broke them.

"Steam's philosophy of, anybody can put their game on it—for a price, but it's not a significant price—that really changed the gaming world. … I think that probably the biggest thing that you can say about Steam is that, for a number of indies, it kept their company alive when they would have otherwise gone under."

But Kuperman attributes the survival of his own indie studio to a different storefront that emerged in the same ferment. "If it wasn't for GOG, there would have been no Nightdive," said Kuperman. "GOG was the one that put System Shock 2 up and really started our company, and that goes to our friend Oleg [Klapovskiy].

"These are relationships that we've had for years, but I don't think people realize just how important they were. I think Good Old Games is something that when you look at the history of games during this time, that they deserve their moment in the sun and the spotlight on them."

A few of my coworkers might yell at me for saying so, but I'm calling this good news. Otherside Entertainment says it's taking a "new direction" with Thick as Thieves, the multiplayer first-person stealth game it announced in 2024: Instead of being a four-player competitive multiplayer game mixing humans and AI enemies as it was originally envisioned, the studio says it is "now focused on two-player co-op and singleplayer."

"When we first revealed Thick as Thieves we intended to focus on PvPvE gameplay. However, as development progressed and the world of Kilcairn has come to life, we found that we were having more fun with solo and co-op play," Otherside wrote about the change on Steam.

"This sharpened focus has allowed us to double down on what makes Thick as Thieves truly special—dynamic stealth gameplay. We’re already seeing how this has strengthened the moment-to-moment action, whether you’re playing solo or with a partner in co-op. We’ve had some fantastic moments while testing the game. Now we’re putting all our energy into making things even better as we build up to launch."

It's something of a surprising change in direction, given Otherside's previously expressed excitement for the multiplayer component, but as I said, I'm calling it a good thing. Otherside's top guys are Paul Neurath and Warren Spector, two groundbreaking immersive sim designers who, ironically, played key roles on pretty much every immsim of note in the '90s except Thief. To have them leading a modernized take on that series (which I dearly love, even the third one) is a tantalizing prospect, but I don't think a competitive multiplayer environment would really play to its strengths, and having to accommodate that aspect of the game would inevitably detract from the elements that could make it truly shine.

Some of my colleagues are disappointed by the change (which is where the yelling at me comes in), and admittedly, multiplayer Thief is an intriguing idea. Even so, I think the overall sense here is that this is probably a good idea. Going right back to when Thick as Thieves was first revealed, PC Gamer's Evan Lahti said he was "enthusiastic" about the game, but "also skeptical that it can attract millions and millions of players, even if it walks and talks like a modern service game, as OtherSide plans."

The fact is that immersive sims, despite being objectively the best genre, remain a niche interest, and making a successful live service game is notoriously difficult; put them together, and yeah, you've got a real challenge on your hands. A singleplayer Thief-style stealth-em-up isn't exactly purpose-built for mass market interest either, but given how so many live service games have struggled briefly and died in recent years, I think it's a safer bet overall—and I think it will also make for a much better game.

But for those who are disappointed about the change, there is hope: In a response to inquiries in the comments, a developer from Thick as Thieves publisher Megabit said "PvPvE may come further down the line, but at launch we're looking at singleplayer and co-op modes only."

Possibly the bigger news amidst all of this is that we're finally going to get a look at Thick as Thieves in action. Otherside has been pretty quiet about the project since that 2024 reveal, but a new gameplay trailer and "some other well-guarded secrets" will be shared at the upcoming Triple-I Initiative showcase, which will take place at 9 am PT/12 pm ET on April 9.

In an interview with PC Gamer, Mega Crit co-founder Casey Yano shared three mysterious plans for new modes that Slay the Spire 2 could see during the next year or two of planned early access development.

The roguelike card game is already bigger than its predecessor—the co-op mode alone is a major addition for what is primarily a singleplayer genre—but it sounds like Mega Crit will be running at least a few mad science experiments to see whether they pay off.

"There are three specific things that I have in mind, which I'll try to explain in a vague way to not give away too much," Yano said.

"The first one is a mode that is meant for people who want to play in a very competitive fashion. The second is for people who want to have the Slay the Spire experience, but do not have the time to do so. And the third is 'What if there were other ways to interact with people, socially or in a multiplayer-like setting, with the current systems that exist today?'"

So what could each of these ideas be?

Slay the Spire 2, like the first game, already features a daily seeded challenge mode, with a leaderboard that logs players' performances. Perhaps this new mode will pair players together for daily runs? Or consider this left-field theory: two players climb the same tower at once, and if their paths converge they have to duke it out.

The sped-up StS experience sounds like a challenge mode that would either condense the experience as a whole or hand you a specific build—or perhaps allow you to pre-select one from your unlocked cards and relics—and then throw you into a series of battles, with no choices to make in between.

And for the third mode concept? Maybe you could drop into someone else's game, Dark Souls-style, just to help them defeat a boss? I'm not sure that jibes with the "current systems that exist today" constraint, but it could be pretty cool.

It's also possible none of these modes ever end up in players' hands. "Not many of our ideas survive the process," said Yano, speaking generally about the studio's open-ended ideation.

"Those are the three ways to interact with Slay the Spire that don't exist right now," Yano said. "Maybe we'll find a way to do that, or maybe having those modes dilutes the core experience and might not make it at all. That's something we need to try to find out."

But it's also possible Mega Crit comes up with something Yano hasn't even mentioned, because co-founder Anthony Giovanetti has just as much input into design.

"I have thoughts, but even Anthony doesn't know what those thoughts are," Yano said later in the interview, specifically around new characters that could be added to Slay the Spire 2. "We're kind of competitive when it comes to the design."

That approach has worked out so far: Slay the Spire 2 has had the most successful launch of any deckbuilder, ever, and it's not even close.

Every morning, Samson wakes up with an enormous problem pressing on his skull: a debt he can't afford to pay back. Either he can take extreme measures to reduce the amount he owes—driving getaways, extorting businesses, taking down rival gangsters—or watch the total grow day by day. Should he ultimately fail to come up with the goods, he and his sister will pay in blood.

High stakes are common in the storytelling of open world crime sagas, and yet they rarely hand that pressure to the player to deal with. Samson, which will be out on Steam and EGS on April 8, is different. Each morning you're given a financial target to meet, and a set number of action points with which to plan your day's activities. Story missions battle for priority with lucrative jobs, and abandoning a gig might lead to debt collectors gathering by your car, ready to knock the cash out of you like a piñata.

It's a rough-and-tumble, roguelike-adjacent structure I've never encountered in the genre before, and was created purely out of necessity. "The reality of the industry hit us about a year ago, and we laid off half the team," says former Just Cause and Mad Max lead Christofer Sundberg. "We didn't 'restructure', we didn't do some 'organisational optimisations' or whatever everyone else called it, we laid off half the team. And those were our friends, and that hurt."

Before that moment, Samson had been intended as a much bigger action RPG—but a newly limited team and budget forced Liquid Swords to take a butcher's knife to its plans. "We just shelved those features and saved them for the next game, depending on the success of this," Sundberg says. "So a lot of those more heavy RPG systems sit there waiting to get back into the game. Maybe for a sequel, or for post-launch—I don't know, really."

It's during this rethinking of Samson that day-by-day debt reduction became the structure of the game. It's something Sundberg largely credits to one of the project's designers, Niklas Norin: "a huge fan of Elden Ring and those difficult games people my age can't play".

It has to be said that, though born from dire circumstances, Samson's new form makes it a far more intriguing proposition. Its setting, the dilapidated and drug-ridden city of Tyndalston, was already an oppressive backdrop of tall buildings and tight corners; now its core loop matches the environment, and offers something new in a genre too often in thrall to GTA. "It's become more unique," Sundberg says. "The original game was a little bit more, I wouldn't go as far as Led Zeppelin, but this is definitely the Sex Pistols."

Fists over bullets

Originally, Samson's skirmishes were built around both fistfights and ranged encounters. "But then when we had to down-scope the game, we took out gun combat, and that made it more interesting," Sundberg says. "It's really come out in a great way, because it allowed us to focus more on hand-to-hand. I think we've had enough running and gunning games where you just pull the bazooka out of your trench coat or tank top or whatever, and it makes things so easy. This game is not about making things easy."

Liquid Swords invented its own gun laws for Tyndalston—which means that only the most dangerous gang enforcers and cops have access to them. "When someone pulls out a gun in the game, it's trouble," Sundberg says. "The best thing you can do is to run. I've found it quite fun actually, to run through the alleys of Tyndalston. It's like a big maze, but you can climb any fence, run anywhere. It's a good, strategic way to escape when you have to, and also a way to explore."

When Samson's recent trailer dropped, PC Gamer writer Rick Lane expressed some skepticism about the brawling, and the number of times enemies appeared to rely on the same attacks in sequence.

"There is depth, but it is repetitive at some point," Sundberg admits. "It just came down to bandwidth. What I've found is you have to be really creative in the way you are using your combat moves and objects in the environment. When I get swarmed by a bunch of enemies, I always try to push my way backwards, sometimes to take painkillers, Max Payne style, but also to grab a bottle or a brick." Lob something heavy at an opponent and they'll be briefly stunned, leaving you an opening. "So yes, it is a bit repetitive, but the more you play, the less repetitive it gets, because you realise you have to strategise."

Samson's driving looks almost as pugilistic, as cars collide with intent and crumple during nasty collisions. It makes sense that there's Avalanche DNA here—namely designer Alex Williams and programmer Josef Sundberg, who were "the leading people in the team that made the vehicle combat in Mad Max".

"The part that we have allowed to slip in from Just Cause is that strange things can happen," Christofer Sundberg says. "When you're chasing someone, they'll sometimes drive into a pillar and explode, and the mission is successful, and not because of you. Because your enemy made a mistake. That's what I think makes open world games so much fun—that you can use the game's flaws to make it fun."

Inevitably, Samson has been dogged by comparisons to Grand Theft Auto. "There are times when you want to put GTA down and pick up something else," Sundberg argues. "And so I see Samson as being like back in the day when action movies were 90 minutes long, not over two hours. I keep on going back to watching Die Hard and Ronin and First Blood and Rambo. I think there's a space for us there, and that's where I want to end up."

It's not just fans who've repeatedly brought up Los Santos in relation to Samson; potential backers have done so too.

"We've obviously tried to find a publisher for years, and we've always heard the GTA fear," Sundberg says. "And then when we looked at the metrics of how much they've played the game, they've only played Samson for 20 minutes. It's 10 minutes past the tutorial, almost. And they're not going into every detail that we have that is unique to our game. It's just, 'Oh, because it's set in a city and it's an urban environment with cars and violence, then it has to be GTA.'"

One publisher rejected Liquid Swords on five occasions. "I bumped into their business development guy at DICE," Sundberg chuckles. "And I asked if we could have a sixth meeting to be rejected once more before we reach the finish line."

Samson is now approaching its April 8 launch at speed. But at times, Sundberg has the air of a man still discovering the joys of his own game. He talks happily about figuring out a way to ditch his beaten-up car during a chase in favour of a faster ride—which can be found reliably parked at Tyndalston's gas stations. "It's a fun balance that I don't even think was deliberate," he says. "But it's good fun. It's a diamond in the rough."

Samson won't be bug-free on release: "We're making it the best it really can be, we can't do more than that". And despite the imminent deadline, Liquid Swords aren't crunching. "We're too old for that," Sundberg says. "So we're super effective during our work hours and then come back the next day. It's like the game, almost."

Nevertheless, there's clearly a lot of pressure. "It's more up here than anything else," says Sundberg, pointing to his head. "I've never been this nervous for a game launch, ever, in 33 years. It's too much at stake. I put my chin out pretty far for the studio and for this game. I made a promise to the team that I couldn't keep a year ago, and now I made a new promise to the team that are just working their asses off to finish this game." What was the new promise? "That we would get this game out."

In the last year, Liquid Swords has taken on a mantra often attributed to the US Marines: improvise, adapt, overcome. "For me personally, it has never been an option just to shut the doors," Sundberg says. "That money that we were asking for before we had to let go of everyone, it doesn't exist anymore. We were asking for around $45 million, which in 2021 and 2022, post pandemic, was too little to ask for. But those days are definitely over."

It's not at all clear to me how things are going to go for Liquid Swords. "The rough ride is not over," Sundberg tells me at one point. "So we'll just hang in there and put chewing gum where the water leaks are in the hull." But it's worth considering that, during recent games industry volatility, thousands of developers have not only lost their jobs but watched their art, code and design vanish into vaults. Locked behind the NDAs of cancelled projects, their hard work will never be resurrected, shown off or appreciated. By making it to launch, the remaining Samson team has at least allowed that work to see the light of day.

"In this day and age, every game released is a victory," Sundberg says. "And, not saying that we have the perfect recipe, but I wish that more developers in our situation could let go of their pride. Take what you have, and then just make the best out of it. If you're being honest, then you will have the respect of the community as well."

Russ Bullock, the CEO of MechWarrior developer Piranha Games, has confirmed that the studio has imposed layoffs, saying that roughly 30% of its employees have been let go.

The layoffs first came to light on LinkedIn (via GamesIndustry), where multiple employees reported being let go; at least two of them said the layoffs came as part of an "economic restructuring" at the company. While Piranha itself hasn't made any announcement about the layoffs, Bullock confirmed the cuts in posts on X.

"Not accurate," Bullock wrote in response to a claim that 60% of the studio had been let go. "It was 30% of the staff which was a horrible day no doubt but we do still continue to work on DLC."

Bullock told PC Gamer that 17 employees were let go in this round of layoffs, but said the MechWarrior 5: Mercenaries DLC it announced in March is still slated for launch in May.

"We are also working on another MW product for release this fall," Bullock said. "Tough week but we continue to adapt to the significant changes in our industry and continue to ship ever-improving products.

"No doubt it is more difficult each year to see very far down the road and what might be in store, but for now we continue to develop and release products we're proud of."

The longer-term future of MechWarrior, at least in its current iteration, seems less certain, however. "I honestly don't know how long we will be able to keep the ship sailing," MechWarrior 5: Clans narrative director Chris Lowrey wrote in another thread confirming the cuts.

"It's been a small miracle that Mercs has received five years of post-launch support as it is. (Very few other single point of purchase games get that amount of post-launch support in our industry.) But those of us still around are still committed to continue support of MechWarrior as long as we're allowed to do so."

This is the second round of layoffs at Piranha in just over a year—parent company Enad Global 7 laid off 38 Piranha employees in January 2025, just a few months after the release of MechWarrior 5: Clans. We gave Clans an 82% score in our review, saying it "delivers a strong story of struggling young warriors alongside the best-ever version of its iconic, lumbering mech combat," but Piranha parent company Enad Global 7 said it "failed to reach new audiences and expand its core audience as anticipated and therefore has not met the necessary sales targets," which resulted in the layoffs.

Amidst yesterday's yearly crop of April Fool's japery, Marathon game director Joe Ziegler broke from seasonal convention on X by sharing an early, fool-free preview of balance adjustments Bungie has planned for upcoming patches in the weeks ahead.

With tweaks planned for weapons, attachments, deployables, runner shells, and more, Ziegler said the wide-ranging slate of updates is intended "to adjust some of the outliers in the combat space that we've both seen in our observations and heard from your feedback across all of the various (and plentiful) channels we watch."

Gear balance seems like a major focus of current Marathon retuning efforts. Bungie's concluded that bubble shields and thermal scopes are too accessible for how dominant they've proven to be in firefights, Ziegler says. Likewise, Snipers—which Ziegler says are "too strong in all scenarios"—are due for a nerf.

Also primed for a depowering is the knife, which Ziegler has acknowledged is currently "scaling too easily to god-like power levels." Considering it only takes a couple upgrades before you're better off bringing a knife to a Tau Ceti gunfight, that's probably a safe call.

Some runner shells, meanwhile, will be getting the opposite treatment, as Vandal and Recon players should soon be enjoying "some more oomph to match their counterparts." Hopefully that extra bit of oomph will mean my tracker drone spends a bit more time exploding anyone but me.

Ziegler also teased a pair of secret new features Bungie is adding to inject some more variety into how firefights can pan out: a "new item to give solos a more common option when downed" and "a new item to give a more merciful option between crews." I'm not sure what mercy might look like when Marathon's interactions skew so hostile, but I'm interested to see it.

We should start seeing these balance changes arrive before long, as they'll be split between Marathon's next patch and another that will follow a week later. And that's evidently not all Bungie has in store, because Ziegler says there will be "a lot more goodies in the coming patches," promising more details when the inevitable patch notes arrive.

"Thank you all again for keeping the feedback flowing," Ziegler said. "We've still got a big list of things we're looking into that extends beyond this that will take us more time to figure out and we'll update you as we continue to build on our evolving plans."

Watch the full interview video above on the PC Gamer YouTube channel, if you prefer.

There are many goofy placeholders across Slay the Spire 2, MS Paint-style art pieces that stand in for the finished products that will come later as development progresses through Early Access, which Mega Crit says the game might take between one or two years.

The deckbuilder's ending is one of the biggest temporary pieces, with all runs culminating in a cutscene with a mysterious figure called The Architect, presumably the designer of the spire. The Architect exhibits indifference to your appearance at the end of Act 3, shrugging off damage before insta-killing you. An end screen then displays the aggregate damage players have dealt to The Architect, which now totals more than 99 billion. It's all very "To be continued..." in presentation.

Speaking with Mega Crit co-founder Casey Yano about what form the eventual ending to Slay the Spire 2 will take, the studio understandably isn't revealing narrative details, but did provide some thoughts on its process and timing.

"I do want there to be a better ending," Yano tells me over a video call. "I do want it to be more extravagant when you win. And I do want the showdown to also be a pretty big deal, we'll say."

Yano says that the ending(s—he used both singular and plural when speaking to me) aren't completely set in stone, and adds that he deliberately wanted them to be shaped by players' responses to the initial launch.

"I do want the endings to be pretty, pretty serious business, but they aren't too designed at this time. I wanted to see the reaction that people have towards the narrative elements to the game that we've brought in. I like to read what people's theories are on what the characters are, if they're good or bad, and things like that. And so we're kind of just seeing what people like, what characters people are attached to. And while we do have kind of like a canonical idea of what is going to happen, we do make minor tweaks to make sure that the ending kind of meets everyone's expectations, and also a little curveball, you know, every now and then, but positive curveballs."

The first Slay the Spire culminated in Act 4, which players had to unlock by completing Act 3 with all three characters. Players then had to collect three keys as they progressed on new runs, opening a new path to a challenging miniboss that preceded the true final boss of Slay the Spire, The Corrupt Heart. Though Slay the Spire 2 inherits many aspects from the first game, it's unclear whether the sequel will adopt this same structure.

Fittingly, Yano suggests that an updated ending might be one of the later things to be added to Slay the Spire 2. "I get the feeling it's going to take a good while for the ending to show up, just because, you know, we have a few other types of content that we want to make. We want these alternative acts and the characters, we want to experiment with the game modes. I think once those are done, we'll probably think about finalizing an ending, probably as we go into V1."

Other roguelikes have taken the approach of adding an ending later in development, notably Hades and Hades 2.

In a somewhat ill-timed announcement (I promise it's not an April Fools joke), the Jak & Daxter-porting OpenGOAL team revealed its complete PC port of Jak 3 yesterday, completing its years-long project to natively port the entire trilogy to PC.

It still requires a legally-obtained copy of Jak 3 to run, similar to an emulator or unofficial porting projects for other games, but OpenGOAL offers a far superior experience to emulation or even original hardware: It's less resource-intensive, has fewer (if any) bugs, and offers more configuration options and modding possibilities (ultrawide ahoy). I used OpenGOAL to replay the first two Jaks on my Steam Deck a few years ago, and they ran like a dream.

OpenGOAL seems to primarily be an exercise in curiosity, mastering the titular programming language, Game Oriented Assembly Lisp. It was developed by Naughty Dog and exclusively used in the Jak trilogy and its spinoff, Combat Racing. Notably, the PSP spinoffs, Daxter and The Lost Frontier, were not made with GOAL or by Naughty Dog, so the OpenGOAL team won't be touching them.

They haven't ruled out Jak X: Combat Racing entirely, though. The team's somewhat out-of-date FAQ page says that "As of now, it's unsure if Jak X will be decompiled." But that seems to have been written before Jak 2 OpenGOAL was finished all the way back in 2023.

Never say die, then, if you're one of the 12 or so people in the world who are really passionate about Jak X: Combat Racing preservation. As for me, I've got to find the time to revisit Jak 3⁠—these are some foundational games for me, and they've got an aesthetic, a vibe, that you really can't find anywhere else. I'd postponed completing my big trilogy replay until the OpenGOAL team's work was done.

Jak and Daxter has aged wonderfully. Jak 2, well, I think the mockery of its edgelord turn is overblown⁠—this is still a weird and whimsical world⁠—but the game treats checkpoints like they're the last precious doubloons of a squandered fortune, begrudgingly sliding one your way every few weeks or so. Jak 2 is frustrating. I hope Jak 3 is better in that regard. I don't remember either way⁠—I literally haven't played it in 22 years.

If you've got your PS2 copy of Jak 2 handy, you should be able to copy an OpenGOAL-ready disk image of it with most Blu-ray drives and a utility like MPF or ImgBurn⁠—both recommended by the PCSX2 emulator community. From there, OpenGOAL has an installer and instructions on its website, including for Steam Deck.

Everything changed when the fire nation attacked… Honestly, that's what it feels like writing this review. Teamgroup's MP44Q PCIe 4.0 SSD has finally landed with me for testing, and it's a curious little QLC number that, once upon a time, would have offered itself up as a nice, affordable, tantalizing opportunity for any would-be budget gamer's rig. The problem is that it's 2026, this sorta QLC feels a bit old fashioned, and SSD prices are through the roof.

It's especially tough as we've seen even superb budget drives such as the WD Blue SN5100 totally hampered by an extraordinarily high price tag. And that's a problem, a real big one, especially for Teamgroup. Side-by-side these drives, although targeted at the same audience, and with similarly high price tags (at time of writing the MP44Q is $305 and the SN5100 is $317), perform wildly differently.

The MP44Q is a sleek little budget PCIe 4.0 NVMe SSD. It's a single-sided M.2 2280 design giving it broad compatibility with all of your consoles, laptops, and PCs, and comes complete in 1-4 TB capacities with no heatsink variant available.

As for that core hardware, you get a Maxiotech MAP1602A controller (a slightly older model, and the same one found inside of the Lexar NM790), and die packages built out of YMTC's 232-layer 3D QLC NAND flash (you "might" find similar density YMTC TLC in Biwin's Black Opal NV7400, or Micron flash who knows…). There's no dedicated DRAM cache here, mostly to keep costs down, but you do get a healthy dose of 40 MB of system memory allocated via HMB, and a big chunk of pseudo SLC, and that's kind of it.

There's no drama, no new flagship parts or anything, just off-the-shelf OEM components—tried and tested and ready to go. What is interesting, however, is the endurance rating, as Teamgroup's got this clocked at 1000 TBW, which for a budget QLC drive is quite impressive. The default rating for most SSDs these days is 600 TBW for 1 TB, and practically all manufacturers abide by that, so it's nice to see an improvement here. Although you still get the same 5-year warranty anyway.

As for pricing, at the time of writing, it's kind of on the high side for a budget drive, especially compared to that SN5100. The reason I've been slamming that comparison so hard in this review is purely because, by the metrics that matter, the SN5100 outclasses the MP44Q in almost every way while being cheaper as it does it.

Benchmarks: SSDs

TemperatureCrystalDiskMark 7.0.0Final Fantasy XIV: ShadowbringersCrystalDiskMark 7.0.0 | RND4k Q1T1 | IOPS3DMark Storage

Teamgroup MP44Q 2 TB l Gen4

54

Sandisk WD Blue SN5100 2 TB | Gen4

61

Lexar NQ780 4 TB l Gen4

62

Biwin Black Opal NV7400 2 TB | Gen4

48

WD Black SN7100 1 TB | Gen4

48

020406080

Peak temperature (°C)

Temperature Data

Product	Value
Teamgroup MP44Q 2 TB l Gen4	54
Sandisk WD Blue SN5100 2 TB | Gen4	61
Lexar NQ780 4 TB l Gen4	62
Biwin Black Opal NV7400 2 TB | Gen4	48
WD Black SN7100 1 TB | Gen4	48

SEQ1M Q8T1 Read (MB/s)

SEQ1M Q8T1 Write (MB/s)

RND4k Q1T1 Read (MB/s)

RND4K Q1T1 Write (MB/s)

Teamgroup MP44Q 2 TB l Gen4

7407

6595

70

288

Sandisk WD Blue SN5100 2 TB | Gen4

7318

6687

107

308

Lexar NQ780 4 TB l Gen4

7446

6601

70

310

Biwin Black Opal NV7400 2 TB | Gen4

7370

6336

75

229

WD Black SN7100 1 TB | Gen4

7116

6820

101

276

02,0004,0006,0008,000

CrystalDiskMark 7.0.0 Data

Product	Value
Teamgroup MP44Q 2 TB l Gen4	7407 SEQ1M Q8T1 Read (MB/s), 6595 SEQ1M Q8T1 Write (MB/s), 70 RND4k Q1T1 Read (MB/s), 288 RND4K Q1T1 Write (MB/s)
Sandisk WD Blue SN5100 2 TB | Gen4	7318 SEQ1M Q8T1 Read (MB/s), 6687 SEQ1M Q8T1 Write (MB/s), 107 RND4k Q1T1 Read (MB/s), 308 RND4K Q1T1 Write (MB/s)
Lexar NQ780 4 TB l Gen4	7446 SEQ1M Q8T1 Read (MB/s), 6601 SEQ1M Q8T1 Write (MB/s), 70 RND4k Q1T1 Read (MB/s), 310 RND4K Q1T1 Write (MB/s)
Biwin Black Opal NV7400 2 TB | Gen4	7370 SEQ1M Q8T1 Read (MB/s), 6336 SEQ1M Q8T1 Write (MB/s), 75 RND4k Q1T1 Read (MB/s), 229 RND4K Q1T1 Write (MB/s)
WD Black SN7100 1 TB | Gen4	7116 SEQ1M Q8T1 Read (MB/s), 6820 SEQ1M Q8T1 Write (MB/s), 101 RND4k Q1T1 Read (MB/s), 276 RND4K Q1T1 Write (MB/s)

Total Loading Time (seconds)

Scene #1 (seconds)

Scene #2 (seconds)

Scene #3 (seconds)

Scene #4 (seconds)

Scene #5 (seconds)

Teamgroup MP44Q 2 TB l Gen4

7.97

0.961

2.166

2.535

1.513

0.795

Sandisk WD Blue SN5100 2 TB | Gen4

7.125

0.852

1.832

2.284

1.425

0.731

Lexar NQ780 4 TB l Gen4

7.869

1.127

2.072

2.465

1.461

0.745

Biwin Black Opal NV7400 2 TB | Gen4

7.398

1.112

1.813

2.286

1.445

0.742

WD Black SN7100 1 TB | Gen4

7.841

1.831

2.411

1.498

0.755

1.347

02468

Final Fantasy XIV: Shadowbringers Data

Product	Value
Teamgroup MP44Q 2 TB l Gen4	7.97 Total Loading Time (seconds), 0.961 Scene #1 (seconds), 2.166 Scene #2 (seconds), 2.535 Scene #3 (seconds), 1.513 Scene #4 (seconds), 0.795 Scene #5 (seconds)
Sandisk WD Blue SN5100 2 TB | Gen4	7.125 Total Loading Time (seconds), 0.852 Scene #1 (seconds), 1.832 Scene #2 (seconds), 2.284 Scene #3 (seconds), 1.425 Scene #4 (seconds), 0.731 Scene #5 (seconds)
Lexar NQ780 4 TB l Gen4	7.869 Total Loading Time (seconds), 1.127 Scene #1 (seconds), 2.072 Scene #2 (seconds), 2.465 Scene #3 (seconds), 1.461 Scene #4 (seconds), 0.745 Scene #5 (seconds)
Biwin Black Opal NV7400 2 TB | Gen4	7.398 Total Loading Time (seconds), 1.112 Scene #1 (seconds), 1.813 Scene #2 (seconds), 2.286 Scene #3 (seconds), 1.445 Scene #4 (seconds), 0.742 Scene #5 (seconds)
WD Black SN7100 1 TB | Gen4	7.841 Total Loading Time (seconds), 1.831 Scene #1 (seconds), 2.411 Scene #2 (seconds), 1.498 Scene #3 (seconds), 0.755 Scene #4 (seconds), 1.347 Scene #5 (seconds)

RND4k Q1T1 Read (IOPS)

RND4K Q1T1 Write (IOPS)

Teamgroup MP44Q 2 TB l Gen4

17195

70309

Sandisk WD Blue SN5100 2 TB | Gen4

26224

75158

Lexar NQ780 4 TB l Gen4

17134

75727

020,00040,00060,00080,000

CrystalDiskMark 7.0.0 | RND4k Q1T1 | IOPS Data

Product	Value
Teamgroup MP44Q 2 TB l Gen4	17195 RND4k Q1T1 Read (IOPS), 70309 RND4K Q1T1 Write (IOPS)
Sandisk WD Blue SN5100 2 TB | Gen4	26224 RND4k Q1T1 Read (IOPS), 75158 RND4K Q1T1 Write (IOPS)
Lexar NQ780 4 TB l Gen4	17134 RND4k Q1T1 Read (IOPS), 75727 RND4K Q1T1 Write (IOPS)
Biwin Black Opal NV7400 2 TB | Gen4	N/A
WD Black SN7100 1 TB | Gen4	N/A

Index

Bandwidth (MB/s)

Average Access Time (µs)

Teamgroup MP44Q 2 TB l Gen4

2874

496.38

63

Sandisk WD Blue SN5100 2 TB | Gen4

3915

672.63

59

Lexar NQ780 4 TB l Gen4

2893

495.11

62

Biwin Black Opal NV7400 2 TB | Gen4

2757

471.23

65

WD Black SN7100 1 TB | Gen4

3238

553.1

55

01,0002,0003,0004,000

3DMark Storage Data

Product	Value
Teamgroup MP44Q 2 TB l Gen4	2874 Index, 496.38 Bandwidth (MB/s), 63 Average Access Time (µs)
Sandisk WD Blue SN5100 2 TB | Gen4	3915 Index, 672.63 Bandwidth (MB/s), 59 Average Access Time (µs)
Lexar NQ780 4 TB l Gen4	2893 Index, 495.11 Bandwidth (MB/s), 62 Average Access Time (µs)
Biwin Black Opal NV7400 2 TB | Gen4	2757 Index, 471.23 Bandwidth (MB/s), 65 Average Access Time (µs)
WD Black SN7100 1 TB | Gen4	3238 Index, 553.1 Bandwidth (MB/s), 55 Average Access Time (µs)

3DMark's storage benchmark comes back with a score that's 1000 points higher straight out of the gate, with a much broader bandwidth figure, and slightly lower access times. That result is all thanks to the random 4K performance. The MP44Q really struggles here, delivering just 70 MB/s on the read and 288 MB/s on the write, making it slower than even Lexar's latest NQ780 QLC drive as well. Compare that to the 107 and 308 MB/s from the SN5100, and there's almost no contest.

Yes, it does eek back a win in the sequential read front being 89 MB/s faster (or 1.21% faster), but that's such a niche use-case for most of us gamers, it's almost not worth mentioning.

That all then translates into game load times too, and although the SN5100 doesn't particularly break any records in that department, scoring 7.125 seconds, the fact that the MP44Q lands in at 7.970 seconds is just, well, less than ideal. Physically, it's cool, mind you, and efficient as a result, but that's not enough to offset those massive deficits in the metrics that matter most.

The First Descendant's launch hype went over like a bag of candy smuggled into a movie theater—exciting at first, but regrettably short-lived. While last July it was celebrating a screaming high of 10 million players in its first week alone, SteamDB shows the game has simmered to around 5,000 concurrent players as I write this. Steam user reviews add up to a "mixed" user rating, and even Nexon's CEO has weighed in on the game's dubious trajectory.

The company discussed the game in a recent capital markets briefing to investors (thanks, GamesRadar), with CEO Junghun Lee giving clear examples of games that did well for Nexon and games that fell off. Where Arc Raiders and MapleStory got kudos, The First Descendant showed up alongside Dungeon & Fighter Mobile on a slide bluntly labeled "What Did Not Work."

Lee said Dungeon & Fighter Mobile "lost its way" after an impressive launch: "The retention mechanics weren't strong enough to hold players long-term." He noted this was the "same issue with The First Descendant: Strong launch, no staying power."

"These are design issues that are not fixed with a patch," he said. "They require structural changes to game mechanics."

The stinger is that while the presentation names a range of plans to keep the Dungeon & Fighter train moving through various upcoming games and updates, Lee doesn't mention The First Descendant again. It's not clear what the company's long-term plans for the looter shooter are, but it sounds like it'll need a more robust resuscitation than jiggle physics or the bizarre AI ads that showed up on TikTok can offer it.

Despite the big numbers it enjoyed when it first landed, The First Descendant was never universally beloved. That "Mixed" user rating goes all the way back to launch week, and PC Gamer staff writer Morgan Park called it a "dollar store Destiny" in his impressions back in 2024.

Is it just me, or does everything feel pretty serious right now? With that in mind, a little whimsy is awfully welcome. Enter the Seasonic Focus ATX 3.1 Sakura Limited Edition.

It's basically an ATX 3.1 power supply with some cherry-blossom themed styling accoutrements, an aesthetic known as sakura in Japan, which is where Seasonic is based.

Not a huge amount is officially known about the Seasonic Focus ATX 3.1 Sakura Limited Edition, which Seasonic has teased on its official X account, merely commenting, "An evolved design, with meticulous attention to detail in every aspect. That cherry blossom color, once more."

The "once more" sign off probably refers to the existing Seasonic Vertex GX Sakura Limited Edition, which also sports those cherry blossom styling vibes.

Anywho, the Seasonic Focus ATX 3.1 Sakura Limited Edition very much looks like a derivation of the Seasonic Focus GX ATX 3.1. That's available in 750 W, 850 W and 1,000 W configurations and complies with the ATX 3.1 and PCIe 5.1 standards. All three version are built into the same chassis within a compact 140 mm depth.

It sports 80 PLUS Gold certification with 90% efficiency at 50% system load, and in a claimed industry first, also offers Seasonic's patented, three-phased thermal control with Digital Hybrid Fan Control functions in three operational stages: Fanless Mode, Silent Mode, and Cooling Mode.

Fully modular cabling, including three PCIe 8-pin connectors, and a 10-year warranty are likewise part of the mix.

The Focus GX ATX 3.1 starts at just under $145 for the 750 W model, and extends to $169 and $204 for the 850 W and 1,000 W models respectively. Perhaps add a small premium on top for the new Sakura Limited Edition.

However, going by the Seasonic Vertex GX Sakura Limited Edition, which is only available in 1,000 W-spec, the full range of wattage options may not be available for the Focus ATX 3.1 Sakura.

We haven't reviewed a Seasonic PSU for a little while now. But the last unit we eyeballed got the thumbs up. There's no word on when the PSU will go on sale, with Seasonic simply saying it's "coming soon." Watch this space then. Or maybe this one.

The AI industry is often seen as pushing the bleeding edge of hardware and tech, but not every AI server is running on the absolute latest accelerator chips. CoreWeave's co-founder and chief development officer (CDO) Brannin McBee recently testified as much on tech and business talk show TBPN (via Ray Wang).

The CDO explained that today we can see "late 2010's SKUs that are available in market on clouds. They're not running that infrastructure unprofitably."

"Our pricing on Ampere went up during 2025, and our pricing on Hopper stayed within 10% or so of where we started the year.

Why is that happening? I think it's largely driven by inference, and ultimately this concept that not every workload needs the latest, greatest piece of compute... There are different types of workloads that need different types of infrastructure, and the market is simply efficiently pairing the right compute with the workload that they need.

So, we see incredibly robust demand all the way back to the Ampere generation... We're really not seeing any deterioration of these older SKU demand profiles, while also seeing accelerating demand for latest-generation infrastructure as well."

'Ampere' refers to Nvidia's circa 2020 microarchitecture which we gamers know as the the one that underpins the RTX 30-series. McBee is referring to the server chips, however. Back at launch, the AI boom hadn't really started, but a couple of years later things would take off with the launch of ChatGPT 3.5. Before long, AI servers were training and inferencing AI using Ampere chips like the A100.

Following this we of course had the RTX 40-series with 'Hopper' architecture, and server chips like the H100. Then RTX 50-series 'Grace Blackwell' GPUs with server chips like the GB100. And later this year we're expecting 'Rubin' server chips.

In other words, things have moved on quite a bit from Ampere. It makes sense that these older chips are still being used, though, because the amount of money pouring into the AI industry doesn't seem to be letting up—something I'm sure we as PC gamers are all aware of thanks to the resultant high memory prices.

Given there's such high demand and chips are flying off the proverbial shelves and being wrapped up tight in ginormous contracts, I expect it's a game of 'take what you can, when you can.' China, for instance, is set to receive a capped amount of Hopper chips, with no Blackwell allowed.

That doesn't mean Hopper is just as good as Blackwell, rather that anything is better than nothing. And if prices rise, that might just be because demand is rising in a still very supply-constrained market.

I'll be honest with you right up front: I loved the Firaxis XCOM games, but I never played the '90s originals. I know, I know. For years, people told me how good they were, and how the newer ones lost some of the magic. And while I may have bounced off of the first Xenonauts, Xenonauts 2 has finally revealed the truth to me. You guys were right all along. This rules.

There's something almost indescribably awesome about that moment in an alien invasion scenario where your scrappy little fighter shoots down a UFO for the first time and everyone cheers. It's like when the cavalry arrives at first light on the third day. We're up against a vastly superior foe, but with this small victory, it feels like maybe we can win this. Maybe there's hope.

Don't pop the champagne just yet, though. This is when Xenonauts 2 likes to take you down a peg, like the time it completely instagibbed my star sniper from halfway across the map even though she was in cover and wearing the most advanced armor my research teams have developed.

The quickload button has never looked so tempting. But what the underlying design understands so well is that the victories never taste as sweet if they're not built on the graves of fallen heroes. Hope can only come shining through when the clouds are dark.

The bones of Xenonauts 2 don't differ much from any other XCOM or X-hyphen-COM or X-COM-like you might have played. You're in charge of a secretive organization of scientists, engineers, and soldiers trying to stop an invasion by technologically advanced aliens—building and defending secret bases, dissecting captured foes, learning to use their technology, and stopping the six major regions of Earth from devolving into panic.

Doomsday

I completely failed to do that about 180 days into my first campaign and had to start over. This is a very difficult game, both tactically and strategically. I did not take to most of its systems intuitively at first. But that forced me to take a step back and think. It forced me to learn and adapt. And I really ended up enjoying that process after getting over the initial frustration. Contemplating what I could have done differently over a pile of dead soldiers is how I eventually won.

Xenonauts 2 is chock full of skill expression. Mastering the 'time units' you use to move, fire, and even rotate the vision cones of your soldiers, and learning how to deal with each of its deadly alien variants—dozens of them—can give you an edge on a tactical level. But that's not enough on its own.

Securing supporters and eliminating infiltrators on the strategic map is a whole minigame that requires you to prioritize reducing panic, gaining more funding, or controlling an entire region to get a unique bonus, like increased passive experience gain. The prices of different supporters change over time, so you can't really rely on a set build order.

Despite this, I was able to stay a step ahead of the aliens almost the entire time by reacting to juicy opportunities intelligently. But this interstellar menace doesn't go down without a brutal fight, and the lessons that contributed towards my victory were hard-won.

The one area where I felt really shot down by the difficulty curve, literally, was knowing which techs are must-haves by certain points in the escalation of the invasion. If you haven't got tier 2 or tier 3 fighter jets by the time the nastier UFOs start showing up, that's basically a game over that might take months to painstakingly play out.

It's not especially enjoyable to have my entire air force grounded, desperately trying to catch up as I watch the xenos have the run of the place.

Scramble fighters

I did love the real-time air combat when the fight was winnable, though. Especially early on, using the throttle controls and roll maneuvers at exactly the right moments allowed me to win some really lopsided dogfights, sometimes without taking a scratch.

Different UFOs require different approaches depending on their capabilities, which creates some really compelling fights when they get mixed and matched. It's one more little area where skill can come into play and add up to a big advantage.

The tactical battles are the star of the show, though. I started with eight and went all the way up to 12 squadmates, taking them to a variety of urban and rural environments reflecting everything from a North African market to a polar storage facility. The depth of the combat is remarkable, too, forcing me to carefully consider the smallest moves.

Room clearing feels tense and authentic, costing time units to kick the door down and look both ways. The safest way to play is to stick together, moving from cover to cover, crouching down every turn and leaving enough time units for reaction shots. But exciting timed missions shook me out of this routine, where I had to take some big risks that had me standing up and pumping my fists when they paid off.

The story is less remarkable. There are a couple twists regarding the aliens, but nothing too crazy if you've read a lot of sci-fi. But it gets the job done, maintaining tension, setting a good narrative pace, and mixing in moments of triumph with eerie periods where you're left guessing about what the aliens will do next.

It's in the moments in between where Xenonauts 2 shows off. The emergent stories, which have burned themselves into my brain—like the corporal I sent onto the UFO bridge to scout, mainly as a sacrifice, who dodged six plasma shots somehow and ended up sticking with me as a colonel all the way to the end. A random number generator that can one-shot a badass can also spare a rookie, and that's where the magic happens.

Squad customization is another rewarding journey, where I teched up from realistic ARs to lasers, gauss rifles, and eventually plasma. I loved seeing my squadmates progress visually from kevlar vests to basically space marines by the end, with powered armor and auto-med injectors that, nevertheless, still didn't keep them from getting annihilated by an unlucky roll now and then. The tension never goes away entirely.

This is coupled with saveable loadouts and certain items like grenades being unlimited once you research them, which maintains some resource management without it getting too tedious or fiddly. It can, however, start to get cumbersome toward the end to kit everyone out. The same goes for the more complex combat turns.

10 soldiers in the midgame feels like a really good number. Once I got the dropship upgrade to bring 12, that felt like too many. And base defense missions, which can have even more than that, were the only ones I really dreaded. They're a micromanagement nightmare with the number of friends and foes present, and they can't be auto-resolved.

Good hunting

Most of the enemies should be pretty familiar if you've played any version of X-COM, but many of their special abilities can be countered in clever ways, waiting for you to discover.

There are these floating egg tank creatures that are impervious to damage in a frontal arc, and they can easily one-shot multiple fully kitted-out soldiers in a single turn. Naturally, they were initially the last enemy I'd want to face in a mission. But there's a catch: they always turn to face the last thing they heard. When I realized I could solo them by throwing a grenade behind them to make them turn around, then unload into their backside with a machine gun, I actually yelled in triumph.

There's no dedicated overwatch ability, but units will take reaction shots if they end their turn with enough time units left. Unfortunately, they're not particularly smart about this and will autofire straight through a teammate to try to hit an alien that pops out.

This is frustrating to begin with, but in the long term, it increases the opportunity for skill expression in a way where setting up a door breach becomes almost like composing a song. Positioning everyone just right, keeping track of how many reaction shots they could take, and setting up contingencies for worst-case scenarios (they happen more often than you might think!) is incredibly satisfying when everything goes according to plan.

Maybe I fell so hard for Xenonauts 2 because I had never sipped X-COM Original Flavor before. But if so, that only speaks to how this uncompromising, dangerous, and detailed sort of tactics puzzle remains today, even without the nostalgia. The way it builds and releases tension across missions and campaign chapters is magnificent. And when the stakes are so high and the price of failure so steep, victory feels earned and tastes so much sweeter.

Working out how to pickpocket in Crimson Desert might have you a bit stumped. After all, the system in this game is very different to any I've seen before—you can't simply sneak up behind someone and have a rummage about in their rear compartment. Nope, there's a trick you need to pickpocket beyond simply crouching and trying to look sneaky.

You'll also need a mask to commit any petty crimes, so I'll explain how to get one of those down below, too.

How to pickpocket in Crimson Desert

Just like stealing in Crimson Desert, you'll first need a mask in order to pickpocket. You can buy these from Back Alley Shops such as the one in Hernand, though you can also earn one through your first bounty mission to capture Jeffrey in Chapter 2, or sometimes loot them from enemies and bandits.

Once you have a mask, you simply have to bump into someone to pickpocket them. When you do, you'll spot a little prompt that appears, either E on mouse and keyboard or X on controller, which lets you quickly pilfer an item from them. This is random, so you'll often end up with somewhat useless stuff, but you'll get better loot if you pickpocket nobles. You can even pickpocket gold bars from them, so it can be lucrative, RNG willing. You can find a load of nobles around the bank in Hernand, or at the various manors in the city limits, so be sure to try your luck.

The only problem with pickpocketing is that you'll be detected doing unless you have a special item (explained in the paragraph below). Guards won't arrest you, per se, but you will have to run outside of the red witness circle or wait for the timer on your crime to run out. You'll also slowly accrue a fine and lose Contribution, though only the same five points you lose for stealing items.

That is, unless you acquire the Great Thief's Gloves by completing the rather elaborate Secret of the Lost Seal faction quest for House Serkis. These gloves have a special buff called Great Thief which prevents you being detected when you steal stuff, though they do have a 30 minute cooldown between uses. If you don't have this quest unlocked yet, you need to progress the other house questlines in Hernand, including House Roberts and House Alfonso.

The Super Mario Galaxy movie released to some shaky reviews; most critics have slammed the sequel for "having no centre", being only "a series of large, vaguely connected explosions", and simply "a way of gouging cash out of families."

But all the awful reviews of the sequel aside, CEO of Illumination Chris Meledandri is still gushing about the movie's somewhat invisible successes, and how great it was working with Nintendo and director Shigeru Miyamoto during an interview with Cherry the Geek TV: "The secret is being able to bring Miyamoto and the team into the center of the process." (via GamesRadar).

"And I'm not just complimenting Miyamoto because he's standing here," Meledandri adds. "Part of me doesn't want to give away this secret because it's like having the secret to Coca-Cola and then giving it away. The secret actually is the inclusion of Miyamoto and the Nintendo artists in the center of the filmmaking process."

I'm not sure working with the original creators on an adaptation means it'll be better is much of a secret. In fact, I'd go as far as to say that any adaptation, whether it be film or TV, should always include some sort of expertise from the source material. But then again the Super Mario movies do tread new ground for Illumination.

"It's not something that we had ever done at Illumination," Meledandri says. "You know, we had made 10 movies at Illumination. We had made all of them completely inside the studio. So, this was the first time that we were going to work in a new way." Past movies include Despicable Me, The Secret Life of Pets, and Sing, all of which didn't have original source material to look back on.

The Super Mario Galaxy movie, like its predecessor, may not have done well initially, but that doesn't mean it won't break any records in the following weekends. The first Mario movie had an equally shaky start but then went on to earn over $1 billion worldwide, so it certainly wouldn't be surprising to see its sequel reach those heights, despite all the bad reviews.

In a new capital markets briefing from Nexon, Patrick Söderlund, the newly appointed Nexon executive chairman and CEO of Arc Raiders and The Finals developer Embark Studios, spoke extensively about everyone's favourite subject in games right now: AI. More specifically, he talked about how Embark's approach to game development and AI can be deployed across Nexon's teams in some form.

Under a slide titled "Redesigning game development," Söderlund explains that "every company has a plan" when it comes to AI, but "most will get it wrong."

"AI may be a race, but the winners won’t be the first movers—the winners will be the ones who understood the challenge," by which he means fully grasping how and why certain AI tools are used. Nexon, Söderlund remarks, understands this challenge from decades in the industry. All of this history is "context—tens of billions of data points," he says, and "AI makes it usable at speed, and at scale."

But, as Söderlund points out, Embark Studios didn't have that history back when it formed to make The Finals and Arc Raiders. The studio paid close attention to one question: "What needs to be done by hand versus what a machine can do more efficiently?"

"Yes, some of that involves AI," Söderlund explains. "But it’s really about encouraging people to use smarter processes, better tools, and to let go of habits that no longer serve them."

"The initial outcome of that process is two games: The Finals and then Arc Raiders. Two games, built with significantly fewer people, at a fraction of the cost you’d expect for a AAA game."

Both games took advantage of AI tools during development, such as controversial, cost-saving AI voices, which are now being replaced (at least partially) in Arc Raiders. Söderlund commented that "a real professional actor is better than AI," which isn't something I ever doubted, but it does raise questions about the value of AI-generated assets and in-game content—and, if Embark recognises that real voice actors are better, why it used AI in its games in the first place.

Nexon president and CEO Junghun Lee, who spoke about the company's Mono Lake AI initiative, called Arc Raiders a "Trojan Horse", a proof of concept that AI-powered game development can lead to huge successes. It's "a gift that contains a shift in the mindset about how technology frees developers and live service teams to spend more time thinking and less time typing," he explains.

Söderlund's comments and overall approach to AI and efficient game development are no doubt a big part of the reason for his promotion at Nexon in the face of rising costs. "Every leader, team and individual in this company has to start asking how they can get more done with better tools, smarter workflows, and less time on things that don’t move the needle," he said. "Nexon developers are meeting Embark colleagues to understand our process. They are taking a step back and look at how they work – not just what they’re building, but how."

It's wild to look at Crimson Desert—a game I reviewed not even three weeks ago—and see it as an entirely different beast to the one that was released in March. Even as I was in the midst of my 75-hour review playthrough, Pearl Abyss was continuing to nip and tuck away at the thing. I don't think I've ever seen a singleplayer game shift and change in such a short period of time.

That's not to say that's a bad thing, mind. Much of the unnecessary friction I had issues with in my review have been fixed. Fast travel points are more convenient and numerous. There's actual private storage now, so I don't have to carry around every unique piece of armour and weaponry in my inventory. You no longer have to worry about accidentally automatically breaking into locked doors with your stash of keys. Do I wish Pearl Abyss was a little less… flimsy in its whole design philosophy? Sure. But hey, the game's a lot less annoying now.

One thing that had been mildly bothering me since the beginning, though, was flying. A giant black, hooded cape that Kliff can don in mid-air to descend through the skies. It should have been the perfect vessel to explore Pywel's sprawling map.

Instead, it was a bit naff.

It consumed far too much stamina, for one. That led to me playing a game of cat-and-mouse with both my stamina and Kliff's health. Fly a tiny bit, start dropping to my death to regenerate my endurance. Unsheathe the wings once more, float around for a second or two. Rinse and repeat until I touch the ground, my final destination not all that far from where I first started.

Nothing about it felt satisfying in the way donning a strange charcoal-coloured cloak and flying around mid-air absolutely should. An uncharacteristically unenchanting experience even further weighed down by how clunky the entire thing felt. Leaping into the air to initiate flight seemed to bring about an incredible resistive force, draining a quarter of my stamina bar while Kliff dabbled in some airborne dilly-dallying, powerless to whatever magical wind drag was holding him back.

Grand flights warped and contorted into clunky gap-closers and an easy way to chuck myself off the side of a mountain without taking any fall damage.

I'm doing all of this moaning in hindsight, thank goodness, because upon booting the game up after its most recent patch, something had changed. Kliff zipped into the air at a moment's notice, snappily soaring across the gap I had expected to instead float my way over with all the grace of an overweight sodden cat leaping between two broken fence panels.

It feels incredible. It feels exactly how it should have at launch—quick and responsive while still maintaining a sense of weight. It no longer sucks the absolute life out of my stamina bar either. It's finally a tool for exploration, one that opens the door for so much more freedom, and it's completely reinvigorated my desire to journey around Pywel on foot.

Getting between two points of interest is wonderfully smooth. Sure I might have to cat-and-mouse a little still—or quickly chug a stamina potion—but I'm finally able to traverse grand distances the way I would expect to from a literal flight ability. I have never felt so giddy about being able to go from one patch of land to another.

All of this is ultimately part of Pearl Abyss' ongoing effort to tighten up how Crimson Desert feels, and it's working wonderfully. Even things like a massively reduced loading cutscene that made me dread any time I had to zip between fast travel points have encouraged me to explore and venture out in ways I was increasingly drawing back from as I racked up the hours in this game.

And it shows that Pearl Abyss was capable of making a game this snappy and responsive all along. When it comes to friction born of clunkiness—feeling like playing a videogame stuck in quicksand or knee-deep in a quagmire—I am more than happy for the developer to discard whatever ethos it was designing this thing around. I'll be taking to the skies far more now, and Crimson Desert is going to be a much more enjoyable game for it.

World of Warcraft's generally made a few great choices when it comes to its raid and rewards structure in the past few years—one of these solid ideas has been the inclusion of story mode, a soloable version of its endgame raids that lets casual players bop their big bads on the head so they can see the cutscenes everyone at Blizzard worked very hard on.

Well, sort of. The issue is, story mode—much like the casual Looking For Raid (LFR) version—keeps getting time-gated. The patch drops, the raid comes out, and both LFR and story mode versions are locked behind a seven-day wait, with LFR in particular releasing its wings piecemeal.

I can somewhat understand the LFR thing—releasing a new raid is a big deal for Blizzard, and fine-tuning a version where difficulty matters slightly less (but still matters!) is something I can see a savvy developer putting off for a while under the guise of making organised raiders feel a little more special having seen all of it.

But story mode's current implementation feels completely counter to what Blizzard's trying to do with The Worldsoul Saga—that is, having a more accessible and easily-appreciated story. I'm mixed on my overall story feelings of Midnight, but I can at least tell that Blizzard's trying to spin an interesting yarn.

I've even enjoyed some character arcs and plot twists, even if I'm still quietly wondering where the heck the Horde is in all of this—the end of the Voidspire raid, in particular, is a really enjoyable knife-twist. I also like that Blizzard is splitting up its raids, into multiple, smaller ones—allowing for a broader, more episodic story over the course of a patch.

And yet, as it stands, experiencing that story for the first time as a casual player involves playing a bunch of neat lead-up quests to the raid itself, building momentum and then—just as the climax reaches its peak—you get hit with a seven-day wait. Why?

You can make the argument that it, as mentioned, is to make players feel rewarded for organising with their buddies and clearing the raids themselves. But that feels like a load of old tripe for me—because I did it.

When the Voidspire raid came out, I brushed off my old raid instincts (I've cleared a Savage tier of FF14 before, so I'm at least baseline confident) and decided to give Normal a try for myself.

And I mean, it was fine. I don't want to dunk on WoW players, because it's not their fault, but compared to a game that essentially makes you play a 500-hour tutorial over several expansions, the overall mechanical literacy of your average WoW player is… let's just call it "not as thoroughly prepared".

I'm mostly talking about the dragons and that goddamn fear cone. Blizzard basically holds up a giant, blaring sign in front of you—early warning that you're going to get targeted with the breath—and still, the groups I played with couldn't manage it. It's alright, though, my dumb arse died on the final boss and I was thoroughly carried, so the cosmic balance is preserved.

My point is, while I broadly enjoyed the raid, I didn't enjoy the sense of pressure to get it done before I was spoiled by the cinematics being uploaded to YouTube—and I certainly didn't feel like my story experience was particularly improved by doing it with strangers whose inter-party politics rival Game of Thrones in their bloody pettiness.

We're long past the point where seeing a raid actually carries any prestige, and as such I just can't fathom why story mode is being held behind the gate—you can make the same LFR argument that Blizzard's just spreading out developer time to balance it, but story mode's explicitly the "mechanics don't matter, have fun" version.

Nothing's ever easy in game development, sure, but I can make the safe assumption that the balance team doesn't have to work overtime on this one. Until then, Blizzard will continue mandating that its story and solo players—the ones it's ostensibly trying to appeal to with the mode in the first place, as well as Delves and Prey—have to wait a full week while playing spoiler dodgeball before experiencing the story as-intended. For what reason? Who's to say.

Yesterday, April 1, Artemis II launched to start its mission to orbit the moon and lay the groundwork for future lunar missions. Important stuff. I'm sure NASA has got all its systems in order, no kinks to iron out, no oversi- oh no, wait, never mind, I see the Artemis II is hooked up with Outlook. And it looks like the Microsoft email and calendar suite is already causing problems for the astronauts:

"I also see that I have two Microsoft Outlooks and neither one of those are working. If you want to remote in and check the Optimus and those two Outlooks, that would be awesome."

Those are the words that recently sounded from Artemis II as one of the astronauts conveyed the issue back to NASA. Though it's worth noting we can't be certain the problem lies with Outlook rather than one of what I'm sure are many other systems at play in a spacecraft. NASA says they'll remotely connect to the computer to see what the problem is.

I can't imagine that much crucial will rest on whether Outlook is working or not, but it does at least go to show even the best and brightest among us can be victim of the simplest IT troubles. Largely because we're at the mercy of our tools.

Those tools will hopefully be improving soon, though. Microsoft has said that the company will be making '100% native' Windows programs, which is a change from the usually more laggy WebView apps, which are essentially browser apps wrapped up to run on Windows. Given the latest Outlook app is a WebView one, hopefully that will get a native re-build, too.

The improvements shouldn't be limited to apps, though, as Microsoft has recently said it's going to work on improving general Windows 11 performance and rolling back AI features. Part of that includes moving non-app elements such as the start menu to the WinUI3 native UI framework rather than WebView and React.

Hopefully no more software issues surface on the Artemis II mission, and hopefully Microsoft has all its improvements ready to ship before the Artemis III launch next year.

Climbing games are having something of a moment: Fare like White Knuckle, Peak, and Cairn have all made big splashes (so to speak) and I maintain that Jusant is one of the most sadly overlooked games to come along in recent years. And now we've got Ascenders: Beyond the Peak joining that mix, with its own take on the cliffside life—but it's a whole lot different than the others.

Ascenders might be most aptly described as, what if Darkest Dungeon was a climbing sim? Thematically, that is. You're not climbing solo, for one thing, but as the leader of a team—and as the leader, that means you'll occasionally have to make some tough calls. Like, for instance, who's going to die.

As you might imagine, that's a fraught call to make, and not just because of the obvious moral considerations. Cutting the rope lightens the load so the rest of the squad is less likely to get hauled down into the abyss, but that's not just because you've got one less climber dangling from a rope: Any supplies or resources they were carrying gets the same express trip to the bottom, which is going to cause headaches later on. There's also the psychological toll on those who remain, which will almost certainly make the continuing descent more difficult.

Of course, you can always opt to withdraw from whatever precarious perch you find yourself clinging to and maybe save a life. But your carefully planned route upward may be lost as a result, potentially risking everyone in the party, rather than just that one guy who doesn't seem to be particularly good at the whole mountain climbing thing anyway.

This would be me. Every time, all the time. It's why I only climb mountains in videogames.

Unlike the other climbing games I mentioned, which are primarily games of dexterity and timing, Ascenders is turn-based and strategy-first. The game features numerous classes of climbers, each of which brings unique tools and abilities to your party that can work together in sometimes very powerful ways. Artifacts collected during successful expeditions can be used during future climbs or to upgrade your base camp, or can be traded instead to other factions. Which factions you work with over the course of the game, which mountains you climb, and when you choose to press forward or retreat will also impact your reputation, the world state, and possible game endings.

And why, exactly, are you doing all this? Dying on tough ascents isn't all that terribly uncommon: Mount Everest is famously covered in frozen bodies of fallen climbers, some of whom end up serving as landmarks for future expeditions. But the motivation in Ascenders goes beyond the classic, "Because it's there." It comes back to those artifacts you're collecting: As the trailer hints, these mountains are "cursed with mysterious Lovecraftian threats hidden within the rock."

"Every expedition is a desperate attempt to reach the summit, understand the forces shaping the mountain, and return alive with the knowledge and artifacts you need for your next ascent," developer Ludogram said. "At all costs, your mission is to forbid them from falling into the wrong hands. Death after death, players will learn to turn the environment into a weapon, mastering each ascent through hard-earned roguelite progression."

That's a long way from the chill, magical mountaineering of Jusant, but even so I have to admit I'm intrigued—and no, not just because I want to find out if I can cut the rope on the rest of my team even if I don't have to. (Not entirely, anyway.) Ascenders is currently expected to launch into early access in Q3 2026, and will be available for PC on Steam.

I am not a prepper per se. But I certainly have prepper sympathies. Maybe the sort where, if I thought I could get it past my partner, I could see myself spending many thousands digging out a bunker in the back garden and filling it with canned goods and a year's supply of drinking water, fuel, and toilet roll. And a toilet. Of course.

But I am also a damned computer nerd, so when I saw Project NOMAD pop up I knew I had to get in on it. Project NOMAD (Node for Offline Media, Archives, and Data) is an entirely offline repository of knowledge, giving you access to a vast library of otherwise net-based goodness without any network availability necessary.

Y'know, just in case the grid goes down for reasons.

These sorts of collections have been doing the rounds on different reddits and forums for a while, but the thing which really sings to me and my nerdy sensibilities is the fact that it's an easy one-shot install that lives locally on your system, and has a straightforward front end you can access via a localhost URL on any browser, whether you're connected to the net or not.

It also sings to me because it's free, where other prepper-facing repos can cost hundreds of dollars for the privilege. I also like the fact it is unashamedly high-end. It's not designed to run on a low-end machine, it's built for precisely the sort of system I'm running it on; a high-end gaming laptop.

And I am also into it because it includes AI. Now, no eye-rolling at the back. Hey, those cries of 'of course it does' are not helpful. But the AI actually is.

The thing with Project NOMAD is that it's been designed to give the user the chance to download a knowledge repository to their own devices, and that can range from a few texts, to a swathe of Project Gutenberg books, medical guides, and the entirety of Wikipedia. Parsing that much knowledge is not easy, especially if you're looking for something very specific, but packaging a local LLM (or multiples if you want to download different models) allows you to use the knowledge base it sets up—as well as add other documents to that knowledge base yourself—and then use the LLM to query it.

But it's not just about having an offline internet of data at your disposal when the WWIII comes home to roost, it gives you the chance to learn in a more structured way, too. By including an education platform, powered by Kolibri—a deliberately offline-first learning platform. You'll need to import your channels online first, but once you've downloaded the entire curriculum you can set up your own little post-apocalyptic school.

Now, I know that I'm still going to need some power source to keep the battery on my Blade 14 rolling, but some good solar panels (if you can still see the sun) or a wee generator would do the job. And I've already queried my NOMAD about how to build my own gasifier already so I can easily generate my own fuel for it.

I've gone big with my install and so, even before downloading the entire US curriculum Khan channel, I've got a setup that's around 500 GB in size. Though that is the entirety of Wikipedia and a ton of survivalist and medical text, too. As well as a detailed Google Maps-esque download covering the entire of the UK. Though I did have to go to Open Labs to download that specific pmtiles file myself as Project NOMAD only gives you the option to download US regions during the setup process.

It does have a tendency to jealously hoard resources like some Capital Wastelander, especially if you have an LLM installed, and it will run automatically when you boot up your machine unless you explicitly tell it not to. Right now, I've got a few Chrome tabs open and the Project NOMAD Command Center open, and it's yomming 12 GB of the 32 GB available to my system.

It also has to be installed on Linux, too. Oh, did I not mention that? Yes, the whole project needs to be installed on a Debian-based Distro, ideally Ubuntu. I've got it running on my PopOS install after a little messing around with Docker, but there's no hope for you if you're running Windows.

Well, if you're persisting with trying to run Windows after the apocalypse then there's really no hope for you anyways, amirite?

But now I, a Linux-user and prepper sympathiser, will be able to continue to read episode breakdowns of old seasons of Survivor on Wikipedia in case I forget who won the immunity challenge in one vital episode of the Cook Islands season. I will also still be able to read The Complete Book of Cheese (with such wonderful passages of prose as, "connoisseurs use gingerbread as a mouth-freshener; and I, too, that sunny day among the Edams, kept my gingerbread handy") and watch videos from The Urban Prepper about the vital 72 hour water supply.

And all because I have loaded a laptop up with a healthy dollop of the interweb tubes.

Now, I just need to figure out how to keep Steam running offline for time immemorial...

We will never get a game quite like 2015's Life is Strange ever again. Ready for the mosh pit, shaka brah? Go fuck yourselfie? I was eating those beans, are you fucking insane? I was eating… those BEANS!? Iconic. Equally as cringe a decade ago as they are now, but it just worked. A small, relatively unknown French studio doing its best to hella capture the essence of American high school life.

At the core of it all was the game's two main characters: Max Caulfield and Chloe Price. Specifically, their relationship (as sapphic or platonic as you well wish) and the ravelling and unravelling of a single week together as Max's newly-unlocked rewind power danced along the timeline of their lives. All of this culminates in the choice for Max to either save their hometown at the expense of Chloe's life, or defy fate forever and see Arcadia Bay destroyed.

Life is Strange is a game that, despite its shortcomings, was an incredibly earnest and heartwrenching tale of teenagehood, tragedy, and consequence. A perfect storm.

One which Deck Nine—who has developed all but one Life is Strange game since Don't Nod's initial debut—has spent the last two years trying to recreate. First with Double Exposure, which brought Max back with new friends, romance options, and time-adjacent powers to explore.

It was a tough job narratively, one I personally found serviceable enough as a staunch Bay over Bae chooser. But it was one which upset many of Pricefield's fans, who felt their decision to save Chloe over Arcadia Bay was minimised. Reduced to a relationship that had drifted apart over a decade, nary a text or occasional hookup to be seen. Something that, to me, feels rather realistic, especially as someone who has journeyed through these games the same age as these two women. But this is also a videogame with magical rewind powers, we can loosen up a little.

And now we're here with Life is Strange: Reunion, the first game to feature Max and Chloe side-by-side since that very first outing 11 years ago. A game that is so clearly Deck Nine's desperate attempt to claw back any good grace with the Pricefield enjoyers that it comes at the expense of every narrative beat set up at the end of Double Exposure, and reduces almost every supporting character to a prop in service of setting up and moving along moments between my two OGs.

Reunited and it feels… okay

This is coming from a fellow Pricefield enjoyer, mind. Did I foolishly direct my romantic interests towards Warren in my very first playthrough of Life is Strange all those years ago? Sure, but I've long since seen the light.

I'm also someone who has never wavered in my decision to sacrifice Chloe. Not all stories have a happy ending. It gives the original game so much more emotional weight, and makes Max's reunion with her in this one far more interesting. A person that she's spent a decade grieving, carving a space out in her heart for, suddenly standing in front of her. It's an incredibly poignant moment—and I admittedly welled up a little when Foals' Spanish Sahara started playing, a reminder of the choice both myself and Max made all those years ago.

The scene gave me high hopes for what Reunion was going to deliver. I'd anticipated fan service and manufactured cutesy Pricefield moments, sure. And they're certainly there. But in pursuit of them, Deck Nine forgot about everything else.

Reunion feels like two stories. The one Deck Nine originally wanted to tell, and the one it felt it had to. The foundations laid with Double Exposure's ending have been demolished. Where I expected Safi's quest to find other people with powers like herself and Max, she instead stropped around campus, hiding her real self in favour of shapeshifting as students and faculty. She toes the line of being interesting—criticising Max's reckless power usage, blind to her love for Chloe to see any external damage she's causing. To Safi included.

Instead it's a bunch of passive-aggressive comments (I rejected her plea for support at the end of Double Exposure, my bad I guess!) that lead nowhere, before the end of the game ties everything up in an incredibly sloppy bow and has her back to her old self.

Where Diamond's nosebleed had hinted towards further cool power-related character development, she's actually entirely absent from Reunion. Relegated to a single (optional) phone call at the very end of the game.

It's all shafted in favour of a mysterious fire that sees the demise of Max's nouveau bestie Moses. And, once again, Chloe. It's admittedly another interesting premise, but one that goes nowhere in Reunion. The eventual climax to who is responsible is awful and nonsensical. A culprit given diabolically little screentime and character development, which makes the reveal a total nothing burger.

But hey! Chloe and Max get to go out on the lake in a little boat at sunset and wax lyrical about their time together as teens. And break into stuff just like they used to. And hold hands. And finally share a little reunion smooch.

They're important moments, don't get me wrong. Ones I loved. But I couldn't help but feel like Deck Nine was carefully tip-toeing towards me to present them. Pricefield kisses in one hand, a white flag waving back and forth in the other.

For me, Reunion's most interesting narrative moments don't even involve Max. Chloe and Moses, and their interactions, are the star of the game. For starters, Moses is perhaps the only Double Exposure character who doesn't feel completely gutted—probably because he fits nice and neatly into Reunion's story—but his dynamic with Chloe is where I got to see the biggest indication of her character growth. Where I get the biggest sense of the kind of woman she's become over on the "alive Chloe" timeline.

And then I'm taken back to a poorly woven story. Attempts to intertwine Safi and Chloe's shared situation. Haphazardly slotting together a fire that has nothing to do with either of them, while also exploring and bringing a critical lens to Max's rewinding, and simultaneously figuring out how to bring it all back to the two who started it all.

Unfortunately, I'm not sold on it. I will always love Max and Chloe, but the characters Deck Nine crafted for Double Exposure deserved better. It's not entirely the developer's fault—the studio is far smaller than it was for the previous game, and I would bet there was a whole lot of narrative pivoting that took place before Reunion's release which has led to a sloppy final product. I just wish Pricefield's farewell had been given the time and care it deserved.

Minecraft creator and opinion-haver Notch has taken to X to complain about Nvidia's upscaling and Frame Generation tech, DLSS. You might have heard of it, particularly its controversial new iteration. However, it's the whole concept of DLSS that seems to be confusing Notch.

"DLSS fundamentally makes no sense," the creator of the best-selling video game of all time opines. "Because the graphics card is too slow to run the game at reasonable speeds, you use THE SAME HARDWARE to run a neural network to generate frames in between the existing ones."

Putting aside the fact that Notch appears to be talking about Frame Generation specifically here, which is just one aspect of DLSS, lumping graphics card hardware into one monolith of a category is a touch reductionist, as specific ML-optimised Tensor Cores handle the...

You know what, the X thread replies have already done a lot of the work for me. So, I'll give you a choice selection of those instead.

An overall point could be made, I guess, that graphics card design and development could (or should) be focussed towards raw raster performance, instead of developing machine learning-specific hardware to assist.

However, with the advent of ray tracing (and ever-more-demanding game engine tech), the industry has shifted away from resource-intensive brute force rendering techniques, and has instead prioritised AI developments to keep the graphical ball rolling.

As Nvidia's vice president of Applied Deep Learning Research, Bryan Catanzaro, explained back in 2023:

"Moore's Law is dead. We don't know as a civilisation how to keep turning the crank on traditional ways of doing things. We have to be smarter."

"You fundamentally realise you have to be more intelligent about the graphics rendering process," Catanzaro continued. "Brute force—let's re-render every frame 120 times a second at 2160p output—that is wasteful because we know that there are a lot of correlations in the output of any rendering process.

"We know that there are a lot of opportunities to be smarter, to reuse compute. And then deliver transformational image quality benefits, things like Cyberpunk, that we could never have imagined doing before."

Our Nick is performing some in-depth testing of DLSS and Frame Generation at this very moment, and has passed me the image above from our own data to represent that very point.

"Everything in the blue box is what the GPU has to do to render one frame," Nick explains. "Everything after that is what the GPU has to do interpolate one additional frame with AI."

That's not to say that DLSS is everyone's cup of tea, or that Nvidia's AI enhancements are flawless. Opinions will vary greatly here. Nor is it entirely unreasonable to rail against the idea of machine learning algorithms generating frames within your games as a catch-all solution—particularly as the tech is not without caveats.

Especially when Multi Frame Generation on lower-spec cards gets involved, or developers rely too heavily on it for reasonable performance.

However, it does all make sense, at the very least. Hope this helps. I hate to be that guy, but sometimes this job just puts you there, y'know?

The Beehive Club is one of Crimson Desert's weirder weapons, letting you envelop your enemies in a cloud of bees. Similar to the Soul Spear, you can't upgrade or slot Abyss Cores into it, but players were initially using it to annihilate bosses. It seems to have been nerfed since the game's release, but it's still a fun weapon (if you like bees) and does alright damage.

There's no guaranteed way to get the Beehive Club, since it's a random drop, but I'll run through your best bet below.

How to get the Beehive Club in Crimson Desert

As the name implies, you can get the Beehive Club as a random drop from smashing beehives you see hanging from trees in various locations. While you can destroy any while out exploring using your bow, the best spots I found were by a mushroom patch at Three Saints' Falls, north of Hernand, literally in between the "Three Saints" and the "Falls" text next to it. There are also a couple of hives in the trees around Pollengarth Apiary to the south of your Howling Hill camp, in the Hernand Highlands.

Though you can't enhance it in any way, such as by adding Abyss Gears, what makes the Beehive Club special is that when you smack enemies with it, the club releases a cloud of bees that do DoT damage. While this has seemingly been nerfed because it was murdering bosses, it'll still do decent damage in the early game. The only drawback is that the club only has 30 durability, meaning it'll only work for 30 attacks before it breaks and you have to go back to smashing hives to find a new one.

If you're struggling with bees attacking you as well (they can be unpredictable like that), remember you can steal a Beekeeping Suit from the shack in the Pollengarth Apiary. If you pursue the Hernand faction questlines, there's also a quest at the apiary for House Grace which sees you kill a honey-hungry bear and get a Beekeeping Suit as your reward. This suit gives you immunity to bee attacks.

Before you get any ideas, though, I already tried smashing all the hives in the apiary and they sadly don't drop the Beehive Club—only the wild hanging hives will work.

AMD compute and enterprise AI VP Madhu Rangarajan told tech news site CRN that they are expecting next-gen AMD Venice servers to increase the performance lead over next-gen Intel Xeon Diamond Rapids, in part due to Intel's decision to ditch for the upcoming server CPU generation.

The AMD VP said, "We understand that our competitor has made some interesting choices on multithreading and so on. That’s going to help us gain even more market share in a broader enterprise market where that has implications to your licensing and many other costs associated with it."

Rangarajan is referring to Intel's decision to ditch HyperThreading—i.e. simultaneous multi-threading (SMT)—in the Panther Cove P-cores that will lie at the heart of the upcoming server processors. This isn't to be confused with Panther Lake (PTL), Intel's recent snappy laptop chips, which use Cougar Cove P-cores, though these also lack HyperThreading. Different cores, different markets, same absence of multi-threading.

AMD's upcoming Venice server architecture is expected to include SMT, with up to 256 cores and 512 threads per chip.

Intel fellow Srini Krishna responded that the company sees "positive momentum with our data center customers... with the performance, security, and efficiency of [the] Intel Xeon 6 family... Ultimately, winning in the data center is about the aggregate value we deliver across security, efficiency, consistency, quality, and overall platform experience."

Last year, Intel's lead x86 CPU architect, Stephen Robinson, explained the decision to ditch HyperThreading with PTL:

"We didn't lose a lot in client because of hybrid and the core count, but we gained a bit in our design and execution... Now you've kind of got something that's maybe a little bit easier and less expensive and maybe can go a little bit faster... If we can get more workloads running on an E-Core then we can bring more efficiency to the whole platform."

The upcoming Xeon server chips will also lack E-cores, however. This is something we've recently heard might be the case for Intel's upcoming tiny mobile chips, the Core 300-series (sans 'Ultra'), though they will use low-power versions of Intel's Darkmont E-cores.

After Diamond Rapids, we can expect SMT to probably return in server chips, however. This is because Intel CEO Lip-Bu Tan explained last year that this is the desire:

"In data center, we are focused on regaining share as we ramp Granite Rapids while also improving our capabilities for hyperscale workloads. To support this, we are reintroducing simultaneous multi-threading (SMT). Moving away from SMT put us at a competitive disadvantage. Bringing it back will help us close performance gaps."

Diamond Rapids is expected to launch later this year and be built on Intel's 18A node. What we've seen of 18A so far, with PTL chips, is pretty impressive, so there are no concerns on the node front. Notably, though, Intel's next-gen Nova Lake desktop chips are expected to be made on a mix of Intel 18A and TSMC N2 processes, not full 18A.

In the PC gaming market, AMD chips hold a pretty significant lead against Intel ones, despite new impressive additions like the Core Ultra 7 270K Plus and Core Ultra 5 250K Plus. There's a reason five of the six CPUs we recommend for gaming right now across all budgets are from Team Red: Intel just doesn't have the same value offering when it comes to pure gaming.

However, when it comes to market share, things change somewhat. Despite AMD climbing significantly, Intel still holds about 70% of the desktop CPU market. And the picture is similar in the server market, with AMD making some serious gains but Intel still retaining the lion's share.

Competitive gaming mice aren't for everyone. In fact, I'd go as far as to say they aren't for most people. Part of that is their often prohibitive price; most won't even get anything out of the milliseconds they can shave off reaction time with higher polling rates, greater DPI, and more accurate actuation. When you're landing firmly in the middle of the end-of-match scoreboard, it's likely not just your aim that's costing you kills—your competition probably have mice that are just okay too.

Still, I'm glad that the competitive market exists because it does a lot to raise up the bottom rungs. Take a look at the classic Logitech G305 that we reckon is the best budget wireless gaming mouse right now. For $26 more at MSRP, it has the same DPI as the Mchose G3 V2 but a lower speed and acceleration. It's almost double the weight, to boot.

Logitech is a good comparison for what I think Mchose is going for. The G3 V2 is certainly a good bit lighter and less thick than the Logitech G Pro 2 Lightspeed, but the aesthetics are similar, with a single logo in the base of the mouse, alongside a white and black mouse wheel. They're also both fairly understated, with just a single light in the mouse to show when it's turned on.

You're also getting the responsive PAW3311 sensor, Omron mechanical switches, and three connectivity modes in that light frame. The value-oriented end of the wireless gaming mouse market gets you a mouse that would be a lot more expensive just a few short years ago, and the G3 V2 doesn't feel like it's making tons of compromises for that price either.

Naturally, this would mean very little without a comfortable feel and solid build quality, yet I'm happy to report that the blue G3 V2 I've had my grubby mitts on performs well on both counts. Mostly, the absence of negatives at this price point is a big positive, and the only thing I've had to get used to is the potential for misclicks thanks to just how ickle this rodent is. I have fairly large hands and definitely felt them drape over the front of the mouse, weighing it down sometimes.

The G305 is similar to the GSV2 in more than just specs, though. The G3 V2 has that classic low-down egg shape and only a handful of extra buttons. It's not particularly flashy in any way, though the light blue model I'm using looks pretty great. You've got two clicky side buttons, your standard left and right clicks, a scroll wheel, and a DPI switch at the bottom. Its Omron mechanical switches feel great to click, with a satisfying ping in response. The same is true of the side buttons, which are easy to press down and result in a satisfying click.

The plastic shell is firm, and even a heavy squeeze on either side gets nearly no flex in response. I threw it into my backpack when heading up to the PC Gamer office and never feared it would break. It even has a small compartment in the base of the model for its 2.4 GHz connector. My only complaint here is that the included USB Type-C cable and connector aren't the same colour as the mouse, though that's not the most reasonable demand to make of a $34 mouse.

While scrolling and browsing generally, its 50 g weight makes it a solid everyday mouse, and the 110 or so hours of 2.4 GHz play per charge is adequate. The texture on top is very smooth and, coupled with its weight, this squeaker glides easily across my desk.

Moving over to gaming, Arc Raiders feels fluid and fast, and perhaps even better than it does with controller's godlike aim assist. Admittedly, the very light and small frame does offer less comfort in my hands during the most hectic firefights than the likes of the chunky 8BitDo Retro R8 I'm used to, though.

In horror shooter Crisol: Theater of Idols, I never found it to lag behind my movements or add any latency. Over long gaming sessions, the connection to the 2.4 GHz connector stayed consistent throughout.

In competitive shooter Counter-Strike 2, I found it snapped to heads fairly well, though it naturally lags behind more expensive, competitive mice. One thing using the G3 V2 cemented in my mind is that the diminishing returns for mouse performance sets in somewhere between $30 to $50.

Take the 8BitDo Retro R8 I use every day as an example. At $50 dollars, you can get up to 8,000 Hz polling while plugged in, and double the DPI. This means for almost 70% more, you are getting a DPI 210% more than the G3 V2 and a polling rate that is 800% better. But is that mouse 2.1x / 8x better overall? My answer is no. They both serve different purposes, and most gamers couldn't justify the battery drain of a 8,000 Hz polling rate anyway.

When compared to the Logitech G503, this G3 V2 is a natural upgrade and well worth considering, but it's not quite head and shoulders above its competition now. Instead, it sits alongside them as an equally good choice.

I've also tested out the G3 V2 Pro, which comes in at a couple of dollars more, in exchange for a PAW 3395 sensor and a DPI of 26,000. They are the same weight and feel identical. If I were picking up a mouse, I'd likely go for the Pro, partly because the difference is so minor in price. The Pro offers the same polling rate, though it does come with a few extra features in the Mchose software.

One thing I do like about the G3 V2 is that you can download drivers online or use its software in-browser. It rarely takes more than a couple of seconds to connect, and from there, I can get any updates I need, plus adjust its six DPI profiles, and change button mapping.

I can also tinker with settings like polling rate, scroll wheel direction, hibernation settings, and key debounce, to adjust how fast I can click. It's easy to use and far less cumbersome than the likes of the Logitech G Hub.

'Un-cumbersome' is a good description of the mouse as a whole. It's budget-friendly, easy to use, uncomplicated, decently specced, and light enough to fling around a table.

If it had launched a few years ago, it would be a clear winner for the best mouse under $60, but now, it's one of many solid mice. And that's certainly not a bad thing.

Cambridge Audio might not be the first brand that comes to mind for desktop speakers, not least when the space has been dominated in the past by the likes of KEF, Ruark and FiiO in recent years.

However, the brand's L/R S were shown off to me at an event in December at its London HQ, and I've been itching to get my hands on them since. After months of waiting, I've finally been able to get my hands on them to judge how they sound and how easy they are to live with as a set of $549/£399 desktop speakers.

I'm a particularly big fan of the way that the L/R S looks, with a stylish and minimalistic finish to it. My sample came in a modern white colourway, although it's possible to get them in black, dark green, dark blue, bright orange, or, for an extra fee, a walnut veneer. My pick based on the models I've seen would be the white or orange, I must admit.

These are on the larger side for desktop speakers, I'd wager, being taller and deeper than competing options from Ruark and FiiO, although they aren't so big that I found them unwieldy on my desk or sideboard. I like the 'floating stand' these speakers come on too, as the chassis tapers off towards the base to give the illusion they're floating—neat.

The front of the L/R S confidently boasts the two drivers, with no option for grilles, be it magnetic or from the factory. Personally, I've always been a big fan of the exposed driver look for added style and intrigue, although I can see why you might prefer a grille.

As is typical with a lot of other active speakers, Cambridge has opted for the 'slave and master' combo, with one speaker handling all of the ports and smarts, with the other connected to it by a pleasant, locking braided cable.

On the one with all the ports and such, you get a pleasant array of inputs with standard RCA line-in, an optical out and a USB-C. The optical out and USB-C inputs support resolutions of up to 24-bit/96 KHz PCM, which is why my SACD rip of Genesis' Abacab wouldn't play… There's even room for a subwoofer via a dedicated port for when you need more thump. You also get two toggle switches so you can 'tell' the speaker where it is, be it near or further away from a wall, or whether it's on a desk or not.

Wireless connectivity is limited to aptX HD Bluetooth, which is switchable with the dedicated 'source' button on the rear, and pairing to my Honor Magic V3 worked decently well. You don't get any streaming smarts, unlike some slightly dearer active speakers, if you wanted some Wifi powers for Tidal, Spotify or otherwise—for that, you'll need to look further up the range.

That's not to say the L/R S isn't happy to take inputs from a streamer for added flexibility. If you chuck a WiiM Mini or a FiiO SR11 with these, you've got a very competent desktop hifi system with streaming smarts and minimal fuss for under £500/$750. For my testing, I not only used these speakers with my main Windows gaming PC and a MacBook over optical and USB-C, respectively, but also with my WiiM Ultra desktop streamer for a more 'typical' use case.

As these cans are bereft of a companion app or similar means of phone-based control, Cambridge provides a small Bluetooth remote to use with these speakers. It provides an easier way of choosing inputs than dealing with the guesswork of using the input switch on the speakers' rear and the associated LED under the right-hand unit.

There are also some basic EQ presets for voice and movie use, with the former upping the prominence of the mid-range and the movie mode bringing a little more bass, not that I found the L/R S needed much more bass.

And now the important bit—how these speakers sound. Well, if the headline is the fact that I was very sad to have to box these up and send them back to Cambridge, that should tell you enough.

For a set of desktop speakers, I was pleasantly surprised by the strength and power of the L/R S's sound signature. These certainly sound bigger than the more modest size of their cabinets, and had no trouble at more reasonable volume levels, filling either my dining room on the sideboard or in my office on my desk. There's 100 W of Class D amplification to play, with each driver getting 25 W—we've got a 21 mm hard-dome tweeter behind a waveguide and phase cap above a 76 mm mid/bass driver in each speaker.

A lot of that is due to the power of these speakers' low end, as they've got some tight bass with lovely extension. Granted, the L/R S isn't designed as the last word in reference speakers, but more as a fun and engaging listen, which they certainly are.

For instance, spinning up Rush's Digital Man demonstrated the presence and depth of Geddy Lee's bass and the gritty guitar riff from Alex Lifeson. The same was true in using an SACD rip of Big Time from Peter Gabriel, which perfectly demonstrated the power and punch of these units, although I found the lower frequencies could encroach a little in the song's busier elements.

The prominent low-end helps the L/R S's gaming chops. It helps provide an immersive touch for explosions and heavy gunfire when I use them in Counter-Strike 2, while they help to prove Los Santos can be a real hellscape of PvP battles when I boot up GTA Online.

I was also impressed with the directionality and precision afforded by these Cambridge speakers for use in music and games. Loading up the new 2026 remix of Marillion's Go!, I got a lovely sense of width and expanse for the track's elements, such as everything from Steve Rothery's guitar riff and solo to the quieter acoustic in the background, and the rounded texture of Pete Trewavas's bassline. The L/R S can also extract a lovely amount of detail.

Likewise, a live cut of George Benson's Affirmation provided further evidence of that lovely sense of space and a wide, precise soundstage. This went for elements such as clapping hands from the audience as the song's intro began, plus the percussion that was further back in the mix against the melodic guitar lines.

A really tricky test for any audio kit that I like to use is a track from seventies jazz fusion band Brand X called Dance of the Illegal Aliens. A good set of speakers or headphones will be precise in its organisation and provide enough room for the track's individual elements, be it the persistent drum hits, percussion or the pounding bassline. Listen to it on a less precise system, and it can turn into a muddy mess. On the L/R S, that certainly wasn't the case, with it acting as a real demonstration of these speakers' width and precision.

Likewise, the more immersive note to the L/R's sound quality helps Forza Horizon 5 with precision of where other cars may be when I'm sat in cockpit view, and I want a little more awareness of where the next punt into the wall from another player may be coming from.

I could go on about the L/R S all day, admittedly, but it'd just be more of the same of me waxing lyrical about them. These are a big brute of a desktop speaker that makes for a real entertaining listen with deep bass, crisp mids and a smooth treble that makes for an ideal companion for games and music alike. Connectivity is good, and they've got some real style, too.

Granted, they are bigger than the likes of the Ruark MR1 MK3s and FiiO SP3 BT, and slightly more expensive, but I have no hesitation in recommending these for an entertaining and stylish set of desktop speakers that I have been very sad to have to send back. I might just have to go out and spend some real money on them.