Let’s get one thing out of the way: I hate Bubsy. Not Bubsy 4D, the interesting, vibrant game I’m actually previewing here, but rather Bubsy himself, the mangy, atrocious feline, a poisonous product of a cynical age of mascot marketing. Born in a dark hour when it seemed practically every video game company tied their fortunes to some clone of a clone of a Sonic sidekick, Bubsy has a well-deserved reputation as a D-tier cash-in character better left forgotten.

Fortunately, I have much more positive things to say about what I’ve seen of Bubsy 4D, a brand-new reconstruction of the furball’s tainted legacy that, while far from perfect, distinctively improves on Bubsy’s dubious past in almost every way. During a little over an hour with the Bubsy Steam demo, I found a great deal worth exploring… as well as some less encouraging signs.

It’s true: there are quite a lot of things to like about Bubsy 4D. It’s a colorful game with some cute creative accouterments scattered around the thematic worlds, like the knitted doily highways in the thread-and-needle themed opening stages. Voice work is solid (even if the script is a bit uneven) and my preview build was mostly glitch-free. The movement mechanics are fairly solid, outside of a little camera frustration on very long jumps to some small targets.

The tutorial area is concise and useful, a rare enough combination in games, teaching valuable skills with minimal exposition. Bubsy’s jump mechanics are logical but take a little getting used to – with various kinds of combinable jumps tied to a face button and both triggers – and it’s nice to be able to practice getting around with minimal risk or consequences before moving into the main campaign. Not that dying is that much of an issue… checkpoints (represented by litter boxes) are fairly generous, and even if he dies, Bubsy rarely respawns all that far from where he last fell.

And the obnoxious little cat gets around pretty well. Bubsy’s main strength is his airborne mobility. He can jump, double jump, leap forward and gently descend like a parachute, and pounce ahead to cover great horizontal distances… or do them all at the same time. Bubsy can also inflate himself into a ball to roll along hills and half-pipes, building momentum to execute even longer jumps. He can pounce on certain sheer surfaces and scramble up them, or claw at special points to gain mid-air distance boosts. He’s very capable of getting around, so much so that he can occasionally get ahead of the camera when stringing his pounce onto the end of a combo.

That pounce doubles as Bubsy’s primary attack, although in truth I found very few threats in Bubsy worth attacking. The demo stages are populated with plenty of steep jumps and acrobatic challenges, but there are very, very few enemies wandering about, and those that are present are braindead and quite easily dispatched. Platformers with few enemies aren’t unheard of (Portal, Super Meat Boy, and Thomas Was Alone come to mind) but it was still a little odd how remarkably vacant the colorful platforming environments felt without the usual crowds of Goombas, Koopas, or Giant Enemy Crabs. And while Bubsy’s level layouts are clean and competent, none of them approach the more subtle complexity of the classic platformers listed above.

And that’s my biggest question about where the full version of Bubsy 4D will ultimately land. The demo stages struggle somewhat in the fundamental level design, not in terms of the aesthetic production, but in the layouts themselves. The worlds, while adequately themed, are not especially well laid out. Most are obvious main paths with clearly-visible spokes and isolated islands. Scattered around the spokes are the most valuable collectables, while along the main path are myriad yarn-balls hovering in space. The results are mid-sized, explorable levels which should encourage curiosity and discovery… but often fail to do so.

The problem is not in the overall method, but its glaringly obvious lack of refinement. It is, in a sense, a return to the PlayStation 1-era world design of many mascot platformers, or perhaps more accurately, the kitchen-sink, haphazard-style popularized by many C64 and Amiga platform games, including classics like the Turrican series. Wherever the inspiration comes from, it doesn’t quite work yet here. The goals of each small section of a level feel repetitive, with the layout either too obvious or the goals not challenging enough. You can tell where you should probably go next, which is a good thing, but it’s often so obvious that the sense of discovery really suffers. There are a metric ton of collectables to seek out, but they’re rarely concealed in interesting ways or blocked by compelling challenges. And the lack of enemies greatly reduces the layered threats that might otherwise enhance the nuance and challenge.

Jumping between towers and collecting baubles is as old as video games, but what separates the mediocre from the classic is the refinement that takes place in implementing these rote mechanics into practice. After playing Astrobot and Donkey Kong Bananza within the past year, I can say that Bubsy 4D’s level design feels unfinished by comparison. The play areas, while visually interesting, are unpolished in their dynamic interactions. Leaping between obstacles and avoiding spikes feels clean, fair, and competent, but rarely feels exhilarating. There’s just enough openness to the worlds to obviously require tremendous testing and refinement on the part of the developers when designing obstacles that can be approached from multiple available angles and elevations, yet it’s painfully obvious this testing process has not come near final fruition. It’s not bad at all, it’s just not nearly as good as you know it could be. Within a few minutes of getting into Bubsy 4D, you realize you’ve done most of this before, but much better… and that’s not a feeling you want a video game to give you.

Fortunately, these are problems that can be addressed with time and attention on the part of the developers. This is a preview of a work in progress, not a review of a finished product, and there’s a lot of room yet for vast improvement before release next year. The potential is certainly there… Bubsy’s MASSIVE horizontal mobility is a great deal of fun in practice, and if the environments continue to be tightened up to match the cat’s core abilities, the team at Fabraz could be on to something really fun.

I do have one specific complaint about a design choice that doesn’t work so well: the demo’s only mandatory timed challenge, a racing romp across some steep blind hills and a bridge that pretty much demands memorization to be successfully completed. I had hoped that in the Year of Our Lord Two Thousand and Twenty-Five we might have moved beyond cramming an unskippable, fairly unintuitive, timed obstacle race into the golden path through a level, yet here we are with Bubsy 4D. These were not fun in the PlayStation 1 era, and 30 years later they are unforgivable sins.

Elsewhere, Bubsy purchases additional abilities from a central hub store run by his annoying niece and nephew, trading collectables for power-ups. My favorite of these was a Wile E. Coyote-inspired power that allowed me to float above empty space for a second like a Looney Tunes cartoon character who’s just run off a cliff, a flavorful and genuinely useful inclusion. It was thematically appropriate and adorable.

Finally, on the audio front, Bubsy 4D’s sample scores ranged from unremarkable to droning. The theme music for the first three stages started with an innocuous quality, but after a few loops became grating. I can only qualify the sound as something that began its life as elevator music that was then run through an N64 nostalgia filter and digitizer, and what came out the other side did little to add to the platforming experience. Maybe that one is just me and others might enjoy it more.

Voice work, on the other hand, is solid, and the script has a lot of nods to the mascot-platformer era. Bubsy maintains a heightened self-awareness throughout his new journey. He’s openly unenthusiastic about his latest quest for the golden fleece. He’s older, tired, and kind of over it all, and the comments that reflect that sense of 40-something exhaustion mostly work. The occasional sarcastic quips and poop jokes seem obligatory nods to the era that birthed Bubsy, but nothing I heard was laugh-out-loud funny. I did not like Bubsy’s family and cohorts, who sort of came across as worse iterations of Sonic’s terrible friends or DK’s annoying extended catalog of characters.

Overall, Bubsy 4D is, thus far, a bright, interesting template for a 3D platformer. Whether it will grow to fulfill its full potential before release is difficult to predict. The obvious care of the developers in treating the character with some reflection, even a possible wisp of dignity, works in 4D’s favor. Hopefully the final product will capitalize on the strong start evident in this demo.

Jared Petty likes writing about how wonderful and silly video games are. You can find him at Bluesky as @pettycommajared and Threads as @pettycommajared.

Glenn Israel, a 17-year veteran of Halo game development, has announced his departure from the series via a startling message that has left fans worried over the future of Xbox’s flagship shooter franchise.

Originally an artist at Bungie, Israel ultimately served as an art director at 343 Studios (now Halo Studios), with contributions to a long list of Halo projects from 2009’s Halo 3: ODST, through Halo: Reach, Halo 4, Halo 5 and Halo: Infinite.

Now, however, Israel has announced he is “officially no longer contributing to the Halo universe” via a post on LinkedIn, and updated his profile to indicate he departed Halo Studios this month.

“As of today and after 17 long years, I am officially no longer contributing to the Halo universe,” Israel wrote. “There is little more I can say for the moment, though I intend to share this particular story in its entirety when it is absolutely safe to do so next year. In the meantime, I have a message for anyone and everyone who needs to hear it.

“I know that the state of our industry seems dire, but never forget that you are *free to choose*. No illusion of security nor promise of wealth or fame or power is worth trading away your health, your dignity, your ethics or values – and no one can force you to. Stay strong, take evidence when necessary, and find where you belong.”

Needless to say, Israel’s statement has prompted concern and confusion among Halo fans, at a time when the future of Microsoft’s once-emblematic series is far from certain.

Microsoft gaming boss Phil Spencer has previously suggested that a new Halo project will launch in 2026, with the expectation being that this is the previously-reported Halo: Combat Evolved remaster that will also see the series launch on PlayStation for the first time.

Fan hopes are high ahead of a much-trumpeted announcement due later this month where Halo Studios has said it will be “sharing more about what we’ve been working on” since it revealed a first look at the franchise’s transition to Unreal Engine 5 exactly a year ago.

With all this in mind, however, fans now say the timing of Israel’s departure seems particularly ominous, with Halo Studios on the cusp of revealing what it has planned for the franchise’s future, and just a few weeks after the quiet departure of Halo Studios chief of staff Melissa Boone.

For Microsoft, meanwhile, which has suffered a tumultuous year marked by layoffs and increasing fan dissatification around price rises, Israel’s public statement likely couldn’t have come at a worse time. Though more worrying still, perhaps, is the developer’s statement of sharing a more detailed account at a later date.

IGN has contacted Microsoft for comment.

Tom Phillips is IGN’s News Editor. You can reach Tom at tom_phillips@ign.com or find him on Bluesky @tomphillipseg.bsky.social

With a week to go until the official launch of Pokémon Legends: Z-A, fans have issued a warning over online spoilers, as datamined details from the game have begun to leak online — including word of even more new Mega Pokémon.

Pokémon Legends Z-A is due to launch on October 16 for Nintendo Switch and Switch 2, but already there’s evidence of early copies filtering out into the wild. And now, as of this morning, IGN has seen evidence that the game is actively being datamined to extract its secrets.

To date, The Pokémon Company has revealed just a handful of the new Mega Pokémon expected to make up the game’s final roster of fresh Mega species. And while an older leak had listed out many of the still-to-be-confirmed Mega Evolutions, there’s evidence today that even more creatures may be included than first expected.

In response to all this, Pokémon fans have now begun signalling that anyone hoping to go into Pokémon Legends: Z-A without seeing spoilers should be mindful of where they click online over the next few days — or log off social media completely. (IGN will not be spoiling any of the new species here.)

Set in Kalos, home of the Mega Pokémon mechanic, Pokémon Legends: Z-A adds the first new batch of superpowered creatures to the franchise’s Mega Pokédex in over a decade, although no all-new Pokémon species are expected.

New Mega Evolutions include the majestic-looking Mega Dragonite — which has also now made its Pokémon anime debut — plus Mega Victreebell, Mega Hawlucha and Mega Malamar.

Post-launch, Pokémon Legends: Z-A will unlock Mega Evolutions for each of the three fully-evolved Starter Pokémon, available via online play: Delphox, Greninja and Chesnaught. And beyond that, you can already buy Pokémon Legends: Z-A: Mega Dimension, a paid DLC which will add two new Mega Evolutions for Raichu.

Indeed, some fans have complained about the costs of obtaining every Mega Pokémon announced so far, something which requires more than $100 to obtain the base game, DLC and a Nintendo Switch Online subscription.

How many other new Mega Evolutions will Pokémon Legends: Z-A include? We’re about to find out.

Tom Phillips is IGN’s News Editor. You can reach Tom at tom_phillips@ign.com or find him on Bluesky @tomphillipseg.bsky.social

EA has insisted players won’t see anything made by generative AI in Battlefield 6.

The use of generative AI is one of the hottest topics in the video game industry, with the pressure on publishers to cut costs and speed up development in order to boost profits despite the risk of backlash from some fans. Indeed, according to a report by The Financial Times, EA’s new prospective owners (the ones who just spent $55 billion to take the company private) are betting on the use of generative AI to do just that.

And EA itself, even before it was bought out, had signalled that it was all-in on generative AI, with CEO Andrew Wilson insisting AI is “the very core of its business.”

For now though, at least when it comes to Battlefield 6, generative AI is taking a backseat. Rebecka Coutaz, general manager of original series developer DICE in Sweden, and Criterion, the UK studio now also a part of what’s collectively called Battlefield Studios, told the BBC that while generative AI “is very seducing,” currently there is no way to work it into the developers’ daily work.

However, Coutaz clarified that generative AI is used in preparatory stages “to allow more time and more space to be creative.”

EA’s approach here for Battlefield 6 is in contrast to Activision’s for Call of Duty. Call of Duty has suffered a number of generative AI controversies in recent years, including the now-infamous six-fingered zombie Santa bundle. Earlier this year, Activision was forced to add an ‘AI generated content disclosure’ to the Steam page for Black Ops 6 after Valve changed its storefront rules.