Project Link Up is set to revolutionise vector design, turning the simple dragging function into context-aware fusion that automatically matches a scene's perspective. (TLDR: constantly redrawing objects at new viewing angles could soon be a thing of the past).

To understand more about Project Link Up, I caught up with Hailin Jin, senior principal scientist and research lead for imaging and design at Adobe, to get his behind-the-scenes insight into the work.

But choosing the best AI video generator for you isn't only a question of visual quality. The increasing number of models available have varying limitations and strengths and weaknesses in different areas. Pricing structures can also vary wildly, and subscription fees often represent only a fraction of the cost.

Many AI video generators operate on credit systems, charge extra for premium models, or limit video length and resolution in ways that can quickly inflate your spend. With that in mind, here we take a deeper dive into the five best AI video generators available today, which we've ranked according to overall value as well as their flexibility and practical usefulness for professional creatives.

Whichever model or models you choose to work with, you'll probably want one of the best laptops for video editing or the best desktop PCs for video editors.

The best AI video generators

Adobe is already a powerhouse in filmmaking thanks to Premiere, which we rate as the best video editing software. As a result, Firefly more than just an AI video generator. Unlike many rivals that function as standalone tools, it integrates into Premiere and the wider Creative Cloud environment, making it significantly easier to incorporate AI-generated footage into existing professional workflows.

The most recent video-generation capabilities are particularly impressive. Prompt adherence is strong, footage tends to look polished and cinematic, and Adobe has invested heavily in making outputs commercially viable for professional users.

As announced at Adobe Max 2026, the native Firefly Video editor has been expanded to make it easier to combine clips with other elements such as images and sound effects, and projects created in Firefly Boards, Adobe's collaborative brainstorming space, can be carried straight into the video editor. There's also Adobe's AI agent, which makes using its software more intuitive and conversational.

The Adobe name also offers reassurance in terms of safety and reliability. The platform's unlikely to disappear or radically change from one day to the next, and output from the native Firefly Video Models is intended to be safer for commercial use since they were trained on licensed material rather than scaped copyrighted video.

On the other hand, if you do want to use other models, they're here in Firefly itself. Google Veo, Kling, Runway and Luma AI's Ray are all integrated and ready to use in Adobe's platform, so you get not only the top AI video generator on our list but access to the second, third and fourth too!

Pricing: Adobe's Firefly Standard plan costs $9.99 per month and includes 2,000 generative credits, enough for around twenty 5-second AI video clips at 1080p quality. Higher-tier plans increase those allowances. Adobe currently prices 1080p video generation at roughly 100 credits per second, meaning a single 5-second clip consumes around 500 credits.

You can also access Firefly video generation through a Creative Cloud Pro plan, which gives you access to Adobe's suite of apps, including Premiere and Photoshop. Just note that you may need to pay for additional generative credits depending on how much footage you're generating, and partner models have different credit costs.

If pure visual quality is your priority, Google Veo has a strong claim to being the most impressive video model currently available. The platform is particularly good at realistic movement, lighting, camera behaviour and environmental detail.

Veo's ability to generate convincing cinematic scenes has made it popular among advertisers, content creators and filmmakers seeking premium-looking footage without a full production budget.

Pricing: Google offers Veo through consumer subscriptions and API access. For developers, Veo 3.1 costs roughly $0.40 per generated second, putting a typical 5-second clip at around $2.00 before retakes. That's easily the most expensive platform in this comparison, but it also delivers some of the highest-quality output available.

You can also access Veo through Google Gemini, Google AI Studio and through Adobe Firefly (see above). Through Gemini, there is a free tier offering just 3 video generations a day, at the time of writing, Google AI Pro costs $19.99 a month for 1,000 Flow credits, which translates to around 100 Lite videos or 10 high-quality Standard videos at 720p with a watermark. Google Ultra AI takes at $99.99 a month and upwards gives you 25,000 credits a month, enough to generate thousands of draft clips at 1080p.

Runway remains one of the most mature AI video platforms on the market. While some competitors may occasionally produce more realistic footage, Runway excels at putting creators in control. The platform includes advanced camera controls, editing features and creative tools that make it particularly appealing to professionals who want more than simple text-to-video generation.

The video editing features enable you to do things such as isolate a subject and remove the background, colour grading, adjusting the depth of field, making a slow motion version of your footage, and 'Inpainting', which means you can easily remove unwanted objects.

Pricing: Runway has several subscription tiers, ranging from entry-level plans through to professional and enterprise options. The credit-based system rewards efficient users but can become costly during experimentation. The Standard plan starts at $15 per month and includes 625 credits. According to Runway's published figures, that's enough for around 52 seconds of Gen-4.5 video generation. A creator producing multiple social videos every week will likely need the Pro plan at $35 per month or higher.

The Chinese platform Kling has become one of the biggest disruptors in AI video generation. It consistently produces impressive results while undercutting many western competitors on price.

Human movement, character animation and stylised content are particular strengths. For YouTubers, social creators and small businesses, Kling often represents the best balance between quality and affordability.

Pricing: Paid plans start at under $10 per month, and a 5-second 1080p clip generally consumes around 40 credits. That works out to roughly $0.40-$0.60 per clip depending on subscription tier, making Kling one of the most cost-effective premium AI video generators currently available and generally cheaper than Veo and Runway. However, as with most platforms, higher-quality generations consume credits faster.

Not every project needs blockbuster-quality visuals. Sometimes speed matters more than perfection, and that's where Hailuo shines. The platform is built around rapid iteration, allowing creators to test ideas quickly and cheaply. Agencies often use it during concept development before moving final production to more advanced systems.

Pricing: Hailuo is typically among the most affordable options available. Standard plans start between $8 and $15 per month depending on promotions and region. A typical 1080p generation consumes about 50 credits, placing the cost per clip around $0.50-$0.75.

The main trade-off is quality, and you might find yourself generating more clips to achieve a satisfactory result, which can offset the initial cost advantage. Additional editing and post-production work may also be required.

The hidden costs of AI video generators

Before diving in to AI video generation, it's worth remembering that subscriptions don't always tell the whole story. A $10-per-month plan may sound inexpensive, but the number of usable clips you can generate could be surprisingly limited.

Making direct comparisons between AI video generators is difficult because some services charge per second of output, while others use credit systems that effectively works out as a cost per clip.

Pricing can also vary depending on which plan you choose and how much you use it. Heavy users are likely to have to pay extra for more credits, and plans are continually in flux. Providers have been making regular changes to their pricing as new models roll out and the competitive landscape changes. The prices being quoted today could easily change in a few weeks' time.

The challenge with any AI video generation is that creating a complete project often requires multiple generations. A single eight-second clip may be excellent, but producing several matching shots for a finished sequence can dramatically increase usage costs. You'll often need to generate dozens of variations before selecting a final clip, turning what appeared to be a modest monthly subscription into a much larger operational cost.

Users will also usually need separate editing software to assemble final productions.

Conclusion

For most creative professionals, we think Adobe Firefly currently represents the most balanced and versatile package while also offering reasonable value. While Google Veo may edge ahead on raw visual realism and Runway offers deeper creative controls, Firefly's integration with professional creative software makes it the most practical choice for real-world production environments, and Firefly does also provide access to Veo and Runway too.

That said, the right tool depends heavily on your needs. Consider Veo for cinematic quality, Runway for directorial control, Kling for affordability and Hailuo for rapid experimentation. Just don't judge any platform by its headline subscription price. In the AI video market, credits, render limits, premium models and editing requirements often matter far more than the monthly fee itself.

Unsurprisingly, this year's Sneaks have a heavy focus on AI across the ten innovations. We saw demos of tools that could make AI image and video editing more precise and transform the workflows of people working in film, digital art, photography and design. Here are the highlights.

• Get Adobe Creative Cloud Pro

Project Scene Sweep

On the video and filmmaking front, Adobe's revealed what it's calling Project Scene Sweep. The AI-driven tool is being designed to understand characters, locations and scenes across a sequence, helping creators apply an edit consistently across cuts and camera angles.

It allows long-form, multi-shot videos to be edited from a single text prompt or edited frame, with the changes applied across each shot, cut, and camera angle. Example use cases cited include post-production fixes like removing a person or prop or swapping wardrobe without a reshoot.

The tool could also help for clearance issues, for example to replace an unlicensed logo or artwork in every shot where it appears, or for product placement after a shoot. Adobe suggests it could even be used to recast a new lead or location for a video or for testing variants for localisation.

Other projects in the works for video editing are Project Sound Bender, which would allow users to describe a sound and pull it out of a finished mix, then explore variations on an interactive canvas, and Project Live Set, which would allow filmmakers to change a subject or environment or redirect the camera as a scene continues.

Project Deep Brush

For digital art and image editing, Project Deep Brush uses AI to understand an image’s shape, colour and lighting and separate images into physically decomposed layers. This would give artists a new level of control for painting and retouching, including the ability to control where paint lands and to make lighting-aware edits in photos.

Users would be able to add things like graffiti, signage, wraps, or branding to surfaces like walls, pillars, and vehicles, making the tool a potential game changer for product and environment mockups. For example, you could put a logo or texture on a curved or angled surface like a bottle, and the edit would follow the surface’s contours and lighting.

You'll be able to try this out for yourself on the new Adobe Labs Playground site.

Project Hands On

Project Hands On is presented as a possible step from text-based editing towards a more fully graphical interface allowing greater control of AI-generated imagery. It analyses what’s in the image and then adds controls, such as control points that can be used to reshape details by simply dragging rather than rewriting a text prompt.

This could make editing AI-generated output more precise and intuitive, avoiding the trial-and-error nature of having to repeatedly rewrite prompts and regenerate an image.

Project Screen Test

Project Screen Test was perhaps the wildest idea teased at Adobe Sneaks. It would allow character designers to talk with their fictional creations to learn more about them and how they might behave in a given situation. Yes, seriously!

The tool would allow creators to develop a character’s look, voice and personality, and then have a natural language dialogue with said character in order to explore who they are and how they might respond in different scenes. The character would effectively audition for you.

The idea is that this conversation could turn character development in generative filmmaking and game development into a more intuitive, expressive part of the storytelling process.

Project Sight Span

Project Sight Span is tool that feels a lot more familiar. It transforms an image into an interactive, navigable world that creators can step inside and explore freely. That exploration could be used to frame shots beyond the original image, and render new images or camera-controlled videos, a little like what some artists are already doing with Gaussian splats. Adobe reckons this could be useful for things like set design and world building.

Project Link Up

Another highlight of Adobe Sneaks 2026 was Project Link Up. With existing vector tools, dragging an object onto another only determines where it lands. Project Link Up turns that drag gesture into semantic fusion, combining an understanding of both objects with drag path, velocity, and creative controls to shape how they influence each other in real time, while keeping the result fully editable as vector artwork.

The idea is that artists and designers would be able to place 2D or vector assets into scenes with automatic perspective matching, reducing the need to redraw objects at new viewing angles. For typography, it would be possible to merge illustrated elements with letterforms while keeping the type editable for branding, logotypes, and packaging design.

For more news from Adobe Max, see what's new in Photoshop 2027.

On paper, the Boulies EP600 can be adjusted to a ridiculous degree. It has an innovative floating lumbar support that slides up and down depending on your fore or aft position, which would make it one of the best office chairs for back pain. In reality, it’s just not that comfortable.

Comfort & ergonomics

The real highlight of the Boulies Ep600 is the lumbar support, a really smart piece of counterbalanced engineering that tracks up and down with your lumbar as you move forwards and backwards on the chair. It achieves mechanically what the Hbada X7 failed to do electronically, with none of that chair’s irritating motor noises and intrusive micro-adjustments. It can be locked in place, too, but I found the floating mode suited me really well.

While the lumbar support is the best feature here, it’s coupled with the absolute worst: the back support. There’s nothing wrong with the support itself, but it’s positioned too far back in relation to the seat pad. If you want back support, then you have to lean right back in the seat, and you’re then an unfathomable distance away from your screen. The headrest is fine, but its bracket either goes higher or more forwards, but never both. As a taller person, I found that the chair was conspiring to position my head as far away from the monitor as is physically possible.

This distance is what got me questioning whether I’ve been sitting properly for all these years. Are you meant to sit like your computer screen is repulsive? It's something that I’ve been aware of in other chairs, but here it’s really noticeable: ideally you should be able to touch your screen with a fully stretched arm when your back is supported, but the Boulies EP600 puts a good few inches between my fingertips and Reddit.

When you want to relax, the chair does lean back further, and while the pull-out leg rest is appreciated, it’s made of rock-hard plastic. I’d sacrifice the leg rest for a chair that leans forwards to offer back support when you’re in those intensely focused sessions — it was a feature I really appreciated when I reviewed the Titan Evo NanoGen.

Elsewhere in the Boulies EP600 it’s business as usual. The armrests flex and pivot all over the place, and even tilt upwards or out of the way completely, but they’re coupled with very, very firm cushions that make you painfully aware of your elbow bones. The seatpad also moves backwards and forwards to accommodate different leg lengths and preferences.

Design & aesthetics

From the front, the Boulies EP600 tends towards the uglier side of office chairs, with its curved back panels resembling an insect: thorax, abdomen, and. It looks better from behind; its curved central column gives it an aesthetically pleasing structure, and its chunky armrests make it feel like the captain’s chair on the U.S.S. Enterprise.

One thing I like about mesh chairs (and this is mesh throughout) is that their transparency helps negate some of the weightiness of the design, and this is the case here. The grey colour scheme Boulies supplied has a slightly Y2K feel, especially in the silver plastic under the armrests and the metallic accent on the rear column. It’s also available in black, which looks even more insectoid.

I almost forgot to mention the pull-out leg rest, because I’d blanked it from my mind, but it’s really hideous; a single piece of plastic that wouldn’t look out of place in a 1970s hospital, which was truly the worst time for healthcare implements.

The Boulies EP600 is never going to win a good-looking chair contest, but this bug-like design language has become fairly standard in mid-range ergonomic chairs, and I don’t think it’s better or worse than any other.

Specs

Assembly

One thing I’ve come to loathe in chair construction is bases that need to be screwed together. I get that it saves packaging, but it’s such a chore! Thankfully, Boulies has overcome this with arms that lever themselves into position when you stand on the centre and, oh my gosh, it’s one of the most satisfying things I’ve ever experienced. More of this, please!

The manual is a little basic, and there are a few bolts that are fiddly to get in the right place, but overall it’s a 15-minute process with very little that can go wrong. The fully built chair feels solid and well-made; there are no creaking parts and nothing wobbles.

At 21kg, it’s reassuringly heavy, and its wide dimensions make it tricky to move through my Hobbity Victorian English home, so this is one that you won't want to move around too much.

Price & value

The Boulies EP600 carries a retail price of $489.99 / £429.99, but at the time of writing it’s $409.99 / £339.99, placing it firmly in the mid-range of ergonomic chairs.

For around the same price, you can pick up the Sihoo Doro C300 Pro, which includes a similar range of features. It lacks Boulies’ dynamic lumbar support and built-in leg rest, but it’s still a supremely comfortable chair with those all-important adjustable armrests.

If you want even more adjustability in a similar form factor, look no further than Hinomi’s H2 Pro. While its price point is a bit higher (it’s currently $569 / £569), it has adjustable points just about everywhere, and I found it really comfortable and supportive once I got the fit dialled in.

Should you buy?

Boulies primarily makes gaming chairs, and it feels as if the EP600 hasn’t quite shaken off its specialisation. It could work well if you’re primarily into gaming or watching movies and occasionally delve into more focused tasks. The flip-up armrests make it good for people who like to draw at their desks or noodle on a guitar to get those creative juices flowing. But designers, editors, and chair review writers who like a more upright position for focused work just won’t find enough support here.

It’s a shame because the lumbar support is smart, the armrests are versatile, and the mesh design is comfortable and cool. It’s also one of the easiest chairs to assemble. A few tweaks, such as better back support and plusher armrests, and the Boulies EP600 would be an instant classic. But as it stands, it’s hard to recommend, and I'm confident that the problem is with the design, not the person who's sitting on it.

This is a bean-to-cup coffee machine. It can grind your beans, brew a wide range of coffees and froth milk to perfection all while looking the part, and it comes from one of the most reliable brands in the mid-range of the market. See below for more details on the deal.

The best early Black Friday coffee maker deals

• Browse coffee machine deals at Amazon

Grinding your beans fresh right before brewing is one of the keys to great tasting coffee. For many, that means choosing between a separate grinder and coffee machine or a large bean-to-cup machine that eats up a good chunk of a kitchen counter. The Rivelia LatteCrema has been designed to do everything without taking up minimal space.

Automatic adjustments and interchangeable 250g bean hoppers make it easy to swap between different coffees, and the LatteCrema Hot Technology lets you adjust foam levels (from flat/hot milk to light or thick, super-fluffy foam) for drinks like cappuccinos, flat whites, and latte macchiatos.

The intuitive one-touch display lets you swipe through a variety of 16 drinks and allows you to store up to 4 user profiles, which can remember your favorite drink strengths and volumes.

Not in the UK? Check out coffee machine deals in your region below.

Space Invaders was the first game I ever played, or at least the first I can remember playing. I played it on a black-and-white CRT with an Atari 2600 (which is why I loved the Atari 2600+), crouched cross-legged on the floor, my face pressed against the TV screen. Our local swimming pool had an arcade because, back then, game cabinets appeared everywhere, but it was off-limits to me at eight.

I’m an adult now, so I can have Space Invaders, which remained one of the best games of the 80s after its release in 1978, on my wrist and play it whenever I like. In colour, too. That’s the pitch behind the My Play Watch series, which brings retro game brands back as digital watches. (For more, check out the indie game aping the visual design of old LCD game and watch gadgets.)

The Mega Man-branded My Play Watch is a different story. I came to the series late, on PlayStation, but of the two, this is arguably the watch with the more interesting game. It’s a reworked version of Mega Man 2, playable on a tiny watch face. Mega Man runs automatically; tap and hold the screen to jump, then release to fire, and beating bosses unlocks new watch faces.

My Play Watch: Mega Man vs Space Invaders

My Play Watch Mega Man: the better game

The blue retro watch and game capture the feel of Mega Man 2 despite the simplified controls, and the bosses and music are still there, bleeping away on my wrist. It isn’t a straight port, though. There are no Robot Master weapons, some enemies and minibosses are missing, and the levels have been changed. Controls can get fiddly when your finger covers the screen.

And yet I’ve kept Space Invaders on my wrist, even though its version of the game is simple. You move the little ship left and right using the side dial and tap the screen to fire. Sure, it's not as easy as playing on one of the best retro game consoles like the new Evercade Nexus, but look how tiny it is. It’s easy to pick up but hard to play while wearing, and the original arcade sounds and pixel-art watch faces sell the nostalgia, even if anyone nearby will tut and raise an irksome eyebrow at the loud bleeps and famously high-pitched 'pew-cssssh' as an Invader turns to pixel dust.

My Play Watch Space Invaders: nostalgia on your wrist

Nostalgia and collectibility are what sell these watches, and they’ve sold me. I love looking down and seeing the little animated Space Invader pixel monster waving back at me, or the bright original cabinet art sat on my wrist.

But they also have to do more than simply callback to my childhood, and that’s where both watches have a smart retro design. The smartest touch is the health tracking. Step count, calories, and heart rate turn into gamified stats, with my daily rates shown right next to my Space Invaders high score. It’s a small thing, but it makes checking my wrist and health updates feel like celebrating a scoreboard.

Which My Play Watch should you buy?

Elsewhere, these watches do what a digital watch needs to do in 2026: show the time, a calendar, a stopwatch and a calculator, but they have no notifications or advanced smartwatch features, which is either refreshing and on-brand for a retro game and watch, or a lack of digital detail. Personally, I’m happy to not have WhatsApps buzzing every 10 minutes. The design is a nice touch, too. Both use UI and fonts inspired by their games, and Space Invaders makes stellar use of pixel-art digits that take me straight back to 1982. The strap is spray-painted with Invaders, too (and each watch includes an alternative strap if you want to tone down the retro branding).

Just keep your expectations in check. This collectable tells the time, not an Apple Watch or Garmin rival. No notifications, a small screen not built for long play sessions but designed to take you back to simpler times and scratch that collectables itch – new editions include Street Fighter 2 and a gloriously obnoxious glowing green and pink Centipede edition that I kind of want for no apparent reason but mostly because these things become a container for our emotions.

At around $79.99 / £79.99, Mega Man is the better game, but Space Invaders is the one I’ve kept on, because, like all things retro, it’s completely personal.

Other My Play Watch editions

They do cost a fair few bob, starting at $1,199 but venturing into the $5000 territory, so they aren't craft machines for the people at this stage.

CNC machines work in the opposite way to the best 3D printers or the best UV machines. Rather than building an object up layer by layer, they carve it out of a solid block with a spinning cutter. That means you can create strong, precise parts in metal, wood and stone, built to last.

I've picked the best CNC machines available to consumers today, along with two promising Kickstarter models that we're excited to see materialise.

The best CNC machine for beginners

Makera launched as a company in 2019, with the aim of lowering the cost of precision desktop manufacturing. Two years later they made a big splash with their first CNC machine, Carvera, which debuted on Kickstarter and is still popular today: we've put it as number two on our list. But the Z1, which launched in 2025, again on Kickstarter, has shaken up the market by driving the cost of entry even lower.

Despite being the cheapest on this list, it does everything a beginner will need. It's built on a cast aluminium frame and fully enclosed. It offers automatic probing and levelling, a built-in camera you can connect to your phone, and an air blast that clears chips as it cuts. And its software offers one-click toolpaths, so you don't need to learn complex software before you get started.

The tradeoff for the cheap price is that it's more limited in terms of size and power. The 200 x 200mm work area also means you can only work with small parts, while the 150W spindle requires shallow, patient passes in metal. Another drawback compared to the Carvera is that you'll need to change tools by hand via a quick-release lever.

If you're seeking your first CNC machine, though, this is our top pick.

The best CNC machine for experienced users

The Carvera has been around since 2021 but remains Makera's flagship. With this CNC machine you get a six-slot automatic tool changer, letting you swap between different milling bits mid-job, saving you from having to fiddle about with your hands. It's capable of cutting and milling wood, plastics, acrylics, carbon fiber, FR4 (PCBs) and non-ferrous metals like aluminum and brass.

Other features include an automated wireless probe for surface mapping and bed leveling, built-in dust collection and a 2.5W laser for engraving and engraving and PCB labelling. The enclosed design keeps noise to around 65dB: not silent, but still quiet for a CNC. You can cut wood, plastics, PCBs and non-ferrous metals like aluminium and brass well, and if you want to make circuit boards at home, Makera will sell you a dedicated bundle for that.

The 200W spindle is modest, and Makera doesn't recommend you try to cut steel or titanium. And while the crowdfunded rivals below promise more power at a similar price, remember that they're not actually available yet. This is, and has years of independent tests and positive customer reviews behind it.

Up-and-comer

Twotrees produces affordable laser engravers and 3D printers, and the ED1 is its latest CNC machine. It offers automatic tool changing, like the Carvera above, but at a more affordable price.

While the ED1 is closer to shipping that the CNC machines below, it's only available for pre-orders, which will ship from January 2027. That means it hasn't yet been independently tested, so at this stage we can only go on the specs.

They do, though, look very impressive. The five-slot changer is said to swap cutters in around 10 seconds, the 600W spindle has three times the power of the Carvera's – Titanium is on its list of materials – and the 300 x 200mm work area is a decent size. You'll also get auto levelling, a built-in camera and AI-assisted software.

Kickstarter products to watch

These CNC machines haven't shipped yet, so we haven't included them in our main list. Both are due to arrive in early 2027, though, so assuming all goes well, they're well worth keeping an eye on.

Infimaker was founded in Shenzhen in 2024 by former DJI engineers, and the K1 is not just its CNC machine but its first product at all. The Kickstarter closed in September, but late pledges are still open, with delivery estimated for February.

So what's different about this one? Well, most desktop CNC machines are 3-axis: the cutter moves left and right, back and forth, and up and down. This adds two more axes: tilting and rotating the workpiece as it cuts, which allows you carve complex shapes like impellers and sculptures in a single setup, rather than having to re-clamp the part by hand.

This CNC machine features a 1.5kW spindle, a six-tool automatic changer and AI-assisted software. Be aware, though, that you'll need a strong desk, as this weighs a hefty 110kg.

Tooldance X1

The Tooldance X1 is a crowdfunded CNC machine with a 1.5kW permanent-magnet spindle that aims to deliver high torque at low speeds, enabling you to work with stainless steel and titanium. It promises the largest work area on this list at 400 x 265 x 180mm, along a nine-slot automatic tool changer that uses RFID tags to track its cutters, and sensors that measure your stock for you.

The Kickstarter is still live, with delivery estimated for March 2027. So as you'd expect, there've been no independent tests yet. If you're in the US, also note that the spindle's output drops to around 1kW on a 120V supply.

FAQs

Creating artwork that can make viewers feel amazed requires what I call ‘epic scale’, which is a technique I specialise in. It focuses on placing a tiny figure in front of enormous architecture, emphasising the contrast between the small figure and the vastness of the environment.

The software I use for my entire 3D workflow is Blender, with additional adjustments and post-processing done in Photopea (see our guide to the best 3D modelling software, and you might want one of the best laptops for 3D modelling).

For more inspiration, see our Blender tutorials.

Gear up

See below for today's best deals on drawing tablets.

This article originally appeared in ImagineFX. You can order back issues.

Creative Bloq is now easier to access than ever before with our on-the-go app, which brings you all the content you know and love from our website, but in a super-streamlined design.
Download the Creative Bloq app for iOS
Download the Creative Bloq app for Android

Since its 1998 release, The Legend of Zelda: Ocarina of Time has been revered as the greatest game of all time, influencing many games, though few have ever tried to replicate its formula beyond subsequent 3D Zelda titles. That is, until Breath of the Wild blew open the doors, reinventing Nintendo's most beloved series in the modern open-world age of player agency and freedom, and its sequel, Tears of the Kingdom, reached new, unimaginable heights.

Returning to the N64 classic then might feel like a quaint relic of the past. So it's a revelation that, for the series' 40th anniversary, revisiting this beloved masterpiece remade for Nintendo Switch 2 has achieved an alchemy that rekindles the magic of what it must have been like to play Ocarina of Time for the first time 28 years ago.

This technically isn't the first time that Ocarina of Time has been remade, but whereas the 3DS update was largely a refurbishment that nonetheless stuck to the original blueprints, this Switch 2 remake has been fully rebuilt from the ground up for a new generation. The most notable difference, of course, is the art direction, which isn't just an exercise in Unreal hyper-realism. There's such attention to detail, from every blade of grass to the ridiculously detailed patterns and textures that adorn Link's hat and tunic. What's remarkable is that, unlike most modern games, all this visual density hasn't come at the expense of readability. I'd argue it's even easier to make out the cracked walls that bombs can blow up, while dungeons like the infamous Water Temple have improved architecture that makes navigation less confusing.

Ocarina of Time gets its character design right

Despite initial divisive responses to the new Zelda character designs (but then, I'm old enough to remember the 'Celda' reactions to Wind Waker), I soon found myself enamoured by the ornate puppet-style aesthetic reminiscent of fairy-tale and fantasy media from the 80s, while the still-mute Link says a lot just with his subtle facial expressions. Admittedly, Sheik is much less androgynous than the original game (though the twist is about as non-spoiler as Luke Skywalker's father), but Zelda herself is also an improvement over the dainty damsel, with a voice distinct from the stereotypical princess in the newer games.

It's also however not a 'reimagining' in the vein of Resident Evil 2 where the original story has been adapted to a modern iteration of the series, nor is it just a primarily graphical overhaul like Metal Gear Solid Delta: Snake Eater that otherwise plays almost identically. After Breath of the Wild and Tears of the Kingdom's vast open world, trekking across Hyrule Field in a matter of minutes should feel ridiculously small, yet I have to give kudos to Nintendo for not succumbing to the temptation of stretching it out and then stuffing it with filler content, though the original side quests, from item-trading to hunting down Gold Skulltulas, are intact.

There's still a majesty to this finite adventure of a courageous boy journeying across time to save the world that still evokes so much, whether it's when you first pull out the iconic Master Sword or just stand in the middle of Hyrule Field and just make out the surrounding locations based on their distinct landmarks without the need for waypoint markers. A maximalist Final Fantasy 7 Remake approach that stretched the narrative with many more things to do would have diluted, rather than enhanced, the experience.

Still, plenty of modern additions make for a much smoother playthrough, from a free-moving camera to 60FPS performance in stark contrast to the original's barely stable 20FPS, and a clear quest log that also weaves a beautiful tapestry chronicling your adventure. There's even a dedicated jump button for Link, which admittedly took me a while to adjust to, since I kept falling off platforms when I expected an auto-jump; you can just dash instead of the original's hack of rolling around ad infinitum.

How Ocarina of Time on Switch 2 mixes the old and new

You can also see some of the newer games' design incorporated, such as being able to quick-select items, though die-hards may lament that this means getting rid of the ability to map three items at once - it's balanced out by having some items like the lenses of truth or iron boots being mapped to respective buttons on the D-pad. Weapons like the slingshot, bow, and hookshot can also use gyro aiming, and while this was introduced in Ocarina of Time 3D, those still retained the original's first-person perspective, whereas now it's third-person like the newer titles, and you can move around at the same time.

Personally, my favourite new feature exclusive to Switch 2 hardware is being able to play the iconic ocarina songs by humming the tunes into the console's built-in microphone. For sure, it's a gimmick, but a very apt one, and I found it so much more intuitive than trying to memorise the button inputs, which is also a testament to how memorable Ocarina of Time's music continues to be.

Zelda's modern polish downside

If there is a downside to all the modern polish, it's that playing through Ocarina of Time is perhaps a little too breezy. That's certainly true of all the target and archery minigames, where, with the intuitive gyro aiming, I could get the top prizes on the first go without breaking a sweat. That also speaks to the absence of a more challenging option such as a Hero mode (you can opt to not increase your hearts, which is easier to do as you have to press a button to pick up heart containers), though an obvious choice would have been a Master Quest mode that remixes the dungeons, which could still happen post-launch either as a free update or paid DLC.

But Ocarina of Time was never as hardcore as Dark Souls; it was more about the pleasure of discovering new items you could put to ingenious use, especially against its brilliantly designed bosses, including an improved multi-stage finale. It's also nice that one of the remake's genuinely new features is a museum of all the enemies you've encountered, which lets you jump into boss rematches and even boss rushes.

If the overall changes to the game and story are relatively minor – voiced cutscenes are fine but hardly a gamechanger, though it's nice that Matt Mercer reprises his role as Ganondorf – it's really a testament that the original 1998 release was already such a defining moment in gaming that it doesn't require a radical overhaul to remind you just how great it still is. More importantly, in a post-BotW world, it's a relief that an adventure of this size can still feel epic and timeless.

The new identity is a fine example of a redesign done right. It feels more modern, more human and approachable, and it remains instantly recognisable, maintaining a clever hidden symbolism.

See our guides to the best laptops for graphic design and the best monitors for graphics artists if you need new tools for your own work.

The Caixa rebrand was the work of the Brazilian agency Tátil. It looked to the bank's history and its existing identity, kept the elements that people most recognised, and turned them up.

The previous logo design had lasted 29 years, so it was doing something right, but it was starting to look drab and dated with the old corporate feel of those muted colours and slightly italicised all-caps letters.

The rebrand turns the colours up with more modern, vibrant hues. Tátil suggests the blue represents enduring solidity while the orange reflects the diversity of Brazil.

The wordmark gets a cleaner lowercase design, and the core X motif, which features in the wordmark and also stands alone, gets a significant refresh.

The letter now more clearly resembles a box when applied in repeating patterns, reflecting the bank's name, which means box (as in savings box) in Portuguese. It also continues to resemble the roof of a house, reflecting the bank's role as Brazil's biggest mortgage lender.

You can see the evolution of the institution's branding in the video below.

"While every app was busy calling itself a bank, we went against the grain and found what only Caixa has, a bank that brings together the social and the commercial, today's needs and the drive to grow," Tátil says of the rebrand.

It's no secret Adobe is going full send on AI, after a full fleet of AI partnerships and Firefly taking centre stage, blossoming into a full-fledged creative studio. It could all start to feel a little existential, but beaming under the stage lights, future Adobe CEO Anil Chakravarthy assures me, "Adobe will meet you where you are”. And I start to believe him.

AI has persisted as a divisive topic among creatives, but Adobe has shown a refreshing self-awareness around the tension. Emerging on stage, vice president Deepa Subramaniam wastes no time addressing that "things feel uncertain” in the wake of AI.

Assurances of “reliability” and “safety” are thrown around, honing in on Adobe's legacy as a brand shaped and "built by creatives that care about the craft”. She assures the sceptics among the crowd (myself included) that for Adobe, "Creativity isn’t a side project; it’s a way of life,” and to my surprise, the typically eye-roll-worthy sentiment soothes me. At its core, it seems Adobe hasn't forgotten about the creatives that shaped its brand.

Deepa raises some valid points. In an increasingly content-driven world, more is asked of us. The humble graphic designer is now expected to be a web designer, motion designer, and social media maverick all in one – something that Adobe's “rich learning ecosystem” can solve, according to Deepa. Creatives are trapped in a battle with waning time and resources, with demand for content ever increasing. If handled correctly, time-saving AI could offer the respite we need, but it raises a moral dilemma. Where do we stop?

For more Adobe news, check out why Adobe's Google partnership hints at a radically different future for Photoshop or take a look at the new AI features coming to Photoshop.

The entry-level specs won't interest professional video editors (see our guide to the best desktop PCs for video editors if you're in that category), but dedicated Nvidia graphics mean this budget setup is ready to handle lighter content creation as well as gaming, and it would mark a step up from a laptop with integrated graphics.

Why choose a desktop PC

Whether it's for gaming or content creation, there are two big advantages of a desktop PC over a laptop. Firstly, you tend to get more bang for your buck in the first place. As a rule, you tend to get more powerful internals, in the form of a superior CPU and GPU and more RAM, compared to what you would get in a laptop at the same price point (of course, some of this is outweighed by the fact that you also need to invest in a monitor and peripherals).

But the biggest advantage of a desktop PC is upgradability. If you find you outgrow the RTX 5060 too quickly, or processes are getting bottlenecked and you need more RAM, or more storage, you can buy the additional internals that you need as you need them and upgrade the configuration yourself, potentially multiple times.

This is a full 5.1.3-channel package with real rear satellite speakers and a wireless subwoofer. This makes it stand out from all those one-piece soundbars, which use virtual processing to try to mimic the effect of surround sound. The result is a fuller, more rounded and more immersive experience (also see our best headphones if you want options for listening alone).

• Browse soundbars at Amazon

A real 5.1.3 soundbar system for wraparound audio

Modern flat-screen TVs are too thin to house quality speakers, often resulting in muddy sound and hard-to-hear dialogue. Soundbars provide a budget, space-friendly upgrade without having to run speaker wires across your room or buy an AV receiver.

Many of them come as a single piece and use processing to try to replicate the effect of surround sound. The LG US80TR 5.1.3 does things properly with physical rear speakers and subwoofer for more immersive sound while still keeping clutter to a minimum.

It comprises the main soundbar, wireless rear satellites and a dedicated subwoofer. With an output of 580W, it can handle quiet dialogue as well as loud sound effects and soundtracks with nuance and sensitivity. The drivers bounce sound off your ceiling to create overhead effects like rain, while the dedicated center channel puts dialogue at the fore where you want it and the subwoofer delivers a punch at the low end, also serving for music and gaming.

Now a new update introduces an attempt to police generative AI content along with more complete AI declarations and more control over filtering. The aim is to make it easier to users to control what they see, but it's got a mixed response from artists so far.

When publishing content on ArtStation, users will now be asked to declare whether generative AI was used in the creation of the piece. Creators can identify their work as primarily generative AI, mixed generative AI and non-generative methods or 100% non-generative.

This relies on users being upfront, and ArtStation says that most are. However, it's recognised that there is a problem with people uploading AI-generated content without flagging it as such.

That leads to the second aspect of the latest update, which is a dedicated channel for reporting AI content. If users come across artwork they think is AI-generated but which isn't labelled, they can report it. ArtStation says that every report will help improve its AI filtering for the whole community.

Users also get more control over what they're shown: they can choose whether to hide declared generative AI content, or to allow mixed generative and non-generative workflows, and they can adjust how aggressively ArtStation filters content based on user reports of suspected generative AI.

The 'Strict' setting hides artwork when just a few reports suggest it may be AI-generated. 'Balanced' filters it out when there’s a moderate level of reports and 'Loose' removes it only when user signals are strongest. By default, the filter will be set to not show any project declared as AI-generated, tagged with 'CreatedwithAI' or tagged with generative AI tools, and suspected as AI will be set to strict.

The platform risks becoming useless if it's impossible to find interesting work because of the sheer volume of low-quality AI-generated content. Many users have welcomed the latest update as a move in the right direction, while others have raised doubts about how effective the measures it will be.

There's a strong reliance on user moderation, which will depend on users accurately identifying AI art. Some may still go undetected completely, while the 'Strict' filtering could remove work falsely identified as AI.

It's not yet clear if artists will be notified if their work is flagged or whether they will have a way to contest that decision. That could be a concern when even long established digital artists like Loish have been falsely accused of using AI over on platforms like TikTok.

The update highlights the difficult balancing act of policing AI content. ArtStation has a big focus on concept art, and some in the industry now see generative AI as an inevitable part of the pipeline in this field. The new Disney deal with Adobe will focus on developing custom AI models specifically for concept work. In that context, KitBash probably sees banning AI art from the platform as counterproductive if not impossible.

Perhaps most exciting is a collaboration with Google DeepMind to develop new AI editing capabilities for Photoshop. The companies are exploring a customised Google image model that could allow users to make near-real-time edits using text prompts, switching between variations without disrupting their workflow. In theory, that could make generative editing feel less like a frustrating cycle of prompting and waiting, and more like working with a traditional Photoshop tool. The catch? It's still exploratory, with no release date announced.

“Adobe’s mission is to empower everyone to create, and our AI innovations are helping more people express their ideas and create in ways that amplify their voice, while maintaining creative control,” said Anil Chakravarthy, president and incoming CEO of Adobe. “Our expanded partnership combines Google’s frontier AI with Adobe’s industry-leading creative and productivity tools at global scale, enabling us to deliver differentiated experiences and bring Adobe’s innovations to more people across Adobe and Google products.”

Meanwhile, Adobe is bringing its creative tools to Google's AI Mode in Search, potentially allowing users to go straight from searching for inspiration to creating something without opening a separate app.

Adobe is also expanding its integrations with Gemini and Gemini Enterprise, enabling users to enhance images, customise designs and work with PDFs through conversational prompts. For designers churning out multiple versions of campaign assets, for example, that could be a genuine time-saver.

Perhaps the most interesting aspect of the announcement, though, is what it suggests about Adobe's direction. Rather than expecting creatives to work exclusively within its own apps, it seems Adobe increasingly wants its tools accessible wherever people are already working, such as Google Workspace.

For more announcements from this week, take a look at our main Adobe Max 2026 page.

This comes alongside a host of Firefly-based AI updates, which effectively turn it into a full creative studio. You can read more about that here.

• Relight adds a virtual light to your image that you drag into position. The Properties panel lets you tweak its colour, intensity and size, and you can add up to three lights to a scene.

• Generative Restyle regenerates an image in a different style, with a slider in the Properties panel to control how far the effect goes.

• The Remove tool also gets an upgrade. It’s now better at taking out an object’s shadows and reflections, so you no longer need to select them as well.

• Quick compare: See your edits easily with a before and after

Creative Agent and a new AI workspace

Adobe is rolling out Creative Agent, an AI agent, across several Creative Cloud apps, Photoshop included. It’s pitched at handling busywork, and can also answer how-to questions, like where a tool lives or how to complete an edit. Photoshop’s AI Assistant also gets a new AI-first workspace built around natural language prompts, with a simpler interface. You can leave it at any time to get back to the full toolset.

The betas designers should look at first

• A tool that converts flattened graphics, such as a flyer or PDF, into editable layers
• A mockup creation tool
• Generative Recompose, which changes an image’s aspect ratio and repositions elements
• Camera Angles, which lets you explore alternative viewpoints of an image by interacting with the canvas
• A tool for removing atmospheric effects such as lens flare and haze
• A timelapse feature for sharing your process

That, of course, raises questions about who used to be handed those repetitive but creative jobs and whether those roles are now no longer viable. But for many creatives, particularly freelancers, handing process work to an AI agent is likely very enticing.

So let's take a closer look at the new Generative Media Tool for those looking to speed up their workflows. Instead of jumping between apps and AI services, editors can generate video, sound effects, soundscapes, and music directly in Premiere's timeline. You can select a range, describe what you need, and add the generated media as an editable clip, so you don't have to break your editing flow to get the material you need. The demo was certainly impressive.

More controversially, Adobe is also bringing supported third-party AI models into the workflow, giving editors more choice in how they generate video but also opening up Adobe's app ecosystem to external AI models such as Kling and Seedance 2.5, which have greyer areas than Adobe Firefly. More practically, a new Generative Remove tool, currently in beta, can delete unwanted objects from a video frame and shot, filling the space with realistic detail, even figuring out where related shadows of a now-omitted object should be removed.

Color Mode gives grading a new home

Color Mode is another big part of the Premiere Pro update. It's a dedicated colour-grading workspace that reshapes the interface around the job, with editable Style Presets that can apply a consistent look across an entire sequence. They're a bit like traditional Looks, only you can get inside them and adjust the individual elements.

You can also work across groups of clips, making it easier to fix inconsistencies between shots without grading each one individually. New on-screen HUDs provide feedback as you make adjustments, helping you stay focused on the footage rather than a wall of sliders.

Adobe is also expanding its AI-powered audio tools, including Enhance Audio, Separate Crosstalk, and Separate Sound Source, to clean up dialogue and separate elements in a mix.

Premiere Mobile and the MCP

Premiere Mobile is getting the Creative Agent in private beta, bringing agentic editing to phones. It can turn footage into editable first cuts, make changes using natural-language prompts and help refine videos while keeping the creator in control.

Adobe is also adding hundreds of thousands of licensed music tracks from Universal Music Group and Warner Music Group to Premiere Mobile on iPhone, with Android coming soon. There's a catch: the music is licensed for personal, non-commercial videos, while commercial projects can continue to use Adobe Stock Audio.

Finally, Premiere is included in Adobe's Local MCP integrations preview. This allows third-party agents to work with native Premiere project files locally, alongside Photoshop, Illustrator and After Effects. Importantly, sequences, bins and other project elements remain editable, meaning you can step in and make changes yourself.

As ever with these tools, the real test will be how much of the creative process editors are comfortable handing over.

The UE 5.9 Preview is Epic Games' first engine release built mainly via agentic workflows. A lot of the updates it introduces are also around AI-assisted workflows, including the use of Unreal MCP, Unreal Assistant and a new AI-driven content creation tool called Unreal Canvas.

"Invite your agents," Epic tells developers in its announcement of the Unreal Engine 5.9 preview, highlighting the ability to work with AI coding agents directly in Editor and explore AI-assisted content creation.

Unreal Canvas is described as a shared workspace where people and agents create and refine content using Unreal Engine and the other connected tools. Its canvases include text, images, videos, and mesh generators that users can control with prompts and a chosen generation service.

The companion plugin Unreal Canvas Bridge exposes Movie Render Graph rendering to Canvas as MCP tools while Unreal Canvas Spatial Annotations places annotations in the 3D scene and captures the scene for use in Canvas.

Other updates in Unreal Engine 5.9 include improvements for animation, rendering and world-building with Lumen World Cache and improved Nanite culling. You can also run projects in browser again through WebAssembly and WebGPU.

Epic stresses that this isn't a conventional preview. It was put together in about a week for the 2026 Epic AI MegaJam, which runs from 12 to 15 November. It says the aim of the release is to learn how to build the engine with multiple people directing swarms of agents working in parallel. In addition, the timing seems intended to allow the jam to serve as a testing ground.

"A handful of people directed fleets of agents to port the AI tooling from our UE6 development branch and implement Web as a target platform. Agents wrote the web platform code, starting from UE4's retired HTML5 platform, with engineers directing, reviewing, and testing the work," Epic says of how it developed the preview.

It adds: "we're already seeing benefits from this approach, and it allows us to deliver more UE5 releases with faster bug fixes and development on targeted features, all without pulling focus from UE6."

Unsurprisingly the heavy focus on AI is causing debate among developers. Those not interested in using AI agents are making a note to steer clear over concerns for the stability of the release. Epic has already clarified that some of the agent-written code still and parts of the web renderer are prototypes that will need to be cleaned up before the full release.

"I think we’ll just be staying on 5.8. LLMs are a great help, but literally everything about this preview sounds like something that crosses a boundary. Focus on optimising your code, what possessed you to add 3D model generation?" one person comments on X.

Others feels that Epic's approach is reasonable. "I'm not a fan of the asset gen stuff. But I do love more integration for agents! Including not trying to force us to use an agent IN the editor," on person argues.

Let us know what you think below. You can learn more on the website (also see our roundup of Unreal Engine tutorials and Unreal Engine courses).

Rather than prompting a model for one AI-made image, you describe what you want in plain language, and the new Agent orchestrates Adobe’s tools and a growing list of AI models to get there. You stay in the director’s chair, with the option to review the agent’s plan before it starts, pick a model yourself or leave it on auto-select, and step in whenever you like. Adobe believes this workflow will bring new projects within reach that were once out of reach and builds on Firefly and other Adobe AI tools added to Photoshop, Illustrator and Premiere Pro over the last three years.

It's a similar Agentic approach we learned about when Nvidia announced its plans for NVIDIA RTX Spark, to connect 3D apps in one AI-controlled space.

Firefly is now design everything

Firefly Design, in beta, is a new editor built from scratch, with vector, image editing, animation, 3D and print on one collaborative canvas. You can work with AI, use Adobe’s precision tools, or move between the two. The image editor gets new tools for objects, text and lighting, including Edit Object, Edit Text, Relight and Retouch, plus a gallery of dozens of Quick Actions for small jobs like removing an object or knocking out a thumbnail.

In a demo ahead of Adobe Max, I learned how it all links up. A Firefly Boards project, with fonts from Adobe Fonts’ library of around 35,000 typefaces, carried straight into the Firefly Video Editor. Adobe Stock supplied the music, ElevenLabs the text-to-speech, and the new cut detection did in seconds what Adobe’s own presenter says can take minutes, even hours. Title motion presets are in there too.

Firefly also offers models from Google, Kling, OpenAI, ByteDance and ElevenLabs alongside Adobe’s own and Topaz Labs’, with recent additions including Kling 4.0 and Dreamina Seedance 2.5. If you already pay for several of these services, seeing them under one roof will appeal. But this also raises more questions about Adobe’s famously ethical AI approach, as the foxes are now in the henhouse. I’ll be interested to see how this plays out.

Back to the details, Firefly Graph is in beta and lets you string tools and models together, save the result and run it again whenever you need to. Builder is still in private beta; you type a prompt, and it spits out a clickable prototype for a site or app, and Adobe claims it won’t wander off your design system. That’s the claim I’d test first, because most AI prototyping tools drift the moment you ask for something unusual.

Adobe is also previewing ‘always-on agents’ (sounds ominous) that keep recurring work moving while you’re away, like organising assets and checking quality. For example, a restaurant owner gives an agent their menus, URL, and brand assets. The agent drafts weekly social posts for approval, learns from feedback, and reports back on performance.

Creative Cloud gets the agent too

The agent is also spreading beyond Firefly into Adobe’s flagship desktop apps, including Illustrator, Premiere, InDesign, After Effects, and Frame.io, though the AI Assistants there are still in beta (in Photoshop only, in the Photoshop beta app). What caught my eye is that Adobe is also opening the door to rivals. The local MCP integrations, previewed now and due in private beta in early 2027, will let you point your preferred third-party agent at native Photoshop, Illustrator, Premiere and After Effects project files on your own computer. Layers, vectors, bins and sequences stay editable, so you can step in and fix things yourself. Plenty of pros already use outside agents, and Adobe clearly wants its apps to do their work where they do it.

Away from the agent, the app updates keep coming. After Effects gets an AI assistant that can make sense of an inherited project, debug expressions and generate images and video. Photoshop gets Relight, Generative Restyle and Quick Compare, and its remove tool now spots associated elements such as shadows. Lightroom adds Glow, one-tap blemish removal and a beta Open Eyes fix for dodgy blinks. Separately, Topaz Labs, which Adobe recently acquired, has introduced two new models, Wonder 4 and Image HDR, for upscaling, denoising, deblurring, sharpening and HDR. Illustrator gets refreshed content libraries and new on-canvas Blend controls, and InDesign gets a co-editing beta, Flex Layout for text frames and rewrite tools.

What's the Firefly Design catch?

The idea that the artist stays in charge is Adobe's way of easing bad news for any creative fearing how AI is affecting them, and I can believe that for filing, resizing, and checking. Whether that holds once the agent makes calls on composition, layout, and style is the big question, and the more capable these tools get, the blurrier that line will be.

Adobe still has to show that more automation really does mean more time spent making things, and that comes down to how much control creatives get once the demos are over. More importantly, with agentic AI now a trend and spreading to more and more creative software, artists and creatives who want to stay AI-free will find it harder than before.

A new wave of desktop UV printers has arrived at affordable prices and manageable sizes. And they offer exciting new creative opportunities to designers, illustrators, crafters and anyone wanting to make stuff to sell on Etsy.

A UV printer jets ink onto an object and cures it instantly with ultraviolet light, so the ink sits on top of the surface rather than soaking in. If you want, you can build it up in layers to create raised, tactile textures. Many of the best UV printers also add white ink for printing onto dark or clear materials, and a varnish channel for gloss and spot effects. Of course, there's also the old fashioned choice of the best 3D printers out there, but make no mistake – UV printers use use a far more sophisticated (and expensive) technology.

In the article below, I've listed the three best UV printers on the market today, along with two Kickstarter projects that are well worth keeping an eye on (and potentially supporting, if you believe in the project). If you're interested in more cutting edge crafting tech, check out our list of the best CNC machines.

Best UV printer overall

xTool is best known for making some of the best laser cutters, and the O1 Omni is its first UV printer, which it describes as a 4-in-1.

You can use it for UV printing onto standard surfaces. It can also make UV DTF transfer stickers for curved and awkward surfaces. And the top-of-the-range, "UV + DT" model can print directly onto cotton (DTG), and onto film for heat transfer to fabric (DTF). If you want to do all of those things, then, you're saving yourself from buying two or three separate machines.

It's also very precise. Rather than relying on a fixed overhead camera, which distorts things towards the edge of the frame, xTool's Pixel-Scan system combines laser sensing with a scanning imager to map the bed one-to-one.

On the downside, the O1 Omni prints a whole job at the height of the tallest object, so mixing a chunky item with flat coasters can leave the lower ones blurred. It's also big and heavy (41kg) so you'll need a sturdy surface to work from. On the plus side, xTool's cartridges are pretty affordable. And overall, if you want one machine that does nearly everything, this is the top dog.

Best UV printer for beginners

Eufymake is a sub-brand of global hardware company Anker, and the E1 is arguably the machine that kicked off the desktop UV boom. It became the most-funded project in Kickstarter history, made TIME's list of Best Inventions of 2025, and is now on sale generally.

If you want something that's easy to set up and use, this is it. The UV printer arrives fully assembled, the cartridge system means you don't have to mess about with syringes or bottles, and it's quick and easy to get from unboxing to a first print. The E1 stacks ink up to 5mm high, allowing 3D texture effects. You can in theory, then, reproduce brushstrokes on a canvas print, raised lettering on a business card or a leathery grain on a phone case. There's an optional rotary attachment for printing on curved surfaces like mugs and tumblers.

It's not perfect: the cartridges are more expensive than rivals', so if you're printing at scale that's worth considering. On the plus side, it's around half the size and weight of the Xtool, making it easy to fit into a home studio. In short, this is the one we'd recommend if you're new to the world of UV printing.

Best UV printer for phone cases

Procolored has been making DTF and UV printers for small print businesses for years. Its UV printer, the Procolored V4, was created primarily for one specific job: printing onto phone cases.

The 120 x 170mm A5 bed with a 15mm height limit is the perfect size for a large smartphone case laid flat. An infrared sensor measures the height of what's on the bed, and you align your designs using printed guides. The V4 is able to print onto cases made of hard plastic, soft silicone and glass. It doesn't just work on phones, either: Other small, flat items such as acrylic keyrings, badges, tiles and metal nameplates work too, as long as they fit within the print area.

Ink comes from bottles not cartridges, which helps keep running costs affordable. The V4 automatically cleans its head every 10 hours when it's left on, using about 1ml of ink a day, to keep the nozzles from clogging.

On the downside, it's slower than the newer machines and there's no varnish channel for gloss or raised textures. The software only works on Windows, and isn't the simplest to use. And for only slightly more money, you can get one of the larger all-rounders above. If you only want to print on phone cases, though, the V4 delivers exactly what it promises.

Kickstarter products to watch

These UV printers haven't actually shipped yet, so we haven't included them in our main list. They're close to doing so though, so assuming all goes well, they're well worth keeping an eye on.

Most desktop UV printers slide the bed back and forth under the printhead. On the Morpho UV Printer, the head moves on a gantry, allowing more space for taller objects: 168mm of clearance, to be specific. The bed will bear items up to 10kg, and Morpho is also promising G7-certified colour calibration for print-shop consistency. It looked pretty good to us at the consumer electronics show IFA this September: here's a short video of it making a bookmark for us.

We don't know much more at this stage, though. This is a crowdfunded product, with shipping estimated between December 2026 and January 2027. But hands-on reports have been positive, and if tall or heavy objects are your thing, it's worth watching closely.

Heygears has spent years making resin 3D printers for dental and audio manufacturing, and with the G1X, it's applying that expertise to UV printing. In its standard form it's a high-end desktop UV printer, but it's built to grow: the Full-3D Pack will add a resin system that lets the same machine build full-colour 3D objects layer by layer.

Even if you never touch the 3D side, though, it looks like a serious UV printer. It uses Epson's industrial i3200 head, which is rated for up to two years and is less prone to clogging, along with eight ink channels for a wider colour range. Again, this is a crowdfunded launch, with shipping targeted for December.

FAQs

What did those poor kids do to upset the editor of these posters? The designs each feature one of the principal cast members' faces in close-up, with their wand held aloft in front of it. Their faces are effectively cut in half by the wand, and some feel they look bizarrely stretched. Poor Ron seems to have been particularly afflicted. Not only that, but each half of the faces appears to be lit by a different light source, making the whole thing look weirdly spliced together. Suffice to say, these aren't among the best print ads we've seen.

"Unfortunately, this is the first major mistake since the beginning of the marketing campaign. It’s already receiving a lot of negative reviews on the official pages. I’m shocked they released it like this, especially after how flawless the attention to detail has been from the very beginning," one Redditor comments. Another adds, "Each poster looks like they took one half of their face from two different photos and combined them along their wands. Absolute terrible editing work, imo."

And to add insult to injury, fans have quickly knocked together an improved take on the design, focussing on just one half of each characters' face. These are arguably more striking and dramatic, and, more importantly, the faces don't look distractingly wide.

Time will tell whether the series lives up to the books, or the original film series. But one thing's for sure: these aren't among the best posters of the year, let alone all time.

This is where new talent, both in front of and behind the camera, gets to show people what they're made of; where fresh and off-the-wall ideas get their first airing in public. A typical festival is a bubbling cauldron of visual styles, storytelling techniques and new voices, which will then gradually filter into mainstream cinema, advertising, music videos and more.

Occasionally too, a festival hit will break out in its own right. Reservoir Dogs started its journey at Sundance; Parasite at Cannes. Even if you've seen neither film, you'll have felt their ripple effects on culture in general.

Wth all that heated experimentation and ambition, you'd think a film festival logo would be another canvas for creativity. But too often, these events seem to just throw together a bunch of elements (a camera, a film reel, a clapperboard) without much thought. So it was a pleasant surprise to see the logo for the Kolkata International Film Festival (KIFF) does something very different.

A single idea

Kolkata, formerly known as Calcutta, is the capital of West Bengal in India. This year the 32nd edition of its annual film festival runs from 16 to 23 December. Recently, its organisers revealed this new logo (shown above), which is the work of Sandip Ray.

Sandip is not only a filmmaker himself, he's also the son of the legendary Satyajit Ray: a director who was also a designer working on book covers, film posters and typefaces. And in this logo, Sandip has done his father proud.

Where the typical film festival logo juggles several references, this one commits to a single idea, and is all the better for it. Heavy black letters are cut into by burnt-orange triangles, each edged with sprocket holes like a strip of film.

If someone had described that to me before I'd seen it, I'd probably have thought "That sounds awful". Film strips are the clip art of cinema: as big a cliché as a popcorn box, and increasingly anachronistic to boot.

But here, the execution is everything. Crucially, Sandip hasn't used the film strips as decoration of the logo, but as its foundation. They're not just a way of adding interest to the logo design; they are the logo design. It's the difference between committing to an idea and not actually having one in the first place.

To my mind, the triangles could be seen as doubling up as play and rewind buttons… although admittedly this might just be me.

Breaking the mould

Overall, the important thing about this logo is that it doesn’t look like anything else. I've spent so many years looking at perfectly formed sans serifs that it's made me weary. And the idea that more and more people are going to use AI tools that serve up an average of this style depresses me even more.

KIFF’s logo, in contrast, is chunky, angular and a bit odd. Some people might dislike it for that very reason. But I love the fact that two colours and one bold idea are all it needs to be instantly recognisable, and that’s got to be an inspiration for logo designers everywhere.

The 8th generation iPad mini upgrades almost everything about the tiny tablet. The headline feature has to be the OLED display, which is brighter, crisper and more vibrant than any iPad mini display we've seen. Meanwhile, minor design changes have resulted in a thinner and lighter body – which is a much more meaningful change than it might sound.

• Get the 8th generation iPad mini for $699 at Amazon

I've always been a fan of the iPad mini. Sure, larger models like the Pro feature M-series chips and ProMotion displays, but sometimes there's no replacement for portability. As the famous saying goes, 'your best iPad is the one you have with you' – and being able to sling the iPad mini into a tote bag is invaluable. And the first thing I noticed when picking it up was the reduction in size. At 5.4mm thick and 255 grams, it's just ridiculously thin and light.

And the second thing I noticed was, yes, that OLED display. While it isn't quite as stunning as the new MacBook Pro's PROLED XDR display, those bright colours and inky blacks look stunning. For creatives, a bright, colour-accurate screen is a must, and now the smallest iPad delivers one in spades.

But there is a glaring omission. In this day and age, a 60hz refresh rate feels slow and stuttering on any device, let alone one that starts at £699. Apple's 120hz ProMotion display has made its way to most of its products by now, but it's surprising, given the scale of the rest of the updates (and the price increase) that the iPad mini hasn't been blessed with one. Reddit, as you might imagine, has not been kind.

But if you can put aside the lack of ProMotion, there's so much to love here. Performance-wise, that A20 Pro chip (which also resides in the iPhone 18 Pro and iPhone Duo) is incredibly speedy and efficient – I'm managing to eke out around ten hours on a single charge.

Another improvement is the sound quality. Apple has somehow managed to place the speakers underneath the display (or as Apple calls it, the 'acoustic display'), and the result is powerfully punchy audio for a tablet so small.

Overall, if you're somebody who already loves the iPad mini and has the cash to burn for an upgrade, the 8th generation model is a winner. Every improvement adds up to a meaningfully enhanced experience, particularly the OLED screen and the weight loss. Now, if Apple could just bring itself to give us a 120hz display, the 9th generation iPad mini might just be the perfect tiny tablet.

Now, it's adding another wireless in-ear headphone to its lineup. The Astrid will sit somewhere in the middle of its current offering, so they won't enter our pick of the best budget IEMs, but they offer an intriguing proposition: three driver technologies for the ability to both adapt sound to your own unique hearing profile and adjust it to how you want it to sound.

They look sharp too with an intricate silver-toned frost-like pattern over the glass fiber earbud and the charging case.

Hybrid earphones are a familiar territory for Noble Audio, which has explored driver combinations in its universal and custom in-ear monitors. Astrid brings that experience to a new wireless design.

There are three driver technologies in each ear piece: an 8.3mm dynamic driver, 6mm planar driver and a balanced armature. The first uses a PU diaphragm with an aluminium-magnesium alloy dome. This works alongside the other two technologies in a hybrid system that takes account of differences between your left and right ears as well as the way your hearing differs from other people’s.

Using Audiodo Personal Sound, a short test checks each ear and creates a personal hearing profile. This is then used to adjust playback to your ears. Meanwhile, the Audiodo Equalizer allows you to adjust bass, mids and treble as preferred: the result is sound that accounts for both how you hear and how you like your music to sound, the company says.

The new TWS has Active Noise Cancellation and a Transparency mode to let through outside sound when needed. Microphone noise and wind reduction are intended to help with calls, while Bluetooth 6.0 and two-device multipoint make it possible to keep a phone and laptop connected at the same time.

Noble cites up to 11 hours of listening per charge and up to 40 hours including the case, depending on settings, with USB-C and wireless charging. There's a choice of silicone ear tips, and the buds have IPX4 splash resistance.

Priced at $539 / £499, the Astrid sits above the $200 Osprey and the Rex5 but below the flagship Prestige Encore II in Noble's growing TWS range. You can order at nobleaudio.com and selected retailers.

The package includes a wide range of tools and materials and even Cricut's BrightPad Go lightbox. It's a big saving, but how does it compare now that Cricut has released the Maker 5, which has a slightly lower starting price? It's worth comparing the bundles available for each as both are among the best Cricut machines you can get.

• Browse Cricut deals at Amazon UK
• Browse Cricut bundle deals at Cricut

Is Cricut Maker 4 still worth buying after the release of the Maker 5?

At this price, yes, I'd still consider the Cricut Maker 4. The Maker 5 was released just two months ago, and it does have a couple of advantages, but the difference in capability and performance is very small. After all, the Maker 4 isn't yet two years old.

One advantage of the Maker 5 is that it has 25% more carriage clearance, allowing it to cut and engrave approved materials up to 3mm thick (compared to 2.4 mm on the Maker 4). It's also around a third smaller in size and nearly 2kg lighter than the Maker 4. It has an open-top design instead of the traditional flip-up lid and introduces a new snap-in Universal Pen holder and an integrated cutting tool housing, whereas the Maker 4 uses a classic Clamp A and Clamp B setup.

However, both machines have the same maximum cutting speed of up to 14.1 inches per second, and the same materials are compatible with both, so there's no issue of the newer Maker being more future proof. The smaller, lighter Maker 5 might be more convenient for smaller desks, easy stowing or travel, but if the size factor isn't a deal breaker for you, Very's Black Friday deal on the older machine may be tempting, allowing you to get started with more tools and materials.

See our full Cricut Maker 5 review and Cricut Maker 4 review for more details on each. For something more compact and travel-friendly, consider the Cricut Joy 2.