Nvidia RTX Spark brings agentic AI to consumer PCs, but hardware costs raise questions

Nvidia's RTX Spark is bringing more agentic AI to consumer PCs, but it feels like a cruel joke in this hardware economy

Abhinav pivoted from a career in banking to pursue his first love in writing. Even while working full-time, he continued contributing as an editor-at-large, a role he has held for more than 7 years. A lifelong tech enthusiast who has built three gaming and productivity powerhouse PCs since 2018, his passion for technology keeps him closely following the semiconductor industry, from NVIDIA and AMD to ARM. His MSc dissertation explored how artificial intelligence will reshape the future of work, reflecting his curiosity about the wider social impact of emerging technologies. The RTX Spark has taken the computing discourse by storm, perhaps for all the right reasons. Taken purely as a feat of chip engineering, it's difficult not to admire. It's not every day that you see a Blackwell-class chip with over 6,000 CUDA cores with a 20-core ARM CPU capable of packing up to 128GB of unified memory in a laptop that's thin enough to disappear into a backpack. It's the kind of marvel that subtly reminds you why Nvidia has the kind of influence over the industry that it does. The broader ambition behind this powerhouse is also evident, and Satya Nadella put it rather succinctly when he stated that Microsoft plans to use the platform to bring, "unmetered intelligence to every home and every desk with Windows." The million-dollar question of the hour, however, is, at what price. Unsurprisingly enough, neither tech giant seems eager to answer that just yet. Nvidia is selling a solution to a problem it helped create RTX Spark arrives at the worst possible moment for consumers RTX Spark seems to have entered the conversation at a rather unusual moment for the consumer hardware industry. While Nvidia and Microsoft speak at length about bringing AI to every desk, consumers have spent the better part of the last nine months dealing with the consequences of a component market that has made all kinds of consumer hardware much more difficult to afford. DRAM prices hit al all-time high and surged by up to 300% globally in March 2026, and this development cannot be seen as a coincidence. The same demand for AI accelerators that made Nvidia the most valuable company on the planet also happens to be one of the major forces behind memory shortages that has been squeezing consumers since June 2025. Key manufacturers from across the industry shifted capacity towards High-Bandwidth Memory (HBM) production for data center hardware, and the downstream effects inevitably reached DDR5 (and now, DDR4) memory, GPUs, and all the consumer devices that use them. This all makes RTX Spark a peculiar proposition, especially given the timing. The company that, at least in part, contributed to make personal computing more expensive, now looks to be offering a $3,000 local inference machine. If that isn't enough to rouse your skepticism, well, there's a little more to it, because it raises another important question. Local AI has a hardware accessibility problem, and the answer to it isn't RTX Spark Nvidia built the most powerful laptop chip to solve a problem no one has Posts By Abhinav Raj Does the average consumer want more AI on their PCs? Or is it just a strategic directive from Nvidia? If you look through the history of consumer technology, the economics are simple enough to follow. Companies build products responding to a consumer demand, and successful categories emerge when the supply follows the demand. For the most part, that relationship has remained reasonably consistent for decades. The AI era, however, appears to have introduced a subtle departure from that economic model. Increasingly, supply constraints and product priorities seem to be shaped less by consumer demand and more by where manufacturers see the greatest opportunity. To understand this, you'd need to look no further than Nvidia's financial results. Of its $68.1 billion quarterly revenue, more than 91% now comes from its data center business. The company that many consumers have for long associated with GeForce graphics cards has, in effect, become something very different, and its apparent strategic pivot is the strongest evidence. Nvidia is no longer primarily in the consumer technology business. It is clear that it's a data center company with a strategic objective to bring its own AI ambitions downstream. Now, whether consumers have been asking for that future is a different question altogether. A pressing and rather difficult question one might want to ask is whether such a company can continue acting in good faith for the remaining 9% of its market. CUDA has no real competition, and Nvidia knows it When there's no viable alternative, the price sets itself The RTX Spark isn't the first major entrant in the category. AMD's Ryzen AI Max 400, codenamed Strix Halo, demonstrated that large unified memory systems for local, on-device AI were viable months before RTX Spark entered the conversation. AMD's premium configurations also cost considerably less than what RTX Spark devices are expected to be priced at. In all ways, it's the more "accessible" proposition. Deals Find Deals on High-Performance Laptops & Workstations Score discounts and limited-time offers on laptops, desktops, monitors, memory upgrades, SSDs, and docking accessories - smart savings for creators, developers, and pros upgrading their work setup. Explore computers & work setup deals to compare prices and claim savings. Deals Explore Computers & Work Setup Deals And yet, most of the conversation revolves strictly around Nvidia, and all of it has to do with CUDA's head start against ROCm. The many years of software investment have created an ecosystem that competitors struggle to match. For many AI developers, there simply isn't an accessible, practical alternative, especially when software ecosystem tends to favor one platform over another. Which brings us back to economics. In a competitive market, viable alternatives keep prices honest, and when one company effectively has a hegemony over the software ecosystem, that mechanism stops working. While RTX Spark may very well represent the future Nvidia envisions for personal AI, it feels like consumers are being asked to fund that vision at a time when the market has never felt more prohibitive to entry. "Every home and every desk" was the cruel joke all along To be fair, none of this diminishes what RTX Spark represents when it comes to engineering. If anything, it makes the proposition more frustrating. A part of the problem is that a platform that's being positioned as a mass-market product (at least, eventually) arrives at a time when personal computing finds itself amid a rapidly exacerbating pricing crisis that Nvidia's own trajectory architected. Whether other configurations of the platform make adoption easier, is something that's worth watching. But for now, the RTX Spark is a demo for the vast majority of the market.

Microsoft built its most powerful Surface ever on Nvidia's new chip, and we saw every RTX Spark competitor coming this fall

Nvidia, alongisde Microsoft, used Computex 2026 to announce RTX Spark, a chip that puts a 20-core Arm CPU and a Blackwell GPU with 6,144 CUDA cores, the same count as the RTX 5070, on a single package with up to 128GB of unified memory. The CPU side was co-designed with MediaTek, and Nvidia rates it at 1 petaflop of AI compute, enough to run 120B-parameter models locally with up to a million tokens of context. It's the consumer descendant of the GB10 Grace Blackwell Superchip that powers the DGX Spark and machines like the Lenovo ThinkStation PGX, except this one runs Windows. In a private showcase hosted by Nvidia, we were able to see every RTX Spark-based machine that was announced. Unfortunately, the story told by every OEM is fairly homogeneous. It arrives in fall 2026, there's no pricing, and spec sheets range from complete to nearly empty, so what we have is what was on display and what each company was willing to say. Still, I've used the GB10 extensively for AI workloads and even gaming, and the 5070 comparison isn't marketing fluff. I've run Cyberpunk 2077 at 1440p on the GB10 through FEX and Proton, two translation layers stacked on top of each other, and it still held playable frame rates at maximum settings. These Windows machines only carry one of those layers, since Prism does on Windows what FEX does on Linux and Proton has no job to do at all, so the overhead only goes down from here. Nvidia says more than 30 laptops and 10 desktops are in the pipeline, with Acer and Gigabyte machines to come as well. Here are all of the machines we know are coming, and what we got to see. Disclaimer: Asus paid for my travel and lodging in order to attend Computex 2026 in Taipei. The company had no input into the contents of this article. Microsoft Surface Laptop Ultra A Surface with Nvidia silicon inside Arguably the star of the show, Microsoft calls the Surface Laptop Ultra the most powerful device it has ever made. Oh, and it's the first Surface with Nvidia silicon since the Tegra-powered Surface RT, the machine that gave Windows on Arm its terrible first impression back in 2012. The company is aiming it at what it calls "world makers," which translates to creators, developers, and AI researchers, and selling it as an alternative to the MacBook Pro with up to 128GB of unified memory and complete CUDA support on board. After years of Snapdragon and Intel Surfaces, Nvidia is the next big frontier. The display is the standout spec. It's a 15-inch mini-LED PixelSense Ultra touchscreen at 2,880 x 1,920 and 262 pixels per inch, with up to 2,000 nits of peak HDR brightness, which makes it the brightest panel Microsoft has ever put in a Surface. Mini-LED is an unusual choice in a lineup where most rivals went OLED, but it gets to that brightness ceiling without the burn-in anxiety of an OLED. The machine comes in under 18mm thick and under 2kg, with the largest haptic touchpad on any Surface to date, in Platinum and Nightfall finishes. Port selection is surprisingly generous for a Surface, too: HDMI, USB-C, USB-A, a full-size SD card slot, and a headphone jack. Microsoft's line is that every port you actually use is on the device, and for once that's not far off. What Microsoft didn't share is storage options, battery capacity, or a price, and given the hardware inside, I wouldn't expect it to be cheap. Microsoft also built its flagship around Nvidia rather than Qualcomm this time, right as Qualcomm's Windows on Arm exclusivity comes to an end. It arrives later in 2026 with everything else. Asus ProArt P16 and P14 The most complete spec sheets of the show While most OEMs kept their cards close, Asus published actual details for both of its RTX Spark laptops. The ProArt P16 (H7607) is the flagship and the ProArt P14 (H7407) is its smaller sibling, and both are CNC-machined aluminum in Nano Black and Neo White finishes. Asus says the new P16 is 13% thinner and 16% lighter than the model it replaces. The displays are what you'd expect from the ProArt name. The P16 gets a 4K 120Hz Lumina Pro OLED with variable refresh and G-Sync, while the P14 steps down to a 3K 120Hz panel, and both hit 1,600 nits of peak HDR brightness with 100% DCI-P3 coverage, Pantone validation, and Delta E under 1. Asus also fitted an anti-reflective coating it claims cuts reflections by 65%, which matters quite a lot on a glossy OLED when it comes to color grading and general display legibility in the sun. The P16 is 12.9mm thick at 1.77kg with a 99.9Wh battery and up to 2TB of storage, while the P14 is 13.9mm at 1.48kg with a 90Wh battery and up to 1TB. Connectivity is properly creator-focused, with three USB-C ports, a USB-A port, HDMI 2.1, a full-size SD card reader, a headphone jack, and Wi-Fi 7. Asus appears to be pitching the P16 directly against the 16-inch MacBook Pro. Memory goes up to 128GB of unified LPDDR5X-9400, which is the most important spec for running large local models on this thing. Asus also confirmed an RTX Spark mini PC alongside the laptops, though it shared far less about that one and had nothing to show for it. Like everything else here, there's no price, and availability is fall 2026 in select regions, but of all the machines I saw, these two are the closest to finished products, if only because Asus alone would commit to a complete spec sheet. Asus later announced a third machine, the ProArt Mini PC, and gave it the most complete spec sheet of any RTX Spark desktop at the show. It's a 150 x 150 x 51mm box whose chassis is nearly identical to the Ascent GX10, the GB10 machine Asus already sells, with up to 128GB of unified memory, expandable M.2 NVMe storage, four USB-C ports, HDMI, and 10Gb Ethernet. Asus is pitching it against the Mac Studio, and it's coming in fall 2026 with no price either. Dell XPS 16 Creator Edition The XPS name is back for this one Dell retired the XPS brand in early 2025, replaced it with the Dell Premium naming scheme, and then brought it back at CES 2026, saying it was getting back to its roots. The XPS 16 Creator Edition is the first XPS to carry Nvidia's full software and hardware stack, so the revived badge's first big swing is an Arm machine with no Intel inside. That's not likely to be a choice that Dell made lightly. The confirmed hardware is a 16-inch tandem OLED display with True Black HDR 600 certification, up to 128GB of unified memory, and a built-in SD card reader and HDMI for I/O. Dell's pitch is 4K 4:2:2 timeline playback, faster exports, and 3D work that would normally choke a thin laptop, which is also the kind of workload this chip was designed for. Beyond that, Dell shared very little: no storage configurations, no battery capacity, and no price ahead of its fall 2026 launch. Dell also brought a second RTX Spark machine that almost nobody is talking about, a small-form-factor desktop with no name attached. The company was very clear that it was a concept and that the design isn't final, so what Dell was really showing was a public commitment to building an RTX Spark mini PC. The unit we saw is a small, square box with perforated mesh sides for airflow, two USB-C ports and a full-size SD card slot on the front. There are no specs or a launch window to go with it, but an SD slot on the front of a desktop tells you the creator pitch isn't limited to the laptops. The commitment makes sense, though. Dell already sells the Pro Max with GB10, a roughly $5,000 mini workstation on the previous chip that runs Nvidia's Linux-based DGX OS, so it knows there's an audience for a small box with 128GB of unified memory. An RTX Spark version swaps that Linux workstation positioning for native Windows, and that's a much bigger potential market than AI developers alone. HP OmniBook Ultra 16 and OmniBook X 14 Two laptops, one disclosed spec HP brought two RTX Spark laptops to Computex, the OmniBook Ultra 16 and the OmniBook X 14, and is calling them the world's thinnest RTX Spark machines. The X 14 measures 13.53mm at the rear and the Ultra 16 comes in at 15.73mm. I'd take the "world's thinnest" framing with a grain of salt, though, since Asus quotes the ProArt P16 at 12.9mm. The measurements likely aren't taken the same way, and we won't know who's right until someone shows up with a tape measure to verify. Thickness is also more or less the only thing HP disclosed. Both machines support the platform's headline capabilities, including 12K 4:2:2 video editing and local 120B-parameter models in the 128GB configuration, but HP withheld display specs, storage, battery, ports, weight, and pricing entirely. There was hardware on the floor, but the details are being saved for closer to launch, with both OmniBooks expected later this year. HP is also building an RTX Spark mini PC of its own, confirmed at the show as a compact desktop on the same platform, with specs fully under wraps and a launch later in 2026. It doesn't have a name yet, and it's easy to mix up with the OmniDesk Mini that HP announced at the same event, which is an Intel Core Ultra machine with no RTX Spark inside. If you see people talking about a "HP mini PC" at Computex, check which one they're about. As with Dell, HP isn't starting from zero here. It already ships the ZGX Nano G1n AI Station, its GB10 machine, in a chassis that measures just 150mm square, so it has built a tiny computer around this silicon's predecessor once already. Lenovo Yoga Pro 9n The 'n' is for Nvidia Lenovo's naming scheme is pretty self explanatory: the Yoga Pro 9i is the Intel model, and the new Yoga Pro 9n is the Nvidia one. It's Lenovo's first RTX Spark laptop, and it follows the same design language as the 9i, with a 15-inch display, an aluminum chassis, top-firing speakers, and a backlit keyboard. The trackpad is unusually large and supports pen input for drawing, which is a strange feature on paper until you remember who this machine is for. The confirmed details read like a creator checklist: an HDMI port, an SD card slot, and all-day battery life that Lenovo won't put a number on. The top configuration gets the full 20-core chip with 128GB of unified memory. What Lenovo hasn't published is the panel's resolution or refresh rate, and I'd caution against assuming it inherits the 9i's display, because nothing official says it does. Lenovo is no stranger to this hardware family either, as the ThinkStation PGX it already sells uses the GB10. I've used mine to serve 80B-parameter models at up to 40 tokens per second and fine-tune my own 7B model, all from the same 128GB unified memory pool this laptop will ship with. That experience is the most concrete preview of the Yoga Pro 9n that exists right now. The Yoga Pro 9n is that capability moving from a Linux workstation into a consumer Windows laptop, and that's the entirety of the proposition behind the RTX Spark. Lenovo hasn't given it an official date or price, but fall 2026 alongside everything else is the safe assumption. MSI Prestige N16 Flip AI+ The only convertible in the lineup The Prestige N16 Flip AI+ is the only 2-in-1 in the RTX Spark lineup of announced devices so far, a 16-inch convertible that rotates between laptop, tablet, tent, and presentation modes. As with Lenovo, the naming appears to be self-explanatory, with the "N" denoting Nvidia. MSI is the launch partner that looks like it took the biggest risk on form factor, and I honestly respect it. Subscribe for in-depth RTX Spark coverage and analysis Join the newsletter to get hands-on breakdowns, clear spec-tracking, and plain-language analysis of RTX Spark hardware and Windows-on-Arm implications. Curated device comparisons and buyer-focused guidance will help you evaluate the new platform. Get Updates By subscribing, you agree to receive newsletter and marketing emails, and accept our Terms of Use and Privacy Policy. You can unsubscribe anytime. The display is a 16-inch UHD+ tandem OLED that stacks two emissive layers to push past 1,000 nits, with touch support and Delta E under 1 for color work. MSI fits a 99.9Wh battery, plus a quad-speaker array and a customizable Action Touchpad. The MSI Nano Pen, a stylus that stows away under the chassis, is also a great little detail for creators. In other words, a digital artist gets a 16-inch pen-input tablet with a color-accurate OLED and a 5070-class GPU behind it, which can run image generation and other local AI workloads on-device instead of requiring the cloud. Storage, weight, and pricing are all undisclosed, with availability in the second half of 2026. Of everything I saw, this is the machine I'm most curious about, mostly because pen-input convertibles almost never get this much GPU. MSI also brought a second RTX Spark machine, and it's the odd one out of the show. The EdgeMesa N AI+ is a mini PC, and where every other machine here is pitched at creators, MSI is aiming this one at edge AI deployments across healthcare, retail, finance, robotics, and smart city infrastructure. The specs lean that way too, with an emphasis on a 10Gb Ethernet port for moving datasets and talking to inference servers, plus one HDMI and three USB-C ports driving up to four displays. Storage, memory SKUs, and pricing are all undisclosed. It's a true platform already Every machine here ships in fall 2026, and not a single one has a price. Estimates put the top-spec systems at a minimum of $3,000 and up, and given that the GB10 machines cost even more than that, I wouldn't bet on these being affordable. The hardware itself doesn't worry me, because I've spent months with the GB10 and unified memory at this capacity changes what you can run locally in a way no consumer GPU can match. Nvidia is also treating this as a platform rather than an experiment. It has committed to three generations of Spark silicon, with the current Grace Blackwell chip followed by a Vera Rubin generation on LPDDR6 and a Rosa Feynman generation after that, and Acer and Gigabyte are lined up behind the launch partners. Whatever happens this fall, Nvidia isn't planning to walk away from Windows PCs after one round. There's a pattern in the branding, too. Nvidia spent a significant amount of time during its keynote and even in its demos talking up 1440p gaming at over 100fps, yet not one of these machines wears a gaming badge. Dell went with XPS rather than Alienware, Asus chose ProArt over ROG, MSI picked Prestige instead of one of its gaming lines, Lenovo used Yoga rather than Legion, and HP went OmniBook, not Omen. Every OEM looked at a 5070-class GPU and decided gaming wasn't the pitch, filing it under creator lines for the laptops and edge AI for MSI's mini PC. Either they don't trust Arm gaming to be ready on day one, or they know exactly who's going to pay $3,000 for the first wave. The open question is Windows on Arm. These machines live or die on Prism emulation, which Microsoft says it tuned specifically for this chip, along with driver maturity and whether anti-cheat and legacy software cooperate. Frame counters and frametime overlays were nowhere to be seen during gaming demos at the show, but the reason for that is that the software is changing a lot right now, and frequently enough that any performance figures would mean very little this far from launch. To be fair, the chip held up its end on Linux through two layers of translation, so if Microsoft holds up its end too, this is the biggest shake-up Windows laptops have had since Apple Silicon forced the industry to take Arm seriously.

NVIDIA's ARM chipset and early Wi-Fi 8 routers: How-To Geek's favorite tech of Computex 2026

Unlike CES or MWC, which are fairly consumer-focused trade shows, the annual Computex Taipei event is traditionally a "B2B" affair. Boring! Still, we made a point to attend, and it's a good thing we did. Some of the greatest tech innovations of 2026 (or maybe even the decade) made their debut during Computex, including a few surprises that may completely change how the average person interacts with computers at a fundamental level. Hello, RTX Spark! Nice to meet you, Arc G3! This is an exciting time for all classes of PC enthusiasts, from Windows on ARM evangelists to hairbrained homelabbers. Of course, the AI geeks get plenty of love, too. So, let's stop yapping and get to the goods. NVIDIA RTX Spark (n1x) A turning point for Windows on ARM (and PCs as a whole, hopefully) NVIDIA absolutely stole the show at Computex 2026, which is really no surprise given the company's status in the AI boom and its near-biological interdependence with Taiwanese manufacturers. Even if the brand had attended this event without a peep of progress or products, the regular appearances of its CEO and co-founder Jensen Huang were a pure highlight for a majority of the attendees, particularly the locals. His lightly-guarded catwalks through the show floor (and comical enthusiasm for dashing off Sharpie signatures on other brands' products) were met with messianic joy, rhythmic chants of "Jensen! Jensen!," and the glow of stock trading apps. We learned a lot about the venue's layout during these appearances, as we were forced to explore uncharted paths just to attend our pre-planned meetings and walkthroughs. But Jensen and company knew better than to arrive empty-handed. Instead, they revealed the RTX Spark (n1x) "superchip," an all-new PC platform that promises to bring high-end agentic local AI to all classes of PC, including consumer devices. It offers a petaflop of AI compute power with up to 128GB of unified memory, enough to power AI agents with up to 120 billion parameters. The closest comparison, at least in NVIDIA's words, is a pro RTX 5070 AI workstation. Though, in our eyes, the most interesting part of RTX Spark is its architecture. This isn't a clone of AMD's x86 Ryzen AI Ultra platform. Instead, it's ARM. Ignoring all of the AI stuff, RTX Spark creates a unique opportunity for Microsoft to finally push Windows on ARM into excellence, as the operating system is no longer bound by Qualcomm's de-facto monopoly on ARM desktop development. It's easy to compare this moment to the introduction of Apple Silicon, which occurred nearly 6 years ago. Yet there are still questions as to whether RTX Spark will begin its life as a prosumer platform (as Apple Silicon did). Its emphasis on AI development and technical similarities to DGX Spark AI, a 2025 dev workstation that currently sells for $4,700, suggest a more enterprise-focused approach. But even if the average enthusiast is priced out of the equation, RTX Spark could give Windows on ARM (and PCs as a whole) the kick in the pants we've craved for the last decade. Microsoft certainly seems to think so. Microsoft Surface Laptop Ultra The flagship NVIDIA Spark machine Microsoft calls the Surface Laptop Ultra "the most powerful thing we've ever made." And while it isn't the most visually exciting device to grace the Surface lineup, the specs are, to Microsoft's credit, outrageous. Surface Laptop Ultra is the first RTX Spark device and potentially the first slam-dunk for Windows on ARM, boasting a 20-core CPU, up to 128GB of unified memory, and a maximum 1 petaflop of AI performance for local LLMs and AI/ML development. The closest comparison is a desktop PC with an RTX 5070 GPU (as we mentioned earlier), although there are no real-world benchmarks at this time and the ARM architecture means that 1:1 comparisons should be taken with a grain of salt. Externally, Surface Laptop Ultra utilizes a gorgeous 15-inch mini-LED PixelSense Ultra touchscreen display with a 3:2 aspect ratio and 2880 x 1920 resolution. Microsoft promises 2000 nits of peak HDR brightness, which is excellent for any laptop. The chassis and keyboard are very MacBook Pro-esque; you get an aluminum body, lattice keys, a haptic trackpad, two USB-C ports (presumably Thunderbolt 4 or 5 though not confirmed), one USB-A port, an HDMI jack, an SD card slot, and 3.5mm audio output. The Surface Laptop Ultra launch event was fairly GPU-focused, with laptops chugging through demanding games like Pragmata, but Microsoft also spent some time showing off the device's repairability. You can remove the Laptop Ultra's backplate, and internal components are tagged with QR stickers for easy lookup. Plus, the SSD and battery are readily accessible. Mild repairability has slowly become a key feature in Surface products, though it's surprising to see this trend continue in a high-end ARM laptop, at least for those of us who've spent any amount of time with a modern MacBook. Pricing and availability for the Surface Laptop Ultra are unknown, but a late-2026 launch date is expected. Microsoft's use of unified memory should alleviate some cost concerns, although RTX Spark makes pricing a bit of a guessing game. It's obviously safe to assume that Surface Laptop Ultra will exceed $1,000, but we could be looking at a product that goes beyond the $2,000 or $3,000 threshold. Intel Arc G3 and G3 Extreme Intel's trying to eat AMD's lunch AMD managed to dominate the handheld PC gaming market with surprisingly little effort. Through this dominance, AMD has also entrenched itself as the premiere processing platform for Linux gaming as a whole. Frankly, it's always seemed like a missed opportunity for the Intel Arc platform, which could have secured an early niche in the handheld market if it hadn't spent so much time banging its head against desktop PCs like a bumblebee that can't figure out how to get through a car window (although we'd be remiss if we failed to mention how Intel Arc excels in homelabbing scenarios). Evidently, Intel finally caught on to the scent of opportunity. It introduced the Arc G3 and G3 Extreme, a set of powerful and power-efficient mobile processors that can rival AMD's Ryzen Z-series chipsets. Leveraging the Panther Lake architecture, these CPUs are essentially handheld-focused versions of the Intel Core Ultra Series 3, a chipset that quietly smuggled high-end iGPU gaming performance and an extended battery life to laptops a few years ago. The Arc G3 Extreme is especially notable, as it borrows the B390 iGPU from the Core Ultra X7 and X9 line (Intel's standard Arc G3 uses a B370 graphics unit). The first Arc G3-powered handhelds will arrive in June 2026, and the first confirmed device is the OneXPlayer 3 (which will launch through an Indiegogo campaign in the coming weeks). Acer plans to stick an Arc G3 in a future Predator Atlas 8 handheld, while MSI has confirmed that the Claw 8 EX AI+ will eventually get an Arc variant. These Arc-based handhelds should be more powerful than AMD Ryzen Z-series machines, at least on paper. The real question is whether game developers (and Valve) will prioritize Arc compatibility at a software level. If games aren't optimized for this platform, then they won't be able to take advantage of the Arc G3 or G3 Extreme's raw power. We'll have a better grasp of the situation once large brands like Acer and MSI put their hat in the ring. Dell XPS 13 The closest thing that Windows users can get to a MacBook Neo The Dell XPS family has always served as a sort-of Windows counterpoint to Apple's MacBook lineup. They're thin and light laptops with a ton of processing power, high-quality screens, excellent keyboards, and wonderful (though occasionally excessive) trackpads. In light of the $600 MacBook Neo, it's no surprise that Dell is pushing to sell a "mid-range premium" laptop of its own; a refreshed XPS 13. Priced at $700 MSRP in its lowest config (or just $600 with a student discount), the base model XPS 13 comes with a minimum Intel Core 5 320 processor, 8GB of RAM, and 512GB of storage. Dell promises 17 hours of battery life, a backlit keyboard (something the MacBook Neo lacks), and a 13.4-inch 1600p display with a 120Hz refresh rate. The most notable thing here, aside from the price, is the build -- XPS 13 is slimmer and lighter than the MacBook Neo, clocking in at 0.5 inches thick and just 2.2 pounds. But like the MacBook Neo, there are some notable shortcomings in the XPS 13's build quality. Dell opted for a mechanical touchpad instead of a haptic trackpad (just as Apple did), and it settled for a basic lattice keyboard layout instead of going hog-wild on the awesome, edge-to-edge keyboard layout that's featured in the XPS 14 and 16. There's also some questions about processing power, as the Intel Wildcat Lake platform hasn't been properly benchmarked yet and 8GB of RAM is questionable for a Windows device in 2026. But the point still stands; Apple decided to stop neglecting the $600 to $1,000 market, and brands like Dell are following suit. If you're a customer shopping in this range, and especially if you're a student who can cash in on discounts or other promotions, the new XPS 13 appears to be your best non-Apple option. It's certainly a welcome addition as computer components and PCs continue to march toward the extreme end of unfriendly pricing. AMD Ryzen 7 5800X3D A 10-year-old processor revived Reinventing the wheel every year is a fool's errand. At least, that's the mindset that we see in most corners of the tech industry. Automakers are happy to stick with a single platform for a decade, Apple likes to reuse its laptop chassis, and legacy software keeps the entire world from falling into chaos. But desktop CPUs tend to disappear from the market after just a couple of years. So, needless to say, AMD's decision to re-introduce the Ryzen 7 5800X3D is kind of unusual. The chipmaker proudly announced a "10th Anniversary Edition" of the CPU at Computex 2026 alongside a revived Ryzen 7 7700X3D. As far as upgrades are concerned, there are none, aside from the inclusion of a Carbice Ice Pad that may provide extended thermal dissipation in some PC builds. This move is clearly a response to rising component costs in the AI era (and the stubborn stagflationary economy, which would have contributed to demand for cheaper PCs in the consumer market regardless of the AI boom). While this CPU isn't a great option for high-end modern PC builds, it's more than good enough for a decent desktop rig, and it should provide a solid upgrade opportunity for anyone currently rocking an older machine with an AMD 400 or 500-series board. Priced at $350 with a launch date set for June 25, the 10th Anniversary Ryzen 7 5800X3D is slightly more expensive than an aftermarket Ryzen 7 5800X3D. The extra cost is a decent trade-off for a fresh warranty, and the CPU's re-introduction should create some downward pressure on aftermarket prices -- used CPUs could get cheaper. ASUS ROG Rapture GT-BN98 Pro A big honkin' Wi-Fi 8 router "Is it a large-scale model, or is it the real thing?" That's the question we kept hearing from passers-by as ASUS introduced us to the ROG Rapture GT-BN98 Pro, a colossal Wi-Fi 8 (802.11bn) router that looks like something you'd find crawling in the mercurial waters of an alien beach. ASUS is getting a huge head-start on the soon-to-be Wi-Fi 8 craze with this monster, boasting two-times mid-range throughput, two-times wider IoT coverage, and 34% lower latency when compared to an unspecified product (we love vague product comparisons). The router's also packed with seven Ethernet ports, specifically a 10Gbps high-speed gaming port, a 2.5Gbps and 10Gbps WAN/LAN jack, three 2.5Gbps ports, and a single 1Gbps connection. We asked why this router is so big, and the answer's pretty predictable; a bunch of 2.5GbE and 10GbE lanes, combined Wi-Fi 8 capabilities, require some decent processing power and heat dissipation. Thermals seem to be a big part of the conversation here, as ASUS took the time to tell us that the new Wi-Fi 8 router will offer better heat dissipation than the Wi-Fi 6E Rapture GT-AXE1600 -- hopefully that means the thermals have actually improved, as it could also mean that the new router simply spits out more heat. ASUS plans to launch the ROG Rapture GT-BN98 Pro at some point, presumably in late 2026 or early 2027. Though, notably, the Wi-Fi 8 standard probably won't be finalized until 2028. That means the ROG Rapture GT-BN98 Pro may lack some functionality that becomes standard to Wi-Fi 8 in the next year or so. Future Wi-Fi 8 routers will be a lot smaller than the ROG Rapture GT-BN98 Pro, as the high-speed connections that are responsible for its size are only necessary in enthusiast and enterprise markets.

Best laptops coming in 2026 after Computex

This story is part of our coverage of Computex, the world's biggest computing conference. Updated less than 0 just now ago Every Computex promises the next big thing, but only a handful of laptops actually feel worthy of the hype. After spending time exploring the show floor and seeing these devices up close, one thing became abundantly clear: 2026 isn't just about faster processors. It's about smarter laptops, better portability, and AI features that are finally starting to feel useful instead of being another sticker on the palm rest. A big part of that shift is NVIDIA's new RTX Spark platform, which made its way into several premium creator machines this year. Rather than diving into its technical details yet again, let's focus on the laptops themselves, because each manufacturer has taken the platform in a very different direction. Microsoft Surface Laptop Ultra If there was one laptop that made me pause for a second look, it was Microsoft's Surface Laptop Ultra. This isn't just another Surface with refreshed internals. It's Microsoft's vision of what an AI-first premium laptop should look like. Powered by RTX Spark hardware and configurable with up to 128GB of unified memory, the machine is clearly aimed at creators, developers, and professionals juggling demanding AI workloads. Even so, its standout feature is its stunning 2,000-nit mini-LED display, which remains perfectly legible even under harsh lighting conditions, immediately positioning it as a serious outdoor competitor to Apple's latest MacBooks. Pair that with Microsoft's refined industrial design and deep Windows integration, and you have a laptop that feels purpose-built for the next generation of AI-powered workflows. Dell XPS 16 Creator Edition Dell's XPS lineup has always balanced premium aesthetics with serious performance, and the new XPS 16 Creator Edition continues that tradition beautifully. Beyond the RTX Spark platform, Dell has equipped it with a dual-layer Tandem OLED display, bringing significantly higher brightness, improved panel longevity, and gorgeous colour reproduction while maintaining the minimalist design language that made previous XPS machines so iconic. It feels purpose-built for creators who spend hours editing photos and videos but still want something refined enough to carry into a boardroom, effectively blurring the line between workstation and luxury ultrabook. HP OmniBook Ultra X 14 and OmniBook Ultra 16 HP's OmniBook Ultra X 14 feels tailor-made for professionals who are constantly on the move but don't want to sacrifice capability in the name of portability. Powered by NVIDIA's RTX Spark platform, the 14-inch machine combines local AI acceleration with a sleek, lightweight chassis that should easily slip into a backpack without becoming a burden. It strikes a nice balance between performance and practicality, making it an appealing option for developers, creators, and business users who need a premium laptop that can comfortably handle demanding workloads while away from a desk. The larger OmniBook Ultra 16, meanwhile, takes the same philosophy and scales it up for users who need more screen real estate and extra horsepower for creative work. With its expansive display, RTX Spark-powered AI capabilities, and focus on sustained performance, it feels better suited for video editors, 3D artists, and professionals juggling multiple applications at once. What impressed me most, though, is that HP hasn't simply built two powerful machines. It has built two laptops that seem genuinely designed around different kinds of users, ensuring there's an Ultra that fits almost every workflow. ASUS ProArt P16 and ProArt P14 Few brands understand creators quite like ASUS, and the latest ProArt family proves exactly why. The ASUS ProArt P16 is clearly built for professionals who spend their days editing 4K footage, rendering 3D scenes, or juggling multiple creative applications at once. Alongside NVIDIA's RTX Spark platform, it packs a gorgeous OLED display with factory colour calibration, giving photographers, filmmakers, and designers the confidence that what they're seeing on screen is exactly what they'll deliver to clients. ASUS has also integrated its suite of ProArt software tools, alongside the physical DialPad baked right into the trackpad, making the laptop feel like a complete creative ecosystem rather than just another powerful machine. The ASUS ProArt P14 takes much of that same DNA and squeezes it into a far more travel-friendly package. Despite its smaller footprint, it still offers RTX Spark acceleration, a stunning OLED panel, and the same creator-first philosophy that defines the larger model. It's the kind of laptop that makes sense for photographers editing on location, content creators working from cafes, or freelancers constantly hopping between shoots and client meetings. More importantly, both ProArt machines reinforce ASUS' long-term commitment to building dedicated creator hardware instead of simply repurposing gaming laptops with a different paint job. MSI Prestige N16 Flip AI+ MSI decided to take a slightly different route with the Prestige N16 Flip AI+, and honestly, I think that's a good thing. Its convertible form factor already makes it stand out, but pairing it with the company's Nano Pen transforms it into something much closer to a digital sketchbook than a traditional laptop. Designers, illustrators, and anyone who spends more time drawing than typing will likely appreciate that flexibility, while RTX Spark quietly handles AI-assisted creative tasks in the background. The result is a machine that feels equally comfortable sketching concepts in tablet mode or editing an entire production timeline from a cafe. Dell XPS 13 Not everyone needs workstation-class hardware, and that's where Dell's refreshed XPS 13 shines. It feels like one of the clearest responses to Apple's aggressively priced MacBook Neo strategy by combining premium construction, impressive battery life, dedicated AI capabilities, and a remarkably portable chassis into a package aimed at everyday users rather than professionals. Dell has also refined the iconic XPS design further, with its ultra-slim profile, minimalist CNC-machined aluminium body, and almost bezel-less display, making it one of the most elegant Windows laptops announced this year. Up front sits a sharp 2.5K display that strikes an excellent balance between crisp visuals and power efficiency, making everything from spreadsheets to Netflix binges look fantastic. Dell isn't trying to build the fastest laptop here. Instead, it's focusing on delivering a polished, premium experience that people will genuinely enjoy carrying around every single day. Acer Swift Air 14 Acer's Swift Air 14 is arguably one of the most exciting thin-and-light laptops announced at Computex, not because it chases flagship specifications, but because of the value it promises to deliver. Positioned as an affordable premium ultraportable with a starting price of just $699, it brings features that would have easily belonged in a much more expensive machine only a couple of years ago. Like the Dell XPS 13, the Swift Air 14 embraces the idea that portability shouldn't come at the cost of capability. It pairs a sleek, lightweight design with dedicated AI acceleration and efficient hardware, making it an excellent fit for students, young professionals, and everyday users looking for a modern Windows laptop without stretching their budget. More importantly, its aggressive pricing shows that AI PCs are finally becoming accessible to the mainstream, and not just reserved for premium flagships. ASUS ROG Strix SCAR 18 Every list needs at least one unapologetically overpowered gaming machine, and this year's honour goes to the ASUS ROG Strix SCAR 18. Configured with up to an NVIDIA GeForce RTX 5090 Laptop GPU and capable of sustaining a staggering 320W combined system power budget, the SCAR 18 is built for enthusiasts who refuse to compromise. Pair that with aggressive cooling and ASUS' signature ROG styling, and you end up with a laptop that doesn't merely replace a gaming desktop. It actively challenges one. The display deserves just as much attention as the hardware inside. ASUS has equipped the SCAR 18 with what it calls the world's first 18-inch 4K 240Hz Mini LED laptop panel, complete with ROG Nebula ELMB technology, over 2,000 local dimming zones, and up to 1,600 nits of peak HDR brightness, delivering a rare combination of ultra-high resolution and esports-grade refresh rates. For gamers who want a machine that can handle cinematic AAA titles one moment and competitive shooters the next, the SCAR 18 feels like one of the most ambitious laptops announced at Computex 2026. Acer Aspire Go 15 Perhaps the most interesting entry on this list isn't the most expensive one. The Acer Aspire Go 15 becomes one of the first laptops to showcase Qualcomm's Snapdragon C platform, signaling a push toward affordable AI-powered Windows machines. Rather than targeting enthusiasts, it focuses on students, first-time buyers, and budget-conscious users who want excellent battery life and modern AI features without spending flagship money. With Acer targeting a price point of under $400, it could become one of the most important launches of the year by bringing local AI capabilities to a market segment that's often overlooked. The best laptops aren't just getting faster. They're getting smarter. If Computex 2026 proved anything, it's that the best laptops are no longer the ones with the biggest spec sheets. They're the ones that actually understand who they're built for. Whether it's an AI-powered creator machine, an affordable everyday ultraportable, or a desktop-replacing gaming beast, the industry finally seems to be trading meaningless numbers for meaningful experiences. And honestly, that's a glow-up both our workflows and our wallets can appreciate.

The Biggest PC hardware trends from Computex 2026

This story is part of our coverage of Computex, the world's biggest computing conference. Updated less than 3 minutes ago Every Computex has its headline-grabbing announcements. There's always a faster processor, a shinier graphics card, or a laptop that's somehow even thinner than last year's model. But after spending several days wandering the halls of Computex 2026, talking to engineers, trying products, and occasionally getting lost between exhibition booths, I came away with a very different takeaway. That said, this year's show wasn't really about individual products. Rather, it was about the direction the industry is heading. Instead of chasing flashy specifications for the sake of marketing slides, manufacturers finally seem focused on solving real problems. The MacBook Neo effect is impossible to ignore Whether companies admit it or not, Apple's MacBook Neo shook up the PC industry by proving that a thin, silent, and premium-looking laptop doesn't have to come with an eye-watering price tag. Its blend of impressive performance, excellent battery life, and aggressive pricing has clearly forced Windows manufacturers to rethink their priorities. Products like the refreshed Dell XPS 13 (2026) and Acer Swift 14 AI are no longer trying to outmuscle bulky gaming notebooks. Instead, they're focused on delivering premium build quality, all-day battery life, cooler thermals, dedicated NPUs, and hardware-level AI acceleration in sleek, highly portable designs while also pushing to make those experiences more accessible to mainstream buyers rather than luxury-only purchases. Perhaps the clearest example of this shift is Intel's Project Firefly, a design initiative centered around building ultra-lightweight AI PCs that maximize day-to-day efficiency instead of brute-force horsepower. The conversation has evolved from asking how much raw performance manufacturers can cram into a chassis to how much performance users actually need before portability, battery life, near-silent acoustics, and affordability become the bigger selling points. As someone who reviews laptops regularly, I genuinely welcome this change. Raw performance still matters, but carrying a power brick the size of a paperback novel everywhere I go doesn't. AI is finally becoming useful If there was one buzzword impossible to escape at Computex, it was AI. Thankfully, this year it felt less like marketing jargon and more like something that can genuinely improve everyday workflows. The best example was the NVIDIA RTX Spark platform, driven by the flagship NVIDIA N1X superchip. Built around a 20-core Grace CPU co-developed with MediaTek, this ARM-powered platform is designed to execute demanding AI workloads locally instead of constantly reaching for cloud servers. Watching Adobe Photoshop intelligently generate assets from simple visual instructions using directional arrows, or seeing Premiere Pro perform near-instant scene edit detection and one-click asset rotoscoping, demonstrated what happens when software developers and hardware manufacturers actually work together. Companies aren't just shipping raw NPUs anymore; they're partnering with application developers to integrate AI directly into creative workflows, productivity tools, and editing software where it can quietly remove repetitive tasks instead of getting in the way. Even discussions around agentic AI workloads reflected that shift. Rather than treating AI as another simple chatbot box, manufacturers increasingly see it as an always-available assistant capable of handling routine work autonomously in the background while users focus on more meaningful tasks. ARM is taking the fight straight to x86 For years, ARM-powered Windows laptops have felt like promising experiments searching for the right audience. Computex 2026 made me feel like that phase might finally be ending. Qualcomm continued pushing its vision forward with the Qualcomm Snapdragon C platform, aggressively targeting affordable AI PCs that combine impressive battery life with dedicated local AI capabilities. At the other end of the spectrum sat NVIDIA RTX Spark, proving that ARM systems can also deliver serious enthusiast-grade creative performance while comfortably handling gaming and AI workloads on the same platform. Perhaps the biggest surprise wasn't just how natural these systems felt during hands-on demos, but how NVIDIA has completely changed the conversation around ARM itself. Rather than positioning it as a low-power alternative to x86, RTX Spark presents ARM as the foundation for a scalable AI ecosystem. Built around a 20-core Grace CPU paired with Blackwell RTX graphics and up to 128GB of unified memory, the same architectural philosophy extends beyond laptops into NVIDIA's broader Grace Blackwell portfolio, including powerful DGX systems designed for AI development and enterprise workloads. It sends a clear message that efficient ARM designs no longer have to stop at thin-and-light notebooks. Technologies like DLSS 4.5 Ray Reconstruction, Microsoft's Prism compatibility layer, and NVIDIA's work with developers on anti-cheat support also made gaming feel surprisingly polished during the demos I experienced. Will ARM replace x86 overnight? Almost certainly not. But for the first time, it genuinely feels less like a compromise and more like a platform capable of scaling from ultraportable laptops all the way to AI workstations, making it a far more formidable competitor than ever before. Current generation hardware isn't going anywhere One of the most unexpected trends from Computex wasn't brand-new hardware. It was companies refusing to abandon existing platforms. AMD reaffirmed its commitment to the AM5 desktop socket through at least 2029, giving enthusiasts a much longer, consumer-friendly upgrade path than many expected. To combat rising component costs, the company also expanded its mainstream graphics lineup with the new AMD Radeon RX 9070 GRE, reinforcing the idea that existing architectures still have meaningful room to evolve instead of immediately becoming obsolete. That philosophy extended well beyond processors and graphics cards. Cooling specialist Noctua showcased the NT-CP1 carbon nanotube thermal pad, a maintenance-free, solid-state alternative to traditional thermal paste that promises consistent long-term performance without drying out over time. The company also previewed its first all-in-one liquid coolers, built around Asetek's mature platform but enhanced with Noctua's own acoustic engineering to reduce pump noise and vibrations, highlighting how refinement is becoming just as important as raw performance. GPU manufacturers echoed the same sentiment with increasingly optimized thermal designs and factory tuning aimed at extracting more efficiency from familiar architectures. For consumers, that's excellent news. The message coming out of Computex wasn't "throw everything away and start over." It was "make what you already own even better." Considering how expensive PC components have become lately, that might just be the most consumer-friendly trend of the entire show. Gaming monitors are growing up Gaming monitors spent years competing in a never-ending numbers race. More hertz. More brightness. More HDR certifications. This year felt refreshingly different. Displays like the Alienware AW3926QW introduced RGB-stripe Tandem OLED technology on a gorgeous 39-inch 5K curved panel, allowing users to switch between pristine 5K clarity at 165Hz for creative work and a lightning-fast 1080p mode at 330Hz, significantly improving brightness and text subpixel clarity along the way. Meanwhile, the MSI MPG OLED 322URDX36 triple-mode monitor demonstrated just how far refresh rates have come, offering a fifth-generation QD-OLED panel with Penta Tandem technology that can scale dynamically across multiple resolution profiles depending on the game genre. Even esports displays continued evolution into absolute precision instruments. The ASUS ROG Strix OLED XG259QWPG Ace pushed tournament refresh rates to a blistering 540Hz while retaining TrueBlack Glossy Tandem WOLED image quality. On the hybrid front, the Acer Nitro XV345CKR P showcased how a 5K WUHD resolution, a 1,344-zone Mini-LED backlight, and Dynamic Frequency and Resolution (DFR) modes can seamlessly serve professional creators who also happen to be hardcore gamers. It feels like the era of buying separate displays for work and play may finally be starting to fade. Handheld gaming PCs are finally growing up Just a couple of years ago, handheld gaming PCs still felt like ambitious experiments trying to squeeze desktop hardware into portable shells. Computex 2026 made them feel much more mature. The biggest story was undoubtedly the Intel Arc G3 Extreme processor, a graphics-first platform based on the Panther Lake architecture and manufactured using the cutting-edge Intel 18A process node. Packing a 14-core CPU configuration alongside 12 next-gen Xe3 Celestial graphics cores, complete with hardware ray tracing and Intel XeSS 3 with Multi Frame Generation, Intel finally looks ready to challenge AMD's long-standing dominance in the premium handheld market. Devices like the MSI Claw 8 EX AI+, the Acer Predator Atlas 8 (PA08-I51), and the ASUS ROG Xbox Ally X20 (20th Anniversary Edition) all reinforced the same message. Better ergonomics, smarter cooling systems (like Acer's 89-blade metal AeroBlade technology), massive 80Wh batteries, refined full-screen Windows 11 Xbox Mode software experiences, and highly efficient silicon mean manufacturers are no longer trying to prove handheld PCs are viable. They're competing to build the absolute best one. As someone who's spent years using devices like the Steam Deck and the original ROG Ally, that's perhaps the most exciting trend of them all. The next big thing is... Practicality? Looking back at Computex 2026, I don't think I'll remember the show for a single processor, graphics card, or laptop. I'll remember it for the industry's changing mindset. For the first time in a while, it felt like companies were exploring how to make PCs better to live with. AI is quietly taking over repetitive tasks instead of demanding attention, ARM is growing into a serious challenger rather than an interesting experiment, gaming monitors are becoming versatile enough to replace multiple displays, and handhelds are finally maturing into products I'd happily recommend without a long list of caveats. If these trends continue, the next generation of PCs won't just be faster. They'll be quieter, more efficient, more affordable to upgrade, and a whole lot smarter about how they use their performance.

Nvidia's RTX Spark is built for AI and creators, not gamers -- and that's the whole point

Ever since a kid, Shaheer has always been intrigued by tech and how its components work, and has always geeked out over new PC component releases. His enthusiasm for tech, gaming, and composing eventually led him to become a writer, a career path he hadn't quite planned or thought of, but it proved to be a natural fit. Nevertheless, Shaheer has been writing for more than 4 years, having penned more than 2000 articles now across multiple accredited publications. Shaheer mostly covers PC components, including in-depth coverage of each component, but his speciality subsumes within hardware of all sorts. Nvidia claims to be reinventing PCs for the first time in 40 years with the RTX Spark -- a super-chip crammed with everything Team Green has learned in over three decades of its existence. Developed in close collaboration with Microsoft and MediaTek, the RTX Spark is a high-end super-chip targeting creators, AI agents, and, lastly, gamers. On paper, the performance sounds great: you're getting 1440p gaming at 100 FPS on the highest graphics settings -- that is, until you realize that Nvidia never really built the chip for gaming. Nvidia doesn't make TVs, but it made the best thing you can plug into one The Nvidia SHIELD TV Pro is perhaps the most powerful streaming device you can buy Posts 21 By Brandon Miniman The RTX Spark is a data center chip in a laptop shell Gaming was never the initial design goal In case you've missed it, here's a brief rundown of the RTX Spark: the SoC combines an RTX Blackwell GPU featuring 6144 CUDA Cores and up to 1 petaFLOPs of FP4 AI performance, alongside a 20-Core GRACE CPU made in collaboration with MediaTek, and a unified 128GB of LPDDR5X memory. This all sounds pretty alluring until you realize that the RTX Spark is a super-chip built on the GB10 architecture, a 2-year-old SoC that Nvidia designed for AI enthusiasts. The GB10 wasn't a consumer-grade processor; it ran a Linux-only AI compute environment designed for AI experiments and related tasks. The RTX Spark brings the same specifications but is intended for more general-purpose uses such as content creation, rendering, AI workloads, and gaming. The specifications required for content creation also make it a decent gaming machine, but gaming was never the design goal. The RTX Spark doesn't make for a good gaming machine The RTX Spark suffers from a lower memory bandwidth The RTX Spark features 128GB of LPDDR5X memory, which may sound like a lot, but compared to GDDR7 memory, it is slower and has higher latency. The RTX Spark's specifications put it up against the RTX 5070 mobile, but the super-chip has a memory bandwidth of 273.2 GB/s, much lower than the RTX 5070 mobile's 384.0 GB/s. The rest of the SoC's specifications focus more on AI-leaning tasks, as evidenced by the higher counts of Tensor Core TMUs and ROPs. Memory bandwidth will be the biggest bottleneck for the RTX Spark, as it also has to share it among the CPU cores. Now, I don't mean in any way that the RTX Spark sounds bad for gaming; it won't be, but that's the least you would expect from a super-chip of this caliber, and this is why I said the RTX Spark for gaming feels like an afterthought. Windows on ARM still isn't there for the average consumer ARM compatibility is still catching up In recent times, we've seen a major shift of manufacturers moving towards ARM. It all began with Apple ditching Intel CPUs for its own silicon, and it's the very same M chips that have compelled manufacturers to shift to ARM for mainstream computers, including Microsoft. The RTX Spark is also an ARM chip, as mentioned earlier, but the thing is that the Windows experience on this architecture is still lacking and nowhere near ready for mainstream adoption. Performance isn't an issue on ARM Windows, but compatibility becomes the biggest shortcoming. The thing is that not all applications are natively supported on ARM and run through the emulation layer -- you're likely to encounter problems and face a performance hit. The RTX Spark will be Nvidia's first step into Windows on ARM, and on the first try, expect more of a groundwork-laying effort than one focused on perfecting it. Pricing may be the biggest hurdle for the RTX Spark There are better and cheaper options for gaming The AI-led memory shortage has vastly affected the consumer electronics market. The RTX Spark features a whopping 128GB LPDDR5 memory, and in times like these, that doesn't come cheap. Morgan Stanley has estimated that the RTX Spark N1X chip for laptops will start at $2,900 or more, and the lower-end N1 chip will start at $1,800. The RTX Spark will ship with variants, and the key phrase is "up to 128GB", not 128GB across all chips. Expect the price to creep higher for laptops with higher memory, likely crossing the $4,000 mark. Get smarter about RTX Spark with our newsletter Subscribe to our newsletter for clear, no-nonsense coverage of the RTX Spark and Windows-on-ARM developments, including technical breakdowns, price vs. performance comparisons, and practical buying guidance to help you judge whether this chip fits your needs. Get Updates By subscribing, you agree to receive newsletter and marketing emails, and accept our Terms of Use and Privacy Policy. You can unsubscribe anytime. If your goal is gaming, the RTX Spark might not be a good investment; for the same price, you can get much better options. The Lenovo Legion 7i Gen 10, featuring an RTX 5070 and an Intel Core Ultra 9 275HX, goes for around $2,400 and can run almost any AAA game at high settings at 1440p, outperforming the RTX Spark based on the current specifications. Lenovo Legion 7i Gen 10 CPU Intel Ultra 9 275HX GPU RTX 5070 Mobile RAM 32GB DDR5 Display (Size, Resolution) 2560x1600 The Lenovo Legion 7i Gen 10 is a premium, lightweight gaming laptop that delivers top-tier performance in a sleek design, featuring the RTX 5070, Intel Ultra 9 275HX, and a 2.6k display, making it capable of running nearly every AAA title at high settings. Brand Lenovo $2400 at Amazon (US) Expand Collapse Nvidia's claims of reinventing PC might have been too much Nvidia claimed that it is "reinventing PCs," which is too bold a claim, especially considering that the RTX Spark is based on a 2-year-old chip. Now, I do appreciate that they're making an effort to advance Windows for ARM -- especially that they're developing a native anti-cheat for ARM, which was the biggest drawback to switching from X86. From what Nvidia has to show so far, it won't change much for gamers. However, considering they succeed in this market, Nvidia might advance into the integrated graphics market for gaming, and that's a space they can dominate, but even then, reinvention seems like a far stretch.

I saw Nvidia RTX Spark in action, and Windows PCs may finally have their Apple Silicon Moment

Windows laptops have been chasing the gold that Apple struck years ago with its M-series chips. Back in 2020, Apple revealed a new lineup of MacBooks powered by silicon that seemed generations ahead in efficiency, integration, and even performance. It was a system that was built around the hardware instead of merely running on top of it. At Computex 2026, Nvidia RTX Spark looked like one of the most convincing responses that Windows has had in years. I got a tour of Nvidia's showcase at the show, and Spark was easily one of the most interesting things I saw. During the demo, the company gave me a close-up look at the RTX Spark as it ran video edits, local AI agents, games, and other hefty workflows. And in each of these tests, it became apparent that this isn't just another laptop chip reveal. More than an AI PC badge The AI PC label has been thrown around so much that it has started sounding like a dead meme on the internet. Every "next-gen" laptop seems to have an NPU to push AI, which honestly looks to be more of a marketing push than something many users would actually use. Spark, on the other hand, is doing things a little differently. It is building around the kind of workloads that are already central to Nvidia's ecosystem. So CUDA, RTX, DLSS, G-Sync, local AI processing, and GPU-based creative work are now front and center. Recommended Videos Looking at the spec sheet, RTX Spark looks packed. It features a 20-core Grace CPU with a Blackwell RTX GPU, 6,144 CUDA cores, up to 128GB of unified memory, and up to 1 petaflop of AI performance. All of this translates to a single processor that is flexible for nearly everything you throw its way. The Apple Silicon comparison is easy When Apple moved the Mac to its own silicon, performance and efficiency bumps were huge, but shifting control back to Apple was even bigger. Now, Apple had the CPU, GPU, memory architecture, operating system, and even software stack working together. The ecosystem was tighter than ever before, and it no longer had to rely on Intel. Overnight, the MacBook started a new PC race after transitioning to M1. They were not simply more powerful, but also practical with the sweet power efficiency. This is the part Windows laptops struggled to match. It had plenty of raw horsepower, sure, and in some ways, far more flexibility than the Mac. Still, it came across as fragmented. One company makes the chip, another builds the laptop, while the third handles the software experience. Even with amazing chips, an unoptimized platform could ruin your experience with a laptop. That's not all, performance can vary wildly depending on thermal designs, drivers, battery profiles, and, most importantly, vendor tuning. Nvidia can't control Windows like Apple with its macOS; however, it has worked closely with Microsoft to make a true Windows flagship of its own, the Microsoft Surface Laptop Ultra. This gives Nvidia a chance to create a high-performance platform with a more stable foundation. Nvidia's RTX Spark Demo Right from the start, Nvidia made sure everyone understood the effort it put into this "superchip". The showcase covered just about every use you can think of. Nvidia showed the chip handling creative workflows, local AI agents, running the heaviest AAA games, and much more. The video editing demo was the clearest example. Nvidia showed Spark working with generative AI tools inside a video workflow, allowing editors to make changes using natural-language prompts instead of digging through timelines and menus. The idea was to automate some of the repetitive tasks that make the process so time-consuming. For someone who spends a lot of time cutting footage, organizing clips, and making tiny revisions, this is one of the few AI demos at Computex that looks pretty useful. A big part of the magic comes down to the hardware. Spark's combination of Blackwell graphics, large unified memory pools, and local AI processing means these tools can run directly on the machine instead of constantly relying on cloud services. Local AI agents were another major focus. Nvidia showed systems capable of running sophisticated AI models on-device, handling tasks that would normally require remote servers. If AI is going to become a tool for everyday workflows, local performance like this could be just as important as raw benchmark numbers. Gaming was unsurprisingly the other pillar. Nvidia still brings an ecosystem advantage that Apple has struggled to replicate, with RTX technologies, DLSS, Reflex, and deep support across the PC gaming market. Both native titles and games running on emulation showed smooth performance. Actual performance figures are still unknown, but the reveal seemed promising. Windows still has to meet Nvidia halfway Spark is promising, but this is not a guaranteed Apple Silicon moment just because the hardware looks good. Apple's definitive advantage came from owning the full stack. For Spark to really work, laptop makers need to build great machines around it. Microsoft needs to make Windows feel ready for local AI workflows. Developers need to support the platform in ways that go beyond benchmark-friendly demos. Pricing and battery life will matter too, especially if Spark laptops end up exclusively sitting in the same premium territory as high-end MacBooks. But after seeing it in action at Computex, I get why Nvidia is going big here. It gives Windows laptops a shot at something they have needed for a long time