Nvidia Targets PC Chip Incumbents With RTX Spark Windows Push

At Computex, Nvidia said RTX Spark will put a Blackwell-class GPU and Grace CPU into Windows laptops this fall, pushing the company from add-in graphics into the processor slot held by Intel, AMD and Qualcomm. The company calls RTX Spark “the most efficient PC chip ever built,” a claim The Verge noted Nvidia made without publishing benchmark charts.

Beyond the processor, Nvidia is packaging its graphics software, on-device AI agents and developer tools into the platform, an attempt to control more of the Windows AI stack than any single chip supplier has held before.

“For forty years, you launched apps. Click. Type,” Chief Executive Jensen Huang said in Nvidia’s announcement. “Local agents. Frontier models. Creative workflows. RTX games. All on a laptop. This is the new PC. The personal AI computer.”

The GB10 design inside Windows

RTX Spark tracks the GB10 silicon in Nvidia’s DGX Spark developer box: a 20-core Grace CPU, a Blackwell RTX GPU with 6,144 CUDA cores and as much as 128GB of unified memory. DGX Spark ran as a Linux machine for developers, while RTX Spark moves the same CPU-GPU package into Windows 11.

Nvidia says RTX Spark can run 120-billion-parameter models with as much as a one-million-token context, render 90GB 3D scenes and edit 12K 4:2:2 video on the top configuration’s 128GB of unified memory. Lower configurations start at 16GB, PCMag reported.

Power consumption is the first number reviewers will test. Nvidia says the chip runs light tasks at “low, low single-digit” wattage and climbs to 80 watts under heavy loads. Aevermann told The Verge buyers should “expect it to be much better than anything you’ve seen before on RTX laptops” and that “you won’t need a charger” for light work. The 80-watt ceiling under load is the figure reviewers will check against the “all-day battery life” pitch.

Microsoft’s Windows on Arm bet

RTX Spark is an Arm-based PC platform, which means legacy x86 Windows software still depends on Microsoft’s Prism emulator unless developers ship native Arm versions. Microsoft is adding Windows security primitives, and Nvidia is bringing OpenShell to Windows so local agents can run under user control.

Software compatibility is the standing question for any Arm-based Windows PC. Nvidia and Microsoft list more than 100 software providers and game partners, including Adobe, Blackmagic Design, Blender, CapCut, ComfyUI, OTOY, Riot Games, Krafton, NetEase, Remedy and Xbox. Aevermann said “all the top games will run on RTX Spark and provide a great experience.” Riot’s League of Legends and Valorant and Krafton’s PUBG matter because anti-cheat systems have been the gating issue for Windows on Arm gaming; Fortnite had already arrived on Windows on Arm.

On agents, Vincent Koc, chief architect at the OpenClaw Foundation, said OpenShell and Microsoft’s security primitives would give users “a fully integrated stack for private, personal agents running on device.” Nous Research Chief Executive Dillon Rolnick put it plainly: “RTX Spark and NVIDIA OpenShell give Hermes users a powerful and secure environment for agents to run and work alongside you. You realize you’re buying a full-fledged assistant, not a typical laptop.”

The Implicator argued in January that AI PCs had stronger silicon than software, citing reports that Microsoft had lowered some AI-software sales targets while PC makers kept adding neural hardware. RTX Spark answers that gap with far more local memory and GPU software, though the consumer software case is still unproven.

More than 40 planned PCs

Nvidia says more than 30 laptops and more than 10 compact desktops are planned, with first systems due this fall from Asus, Dell, HP, Lenovo, Microsoft Surface and MSI, and Acer and Gigabyte models to follow. PCMag reported an initial six-premium-laptop wave, while TechRadar listed eight named RTX Spark laptops already announced.

Microsoft is fronting the launch with its own hardware. Andrew Hill, a Microsoft Surface product leader, told The Verge the Surface Laptop Ultra is “the most powerful thing we’ve ever made.” Aevermann added that “RTX Spark is going to be a family of products that are going to attack a lot of different price points” and said, “The overall market opportunity that we see is quite large.”

In Monday trading after the announcement, Intel fell 4%, AMD dropped 3% and Qualcomm slid 6% while Nvidia rose 4%, according to Business Insider. Chris Versace of TheStreet Pro called it “a move [that] will strike at the heart of the PC business at Intel and AMD” and make Qualcomm’s PC push “far more challenging.” The Implicator covered Nvidia’s Intel x86-RTX collaboration as a system strategy that gave Nvidia PC optionality without committing to Intel foundry risk.

Forrester analyst Charlie Dai told the BBC the announcement marked a “paradigm shift” from “component supplier” to “architecture owner in the PC market,” adding it would “directly challenge Intel, AMD, and Qualcomm and raise competitive pressure on performance, efficiency, and AI integration.” CCS Insight analyst Ian Fogg offered the counterweight, saying the new class is “likely to come with a significant price tag” and will target buyers seeking “workstation-class performance.”

Premium pricing and the local-agent test

Nvidia and Microsoft did not disclose laptop prices, and Aevermann said the first wave is aimed at premium segments. That matters because, as PCMag and HotHardware noted, high-memory configurations arrive during an ongoing DRAM and NAND shortage tied to AI demand.

Nvidia’s narrower case rests on the agents themselves: software that runs privately on a device needs local compute, memory and security, and RTX Spark supplies more of each than a typical thin-and-light laptop. Qualcomm Chief Executive Cristiano Amon, speaking around Computex, made the same point from the rival camp. “Two years ago we talked about how AI will change the human computer interface, and as a consequence will change the architecture of all of our personal computing devices. And that is starting to become a reality in 2026.” He called it “the year of agents” and said, “All of these devices today, they have been built for actions initiated by the user, not by the agents.”

The first hard evidence comes this fall, when reviewers can measure price, unplugged battery life and Prism emulation performance, and when buyers learn whether the promised 30-plus laptops ship beyond the initial six premium models PCMag described. Nvidia and Microsoft have not said when laptop prices will be announced.

Samsung, SK Hynix and Micron chase AI memory as buyers wait to 2027

The memory shortage is likely to last until around 2027 as Samsung, SK Hynix and Micron add capacity too slowly to match demand. The three suppliers control about 90% of DRAM, and current expansion pl

The Implicator

China’s AI Suppliers Cannot Build Fast Enough for DeepSeek

On May 12, Xiang Xiaotian gave Bloomberg the sentence Chinese AI investors did not want to hear. The Shanghai Chengzhou Investment Management director said component bottlenecks were "unlikely to be r

The Implicator

The Silicon Squeeze: How AI's Memory Appetite Is Cannibalizing the Tech Industry

The uncomfortable part for anyone planning to buy a laptop, upgrade a server, or manufacture smartphones in 2026 isn't that memory chips are expensive. It's that the shortage driving those prices has

The Implicator

Marcus Schuler

San Francisco

Editor-in-Chief and founder of Implicator.ai. Former ARD correspondent and senior broadcast journalist with 10+ years covering tech. Writes daily briefings on policy and market developments. Based in San Francisco. E-mail: editor@implicator.ai