The N1 and N1X chips that were floating around in a bevy of reports and rumors are officially called RTX Spark, as NVIDIA has lifted the veil for its first consumer-grade chips that will power a truckload of laptops later this year for the Windows on ARM platform. Thanks to the company’s collaboration with MediaTek, the RTX Spark houses a ton of CPU cores and RTX Blackwell cores to deliver incredible compute, graphics, gaming, and AI performance. Here are all the details.
The RTX Spark is also designed for tackling AI workloads, being able to run 120 billion parameter models with 1 million context and more
At the NVIDIA GTC Taipei keynote, founder and CEO Jensen Huang says that RTX Spark is the result of the GPU manufacturer’s 30-year innovation, with all of its technologies combined to bring together its first Windows on ARM platform.
“The PC is being reinvented. For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask — and the PC does the work. RTX Spark brings everything NVIDIA has built — CUDA, RTX, our AI platform — into a single superchip. Local agents. Frontier models. Creative workflows. RTX games. All on a laptop. This is the new PC. The personal AI computer.”
With up to a 20-core Grace CPU that’s designed and built with the help of MediaTek, NVIDIA promises better single-core and multi-core performance, though evidence of these gains isn’t really there since the RTX Spark fails to compete with Apple’s three-year-old M3 Max in Geekbench 6’s single-core and multi-core test results. Then again, it’ll be interesting to see what scores commercial units gain.
Alongside the 20-core Grace CPU, NVIDIA has crammed in its Blackwell RTX GPU that sports 6,144 CUDA cores and delivers 1 petaFLOP of AI performance. While various predictions claim that the GPU will be on par with a laptop RTX 5070, we’ll have to see the performance capabilities ourselves to make any judgments. The unified memory architecture brings up to 128GB of LPDDR5X RAM with a bandwidth of 600GB/s. with five lanes of PCIe Gen 5.
This memory architecture is enabled by NVIDIA’s NVLink C2C, which is similar to Apple’s UltraFusion. The company also states that the RTX Spark is ideal for creative professionals and gamers, being able to render 90GB 3D scenes, along with editing 12K video, while also playing AAA titles at 100FPS at the 1440p resolution, with ray tracing and DLSS support.
With its compute and graphics performance, not to mention increased memory count and bandwidth, the RTX Spark is also designed for running AI models, with NVIDIA stating that a full AI stack with 120 billion-parameter models with 1 million context can be run. Also, thanks to being fabricated on TSMC’s 3nm process and housing 70 billion transistors, the RTX Spark is expected to be highly efficient and will power Max-Q laptops with an all-day battery life.
About the author: Omar Sohail is a reporter and analyst for Wccftech's mobile section, specializing in the technology and business of the mobile industry. His expertise lies in the intricate hardware supply chain, covering developments in semiconductor manufacturing, chip lithography, and camera sensor technology.
Follow Wccftech on Google to get more of our news coverage in your feeds.
