Microsoft Surface Laptop Ultra Review: With This Slick Flagship, the RTX Spark Era Is Just Getting Started - PCMag UK

The Surface Laptop Ultra is the flagship of the new Nvidia-powered RTX Spark laptops, bringing the power of local AI and agentic tools to Windows. It is built as much for local AI processing and development work as for basic productivity.

The base model employs Nvidia's RTX Spark N1X processor with 18 Grace processor cores and a Blackwell GPU with 5,120 graphics cores. It is paired with 24GB of unified memory and a 512GB SSD for storage.

The model reviewed here has a higher-tier version of the N1X chip, with 20 Grace processor cores and 6,144 Blackwell graphics cores. It is complemented by 64GB of unified memory and 1TB of storage, selling for $4,299.99.

A top-spec model with 128GB of unified memory is available for $5,899.99. Microsoft allows the 64GB model to be configured with up to 2TB of storage, but the 128GB model with only 1TB.

Unified Memory Architecture

The most important feature of the Spark SoC is its unified memory shared between the CPU and GPU. This single pool of memory is shareable between the processor and graphics, enabling support for local AI models with billions of parameters and large context windows.

This design is similar to what Apple has done with Apple Silicon, combining an Arm-based processor and powerful discrete-GPU-grade graphics hardware on a single chip die. It is also similar to what Nvidia has done with its AI developer desktop, the Nvidia DGX Spark.

RTX Spark is essentially server-grade hardware scaled down into a laptop. The Surface Ultra's contribution is that Nvidia has partnered closely with Microsoft to ensure that all of this AI capability supports Windows.

Design and Build

The laptop features top-notch build quality. It is a bold statement about the future of Windows software. Early adopters should take a close look, though everyone else should consider the Apple MacBook Pro 16-Inch (2026, M5 Max) for sheer AI moxie.

All should keep a close eye on how the RTX Spark platform develops. It signals a huge shift in the evolution of personal computing.