NVIDIA N1 & N1X Chips: Everything You Need to Know

If you've seen any news headlines mentioning NVIDIA's RTX Spark ecosystem, the NVIDIA N1 and N1X directly correlate as they're the hardware within that allows RTX Spark to shine. When evaluating the N1/N1X chips themselves, they showcase themselves as an intriguing piece of hardware that strays away from traditional architecture and provides excitement as to how else they'll be integrated into systems within the industry.

The NVIDIA N1 and N1X are both SoCs (system on a chip) that are infused with an ARM-based CPU and NVIDIA's Blackwell GPU architecture. Unlike traditional desktop PCs where they're "modular", which allows you to pick and choose different components (RAM, CPU, GPU), the N1 and N1X already have that all combined on a single piece of silicon.

The NVIDIA N1 and N1X chips have different capabilities, thus will be used in differing target devices:

• N1 - for thin & light laptops and mini PCs.
• N1X - for high-end creator laptops, mobile AI workstations, and compact desktop systems.

This is a similar approach to how AMD & Intel compose their products - there are specific CPUs for more casual usage (ex. Intel Ultra 3, AMD Ryzen 3), whilst there's other models offered for more extensive workloads. To see which one suits your needs better, you can check out our NVIDIA N1 vs. N1X: What's the Difference article.

The main thing that separates the NVIDIA N1 and N1X from the rest of the hardware on the market is its balance between three resources: compute power, memory access, and power/thermal budget. The N1 and N1X provides a breath of fresh air from the traditional splits between those resources as they're all unified on a single chip, unlocking a whole new level of how the resources talk to each other. Benefits of the balanced architecture include:

• Single, massive unified memory architecture (UMA) so the GPU can instantly access 100GB+ of memory for AI models & 3D Rendering without having to wait for data transfer over PCIe. This is an absolute game-changer for local AI and LLMs, especially for a laptop/compact system.
• High power efficiency with its 45W - 80W power budget that is constantly shifted by the millisecond to whichever engine needs it the most (ex. 80% to GPU during gaming, or 100% to CPU during compilation workload)
• Considering the N1/N1X chips have their CPU/GPU on the same chip, it can be cooled by a single cooler/heatsink rather than needing separate coolers for the CPU and GPU. This allows for full N1/N1X systems to be built in compact form factors as it doesn't need as much physical space for excessive cooling.
• The N1/N1X chips have thousands of Blackwell CUDA cores integrated on them, offering performance up to par with a laptop RTX 5070 GPU. Having the Blackwell CUDA cores directly on the SoC reduces footprint by over 40%, further reinforcing the ability to create thin laptops and compact PCs. Battery life is also lengthened, providing the potential for all-day battery life whilst having gaming laptop performance.

Most current SoCs (like Apple M4 or Qualcomm Snapdragon X) feature integrated GPUs designed primarily for efficiency and moderate graphics tasks. Standard x86 laptops usually require a separate, power-hungry discrete graphics card to handle heavy gaming or 3D rendering.

The NVIDIA N1/N1X's integrated Blackwell CUDA cores can enable it to perform for up to a RTX 5070 laptop, which is unheard of for an integrated graphics chip.