Nvidia N1x SoC benchmarks trail Apple M3 Max

🔑 Enhanced Key Takeaways

•The Nvidia N1x SoC is fabricated on TSMC's 3nm process and integrates a Blackwell-architecture GPU with 6,144 CUDA cores, which is comparable to a desktop RTX 5070.
•The N1x is specifically designed for Windows on ARM laptops and is a mobile adaptation of Nvidia's GB10 Superchip, previously seen in the DGX Spark mini-PC.
•Nvidia's entry into the Windows on ARM market with the N1x aims to directly compete with Qualcomm's Snapdragon X Elite, leveraging Nvidia's extensive CUDA software ecosystem for AI and graphics acceleration on Windows laptops.
•The N1x features a 20-core ARM CPU, configured with 10 performance and 10 efficiency cores based on Arm v9.2 architecture, and supports up to 128GB of LPDDR5X unified memory.
•Major PC manufacturers including Dell, Lenovo, and ASUS are anticipated to launch laptops powered by the N1x SoC, indicating broad OEM support for Nvidia's new platform.

📊 Competitor Analysis▸ Show

Feature/Spec	Nvidia N1x SoC (Top-end)	Apple M3 Max (16-core CPU/40-core GPU)	Qualcomm Snapdragon X Elite (X1E-84-100)
Manufacturing Process	TSMC 3nm	TSMC 3nm (N3B)	TSMC 4nm (likely)
CPU Cores	20-core (10 Performance + 10 Efficiency) Arm v9.2	16-core (12 Performance + 4 Efficiency)	12-core (8 Performance + 4 Efficiency)
GPU Cores	Blackwell architecture, 6,144 CUDA Cores (48 SMs), RTX 5070-class	40-core GPU	Adreno GPU
Neural Engine/NPU	180-200 TOPS AI computing power	16-core Neural Engine, 18 TOPS	45 TOPS NPU
Memory	Up to 128GB LPDDR5X unified memory	Up to 128GB unified memory (LPDDR5-6400)	Up to 64GB LPDDR5X unified memory
Memory Bandwidth	~273 GB/s (GPU) / 300 GB/s (NVLink C2C)	400 GB/s	136 GB/s
TDP/Power	45W-80W (N1X), 18W-45W (N1)	Not explicitly stated as TDP, designed for laptops	~20-30W (typical for laptops)
Target OS	Windows on ARM	macOS	Windows on ARM
Key Differentiator	Full CUDA ecosystem, RTX-class integrated graphics	High performance-per-watt, integrated ecosystem	Strong battery life, Windows on ARM pioneer
Geekbench 6 CPU (Prototype)	Single-core: ~3096, Multi-core: ~18837 (N1x)	Single-core: ~3000-3200, Multi-core: ~21000-24000	Single-core: ~2900-3000, Multi-core: ~15000-16000
Geekbench OpenCL (GPU)	46,361 (N1x prototype)	~37,500	N/A

🛠️ Technical Deep Dive

Architecture: The N1x SoC combines a CPU die designed by MediaTek with a GPU die built on Nvidia's Blackwell architecture in a 2.5D package.
CPU: Features 20 Arm v9.2 cores, split into two clusters of 10 performance cores (Cortex-X925) and 10 efficiency cores (Cortex-A725), sharing 32 MB of L3 cache.
GPU: Based on the Blackwell architecture, it includes 48 Streaming Multiprocessors (SMs) for a total of 6,144 CUDA cores, which is the same shader count as the desktop GeForce RTX 5070. It also features fifth-generation Tensor Cores with NVFP4 precision and dedicated ray tracing cores.
Manufacturing Process: Built on TSMC's 3nm process.
Memory: Supports up to 128GB of LPDDR5X unified memory across 16 channels, with a reported memory bandwidth for the GPU of approximately 273 GB/s.
Interconnect: The CPU and GPU chiplets are connected via Nvidia's NVLink C2C interconnect, providing 300 GB/s bidirectional bandwidth.
AI Performance: The N1x is expected to deliver 180-200 TOPS of AI computing power, exceeding Microsoft's Copilot+ certification requirement of 45+ TOPS.
Power Envelope: The N1x models are designed to operate within a power range of 45W to 80W, while the standard N1 chips target a lower 18W to 45W range.
PCIe Support: Both N1 and N1x processors feature PCIe 5.0 and 4.0 connectivity, with N1X offering 8x PCIe 5.0 lanes and 3x PCIe 4.0 lanes.

🔮 Future ImplicationsAI analysis grounded in cited sources

Nvidia's N1x will significantly intensify competition in the Windows on ARM PC market.

By bringing its powerful Blackwell GPU architecture and the CUDA software ecosystem to Windows on ARM, Nvidia directly challenges Qualcomm's established position and offers a compelling alternative for AI and graphics-intensive workloads.

The N1x could accelerate the adoption of Windows on ARM for 'AI power users' and potentially redefine the 'AI PC' category.

With its high AI computing power (180-200 TOPS) and RTX-class integrated graphics, the N1x provides a robust platform for on-device AI and demanding applications, potentially attracting users who require more than current x86 or other ARM solutions offer on Windows.

Nvidia's re-entry into the consumer PC CPU market could lead to a more diverse and competitive laptop landscape.

The N1x, alongside the N1, targets different performance and price segments, potentially opening doors for new form factors like high-performance handhelds and a broader range of OEM PCs, fostering innovation beyond the current x86 and Apple Silicon dominance.

⏳ Timeline

2010-09

Windows demonstrated running on Nvidia Tegra ARM, showcasing early collaboration.

2012-10

Microsoft launched Windows RT on the Surface RT tablet, powered by Nvidia Tegra 3 CPU.

2023-10

Reuters reported Nvidia was developing ARM-based CPUs capable of running Windows OS.

2024

Internal Nvidia documents related to N1/N1x development were dated, indicating ongoing work.

2025-07

Nvidia N1x SoC appeared in Geekbench OpenCL listings, revealing GPU capabilities and 20-core CPU setup.

2026-05

Nvidia, Microsoft, and ARM posted coordinated social media teasers for 'A new era of PC' ahead of Computex, with N1x specs leaking shortly after.

Nvidia N1x SoC benchmarks trail Apple M3 Max

⚡ 30-Second TL;DR

Key Points

🧠 Deep Insight

🔑 Enhanced Key Takeaways

🛠️ Technical Deep Dive

🔮 Future ImplicationsAI analysis grounded in cited sources

⏳ Timeline

📎 Sources (27)

👉Related Updates