Nvidia RTX Spark Analysis: Finally an answer to Apple's M5 Pro & M5 Max?
Nvidia's RTX Spark platform has finally arrived, marking the company's first foray into the SoC market. This architecture mirrors the approach taken by Apple: a custom ARM-based CPU paired with a "Blackwell" GPU and a unified memory pool of up to 128 GB. We tested the RTX Spark N1X in the Asus ProArt P16 and Microsoft Surface Laptop Ultra to see how it stacks up against Apple M5 Pro and M5 Max.
The N1X Architecture: ARM meets Blackwell
The RTX Spark N1X is a sophisticated hybrid. Its CPU, co-designed with MediaTek, is built on TSMC's 3nm process and features up to 20 cores (10x Cortex X925 performance cores and 10x Cortex A725 efficiency cores). The GPU side is a powerhouse based on the Blackwell architecture (GB20B), featuring 5120 or 6144 CUDA cores running at up to 2535 MHz. This integration allows the chip to leverage cutting-edge software features like DLSS 5 and Frame Generation.
A critical technical detail is the requirement for ARM-based Windows, which may impact software compatibility for some users. The system supports up to 128 GB of LPDDR5x-9400 RAM with a 300 GB/s bandwidth, comparable to the Apple M5 Pro. Furthermore, the CPU and GPU tiles are linked via NVLink, providing a massive 600 GB/s of interconnect bandwidth.
Performance: CPU and GPU Benchmarks
In single-core workloads, the RTX Spark N1X is highly capable, outperforming AMD Zen 5 and rivaling Intel Panther Lake. However, Apple remains the gold standard here, with M5 Pro and M5 Max delivering significantly higher single-core scores. In multi-core scenarios (Cinebench 2024), the N1X competes closely with AMD Strix Halo and even slightly edges out the Apple M6, though it still trails the high-end M4/M5 Max models.
The GPU performance is where Nvidia shines. In synthetic tests like 3DMark Steel Nomad, the RTX Spark delivers performance on par with a mobile GeForce RTX 5070, outperforming the AMD Radeon 8060S by roughly 50%. In real-world gaming, such as Cyberpunk 2077, it performs similarly to an RTX 5060 or 5070, making it a formidable option for gamers, especially when utilizing DLSS. In creative workflows, while Apple holds the edge in CPU-heavy tasks like Adobe Photoshop, the Nvidia solution excels in memory-intensive projects (like large RAW photo editing) due to its massive unified memory capacity.
Verdict: A Strong Debut for High-End Laptops
The RTX Spark is a very solid first effort. It isn't a total
Гика
