NVIDIA RTX Spark SoC Launches for Windows with AI Focus
NVIDIA enters the Windows silicon market with the RTX Spark chip, targeting AI developers and content creators with strong GPU performance.
Key points
- NVIDIA launches the RTX Spark SoC, featuring up to 20 cores and 6,144 GPU cores, for Windows devices.
- The chipset is optimized for on-device AI workloads, competing directly with Apple Silicon in the high-end market.
- Initial benchmarks show strong multi-thread CPU performance and superior GPU capabilities compared to rivals.
NVIDIA has officially launched its RTX Spark system-on-chip for Windows devices, marking the company's return to the Windows silicon arena. The new hardware, developed in collaboration with MediaTek, begins shipping to consumers this week through six hardware partners, including a Surface Laptop Ultra configuration.
The RTX Spark architecture features an 18 or 20-core design with between 5,120 and 6,144 GPU cores. It supports up to 128GB of unified memory, providing significant processing power for applications, games, and artificial intelligence experiences. This launch represents a strategic shift for NVIDIA, aiming to compete head-to-head with Apple's MacBook Pro lineup.
Design and Optimization Strategy
Microsoft has worked closely with NVIDIA to optimize the Windows platform specifically for the RTX Spark chipset. This collaboration aims to ensure a smooth launch for high-end content creators and developers. Third-party vendors, including Adobe and Blender, have also mobilized to optimize their products and frameworks for the new hardware.
Anti-cheat software providers and other framework developers are similarly adapting their tools for the RTX Spark architecture. NVIDIA positions this chipset as the first for Windows designed from the ground up to run local AI experiences securely and with high performance.
Performance Benchmarks and Context
Testing of the top RTX Spark configuration reveals strong GPU performance and solid CPU capabilities. The CPU performance is comparable to the Apple M4 Pro and Snapdragon X2 Elite, with multi-threading being its stronger asset. However, it trails behind the more powerful Apple M4 Max and Snapdragon X2 Elite Extreme chips in certain metrics.
The primary focus of the RTX Spark is on-device AI, capitalizing on the growing demand for large language models and agentic workflows. NVIDIA believes this chipset allows users to run these AI experiences locally without sacrificing performance, appealing to AI developers who previously favored Apple Silicon platforms.
What should you do?
Security and technology professionals should monitor the optimization status of critical third-party applications for the new RTX Spark architecture. Organizations evaluating high-end Windows devices for AI development should review the specific core configurations and memory limits available from the six hardware partners.
Users interested in local AI workloads should verify that their specific use cases are supported by the current drivers and software frameworks provided by vendors like Adobe and Blender. IT departments should assess whether the multi-thread performance advantages justify the transition from existing Apple or Snapdragon-based devices.
WATCH FOR:
- Official driver updates from NVIDIA for the RTX Spark chipset.
- Performance reviews of specific laptop models from the six hardware partners.
- Compatibility reports from major software vendors like Adobe and Blender.
Background: Large language models
Large language models are a specific type of neural network trained on text. Generative AI is the broader umbrella term for any system that creates new content, including images and code. Not all generative AI uses large language models, but most modern text generators do.
What to do and how to stay safe: NVIDIA
- Hardware-backed key storage prevents attackers from extracting credentials even if they gain physical access to the device.
- Mobile device management enforces consistent security policies across diverse operating systems and hardware configurations.
- Application sandboxing limits the damage a compromised app can cause by restricting its access to other data and system resources.
Mobile device security requires a layered approach combining hardware, software, and policy controls. Enable encryption, use strong passcodes, and keep your operating system updated to protect your data.
Step-by-step guide: Mobile Device Security: How to Protect Data Beyond Passwords
General security guidance from the Threat Times newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
What is the RTX Spark chipset?
It is a new Windows system-on-chip from NVIDIA, developed with MediaTek, featuring up to 20 cores and designed for AI and content creation.
How does RTX Spark CPU performance compare to competitors?
It is comparable to the Apple M4 Pro and Snapdragon X2 Elite, with strong multi-thread performance, but trails the Apple M4 Max.
Who are the hardware partners for RTX Spark?
NVIDIA has six hardware partners launching devices this week, including a Surface Laptop Ultra, aimed at high-end creators and developers.
