Nvidia introduced the RTX Spark chips and new Windows devices carrying these chips at the Computex technology fair in Taiwan. The company's goal is to create a more powerful and efficient laptop class better suited for running native artificial intelligence.
The new systems combine unified memory architecture, RTX graphics units, and an Arm-based processor called N1. This structure stands out as a new option on the Windows side for AI enthusiasts and users seeking high performance.
Highlights
- Nvidia announced the RTX Spark series at Computex.
- Devices will come with unified memory, RTX graphics, and N1 CPU.
- The new models are expected to emphasize native AI usage in the Windows ecosystem.
- Top-tier models are rumored to offer up to 128 GB of memory and processors with up to 20 cores.
- Prices are said to potentially exceed $4,000 in the high segment.
Search for a competitor to MacBook Pro
The text emphasizes that these devices could form an alternative on the Windows side comparable to the MacBook Pro. Up to 128 GB of unified memory and graphics power approaching the RTX 5070 level set an unusual performance target for portable computers.
This approach is described as an important step for users who want to run large language models and other AI workloads locally. Privacy and speed advantages are highlighted especially for those who prefer processing on the device rather than in the cloud.
New partners coming to the Windows ecosystem
Nvidia is opening its technology to Windows manufacturers like HP, Asus, Dell, Lenovo, and others. This is expected to result in the RTX Spark architecture appearing not only in a single model but in different designs and price levels.
On the Microsoft side, the model called Surface Laptop Ultra draws attention. Featuring a 15-inch Mini-LED screen, various ports, and a body design similar to the MacBook Pro, the device gives a new direction to the Surface line, which has long not been performance-focused.
Battery, memory, and performance balance
Battery life has long been a challenging area for Windows laptops. Dependence on discrete graphics cards has put pressure on battery life and fan noise. Although new Intel Core Ultra Series 3-based models have started to improve this balance, the memory limit remains at 64 GB in most configurations.
The structure offered by Nvidia stands out not only with memory capacity but also with the CUDA software layer. Developers' ability to directly access GPU cores could make the Windows side more ambitious in native AI operations.
Prices point to the high segment
The new computers are not expected to be affordable. Some estimates show that high-end RTX Spark models could exceed $4,000. This level points to a price range similar to powerful MacBook Pro models.
The company is also preparing to add small form factor desktop computers to this family. Lower hardware options are expected to appeal to gamers, content creators, and users focused on both gaming and productivity.
New competition for Intel, AMD, and Qualcomm
Nvidia's move could increase competitive pressure on Intel, AMD, and Qualcomm in the Windows PC market. The proliferation of devices focused on native AI, with high memory and powerful graphics processors, could also change computer usage habits.
The text says AI models may be run directly on devices in more workflows in the future, which is especially important for privacy and practical use. The company's move brings a new performance class long missing on the Windows side to the agenda.
Context: Nvidia holds a strong position in AI processing power in data centers. With RTX Spark, it aims to bring this experience to local computers.