NVIDIA’s RTX Spark announcement revealed that the chipset’s 20-core CPU would be paired with a Blackwell GPU, but there were scarce details surrounding the graphics processor. Fortunately, a closer look at the latest Windows 11 drivers brings us one step closer to learning about the SoC, specifically its CUDA core count. Based on the latest findings, the RTX Spark will arrive in two versions.
The less powerful RTX Spark will offer 5,120 CUDA cores, with the GPU possibly teaming up with an 18-core CPU
The first GeForce graphics driver for the RTX Spark has been uploaded by NVIDIA, with version 616.00 said to be a Developer Preview rather than an official release. While the driver isn’t intended to be used on consumer laptops, that hasn’t stopped anyone from snooping in on the more useful information.
Upon further investigation, Windows Latest spotted a nv_surface_woa.inf file that contains all the information there is to know about the device on which the driver is said to be installed. The .inf file includes two RTX Spark configurations labeled “NVIDIA RTX Spark N1X,” with one of the units featuring a 6,144-core Blackwell GPU, while the other features 5,120 CUDA cores.
It’s likely that the top-end version with the highest number of cores will be paired with MediaTek’s 20-core CPU and 48 Streaming Multiprocessors, while the slower variant will be kitted with an 18-core CPU and 40 Streaming Multiprocessors. Thankfully, NVIDIA has said that the maximum memory support will be 128GB of LPDDR5X RAM, so consumers who want to sacrifice a little bit of compute and graphics performance but want the full RAM count will be happy to learn about this similarity.
It’s important to note that the number of CUDA cores isn’t an indication of the RTX Spark’s performance, as the version with 6,144 cores matches this specification with a desktop RTX 5070 and laptop RTX 3080. Similarly, a laptop RTX 3070 features 5,120 CUDA cores, the same as the less capable RTX Spark, but we’ve reported that both Alan Wake 2 and PRAGMATA appear to run fluidly on the platform, and that too with path tracing enabled.
About the author: Omar Sohail is a reporter and analyst for Wccftech's mobile section, specializing in the technology and business of the mobile industry. His expertise lies in the intricate hardware supply chain, covering developments in semiconductor manufacturing, chip lithography, and camera sensor technology.
Follow Wccftech on Google to get more of our news coverage in your feeds.
