RTX Spark won’t be a “one-off” launch, and as NVIDIA mentioned during the chipset’s official announcement, the company is preparing successors, starting with the Vera Rubin Spark architectures. Pretty soon, that compute performance gap of the Cortex-X925 clusters that everyone is grumpy about will become a memory. 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
RTX Spark Is Apple’s And Qualcomm’s Kryptonite In One Crucial Area And These Two Only Have A Year To Get Things Right… But now, the facility is aiming to become the Crown Jewel of the next chapter in semiconductors: Glass Substrates and Silicon Photonics. Most recently, Intel began offering silicon photonics manufacturing at Rio Rancho to its external foundry customers. …
… Apple has also detailed the architecture of its Private Cloud Compute PCC framework , going on to note: What's new with PCC on Google Cloud is the implementation: "NVIDIA Confidential Computing with NVIDIA GPUs, Intel CPUs with TDX, and Google's Titan chip." Apple states that while the AFM Cloud is… …
… NVIDIA proclaims that the feature "preserves the confidentiality and integrity of AI models deployed on Rubin, Blackwell, and Hopper GPUs," while enabling "sensitive AI workloads to run securely at scale with near-native performance, even in shared or cloud environments." This step should help Appl… …
… Even AMD’s 16-core Ryzen AI Max+ 395 is no match for the RTX Spark either, with only the power-hungry Intel Core Ultra 9 285HX and M5 Pro getting the better of their new rival In the Clang benchmark, which was shared by @lafaiel on X, it was compared against a host of laptop processors and various … …
… Gaming continues to be a weakness for both Apple Silicon and Snapdragon, with only a decent experience possible by spending top-tier funds on the M5 Max Before the arrival of RTX Spark, having heaps of memory was an Achilles Heel for desktop and laptop GPUs when running AI models locally. …
… All of these come together in the NVIDIA Vera Rubin NVL72 rack, which offers some impressive uplifts versus Blackwell as detailed below: 5x NVFP4 Inference 3.6 EFLOPS 3.5x NVFP4 Training 2.5 EFLOPS 2.5x LPDDR5x Capacity 54 TB 1.5x HBM4 Capacity 20.7 TB 2.8x HBM4 Bandwidth 1.6 PB/s 2x Scale-Up Bandw… …
… With renewed emphasis on world-class process technology, advanced packaging, purpose-built AI silicon, and a thriving foundry business, he is steering the company toward leadership not just in its own products, but as a vital partner for the industry. …
… All of these come together in the NVIDIA Vera Rubin NVL72 rack, which offers some impressive uplifts versus Blackwell as detailed below: 5x NVFP4 Inference 3.6 EFLOPS 3.5x NVFP4 Training 2.5 EFLOPS 2.5x LPDDR5x Capacity 54 TB 1.5x HBM4 Capacity 20.7 TB 2.8x HBM4 Bandwidth 1.6 PB/s 2x Scale-Up Bandw… …
… Intel is also preparing its Xeon onslaught against NVIDIA Vera, and the custom silicon threat is also rising with major AI firms developing their own CPUs for Agentic AI. …
… The HBF standard is ideal for inferencing workloads, which matter more today due to a surge in Agentic AI. The higher capacity also offsets some of the KV Cache constraints from the main compute chip. …