Zak Killian

When you read that headline, it probably sounds like former Intel CTO and CEO Gelsinger’s got a case of sour grapes. Of course, the man’s far more mature and experienced than that. His comments came while speaking to the Acquired podcast as an invited guest at NVIDIA’s GTC 2025 conference, where the GPU vendor unveiled Blackwell Ultra GB300. Gelsinger, speaking in the pre-show podcast, said this to the hosts: Jensen and I had numerous conversations

It’s been six months (yes, really) since we were first introduced to the Zen 5 CPU architecture with the Ryzen 9000 series desktop CPUs, and it’s been nearly a year and a half since AMD unveiled the Threadripper Pro 7000 series processors. Isn’t it about time for a new Threadripper generation? Apparently so, because those parts are seemingly already in shipping containers, according to the latest leak. Spotted by the ever watchful eye of a community

AMD’s just launched its Strix Halo high-end mobile APUs, but progress marches ever forward, and so it goes that we are now looking at leaks concerning the Strix successor, supposedly known as Medusa Point. Based on the Zen 6 “Morpheus” CPU core, Medusa Point has been said to sport a twelve-core CPU CCD as well as an eight-WGP GPU based on the potent RDNA 4 architecture. Well, play the extremely loud incorrect buzzer noise, because

Imagine, for a moment, that you had a brand-new M.2 SSD rated for sequential transfers of over 15 GB/second. Obviously, it’s the latest design, using four lanes of PCI Express 5.0 to achieve this immense throughput. Alternatively, it could be an x1 design—using PCIe 7.0. That’s right kids: PCIe 7.0 offers per-pin transfer rates of 128 GT/s, meaning that an x16 configuration offers 256 GB/second of bandwidth one way. If you’re thinking that we just

Building out the datacenters—or as NVIDIA calls them, “AI factories”—required for training and operating state-of-the-art AI models is extremely expensive. So much so that even hyperscalers like Google and Microsoft are going to have to plan these purchases well in advance. To that end, NVIDIA offered a sneak peek of its upcoming products at today’s GTC opening keynote, and you might want to sit down for this one. The company’s extant Blackwell processors are most

NVIDIA’s Blackwell GB200 is an absolutely monstrous processor with up to 10 petaflops of dense FP4 tensor compute and 192GB of lightning-fast HBM3e memory delivering 8 TB/second of bandwidth per GPU. That number, though, that 192GB, is still limiting for hyperscalers building the latest, greatest AI models. For those companies that need ever greater memory specs, NVIDIA has just revealed the Blackwell Ultra GB300 GPU. NVIDIA CEO Jensen Huang announced Blackwell Ultra today on stage

Intel has achieved a significant early milestone in its Arizona semiconductor fabrication plant, with the first wafers successfully processed on its advanced 18A process node. This achievement, announced at the same time as its March 13th conference, signals positive momentum for the company’s manufacturing roadmap. The news first surfaced publicly via LinkedIn, where Pankaj Marria, an Engineering Manager at Intel and member of an internal team apparently known as the ‘Eagle Team,’ celebrated the achievement.

The last few weeks for AMD have been pretty remarkable. The chipmaker is firing on all cylinders: the Ryzen 9 9900X3D and 9950X3D have arrived to universal praise, the Ryzen AI Max “Strix Halo” processors finally launched to wide acclaim, and the Radeon RX 9070 series graphics cards brought the RDNA 4 architecture and a much-needed injection of sane pricing to a downright brutal graphics market. AMD’s David McAfee agrees, stating outright that it has

One by one, the barriers to gaming bliss on Snapdragon laptops are falling. Back in November, a new build of Windows brought support for Intel’s Advanced Vector Extensions to the Prism emulator that is used to translate x86-64 program code into Armv8 instructions that the Snapdragon CPUs can execute. Now, Epic Games has announced that it is releasing a new version of the Epic Online Services SDK that adds Arm support. That SDK includes Epic

For decades, the semiconductor industry has been laser-focused on shrinking silicon transistors, but Peking University researchers believe the future might lie in changing materials entirely. In a newly published paper, the team presents a 2D gate-all-around (GAA) transistor that they claim beats the latest silicon-based designs from Intel and TSMC in both speed and energy efficiency. Built using bismuth oxyselenide (Bi₂O₂Se), a high-mobility 2D material, the transistor operates at lower voltages and could enable ultra-dense