Tech Logic

Intel 18A-P Enters Trial Production: Confirmed PPA Gains and Unverified Detail Boundaries

Intel has announced that its 18A-P manufacturing process has entered the test/risk production stage. The core fact confirmed by three sources is that the node continues to build on 18A and is described as superior to 18A in power, performance, and area (PPA); one source provides specific figures, saying it delivers a 9% performance improvement at the same power level and an 18% reduction in power consumption. One source also mentions enhancements to GAA and backside power delivery (BSPD/Backside Power Delivery) technologies. However, the event summary’s references to PowerBoost, 0.5V, CFET, and a broader roadmap could not be confirmed from the provided sources.

Jun 24, 2026

HKU Team Unveils Cryogenic Brain-Inspired Chip Research: SiC Transistor Simulates Neuronal Spikes at 10 mK

A University of Hong Kong team has announced research based on a Nature Communications paper, reporting a cryogenic neuromorphic circuit in silicon carbide (SiC) MOSFETs that can operate in ultra-low-temperature environments. A single transistor can mimic neuronal spiking behavior, with potential applications in quantum computing control and deep-space exploration electronics. Three sources are broadly consistent on the paper title, materials, research leaders, and the core low-temperature NDR/spiking behavior. Claims such as “world-first” and the exact scope of applications are presented with promotional differences in some sources, and no further quantifiable data were provided for verification.

Jun 13, 2026

Microsoft unveils second-generation topological quantum chip Majorana 2, says qubit lifespan has increased by about 1,000 times and points to commercialization in 2029

Microsoft has announced its second-generation topological quantum chip, Majorana 2. All three sources confirm the core claims: qubit lifespan is about 1,000 times longer than in the first generation, averaging around 20 seconds and reaching up to one minute in some cases; the company is also pointing to 2029 as the timeline for scalable, commercially viable quantum computers. Based on the provided sources, the TSO verification conclusion is that the three sources are highly consistent on the key facts, but Sources 1 and 2 do not provide verifiable background details, and the concerns about reproducibility and the lack of peer review appear only in the event summary and cannot be confirmed from the given sources.

Jun 5, 2026

Huawei Introduces the Tau Scaling Law: Pursuing New Paths to Improve Chip Efficiency Through Advanced Packaging, 3D Stacking, and Logic Layering

At the end of May 2026, Huawei publicly explained its new chip concept, the “Tau Scaling Law” or “Tau Law,” emphasizing that it would no longer rely solely on continued transistor miniaturization, but instead improve chip density, performance, and efficiency through advanced packaging, 3D stacking, and LogicFolding. Reuters and The Wall Street Journal both confirmed this direction; related DIGITIMES coverage mentioned the role of glass substrates in this approach, although the main article did not directly expand on Huawei’s chip technology. Some details, such as whether glass substrates are central and the exact scope of industry impact, could not be fully confirmed from the available sources.

Jun 4, 2026

Lisuan Tech’s First In-House GPU, the LX 7G100, Sells Out and Draws Reviews: Compatibility Is Taking Shape, but Performance Still Trails Mainstream Nvidia GPUs

After the launch of Lisuan Tech’s first self-developed graphics card, the Lisuan LX 7G100, media outlets and reviewers focused on its sell-out status, compatibility, and performance. Three sources consistently confirm that the card has been sold publicly and that the company said the first 30,000 units were sold out. It also appears to support Windows, DirectX 12, Vulkan, OpenCL, and OpenGL. On performance, sources indicate it can come close to the Nvidia RTX 3060 in some tests, but it still struggles to compete with mainstream Nvidia GPUs overall. Claims that it is explicitly positioned against the RTX 4060, as well as more complete benchmark details, could not be fully confirmed from the available sources.

May 27, 2026

AMD says FSR 4.1 will expand to older Radeon GPUs, with RDNA 3 in July and RDNA 2 in early 2027

AMD announced on X via Jack Huynh that FSR 4.1 will first roll out to RDNA 3 graphics cards and then reach RDNA 2 in early 2027. Three sources broadly agree on the core timeline, though they differ in wording and some model- and technology-specific details could not be independently verified.

May 20, 2026