MicroCloud Hologram Advances Deployable Quantum Recurrent Neural Network Technology
MicroCloud Hologram Inc. (HOLO) announced a Quantum Recurrent Neural Network (QRNN) built around a novel Quantum Recurrent Block (QRB) architecture designed for noisy intermediate‑scale quantum (NISQ) devices. The QRB acts as a modular, repeatable subcircuit that drastically reduces coherent‑time consumption, while an interleaved‑stacking scheme reuses the same block across time and feature dimensions. A hybrid quantum‑classical training loop lets classical optimizers update QRB parameters, delivering higher prediction accuracy on time‑series benchmarks than conventional recurrent neural networks. The company positions the QRNN as a first‑generation quantum‑advantaged model for sequential learning tasks.
CoreWeave Announces Agreement to Power Perplexity’s AI Inference Workloads
CoreWeave and Perplexity have signed a multi‑year strategic partnership to run Perplexity’s inference workloads on CoreWeave Cloud, using dedicated NVIDIA GB200 NVL72 clusters. The deal includes deployment of Perplexity Enterprise Max, which adds advanced search, research, and data‑visualization capabilities for...
DOE Advances Domestic Capabilities for Producing Quantum Materials
The Department of Energy announced that Pacific Northwest National Laboratory has built systems to turn commercially sourced isotopically enriched silicon and germanium into high‑purity silane (SiH₄) and germane (GeH₄) gases. These precursor gases are essential for quantum information science and...
Nebius Secures Approval for Its First Gigawatt-Scale AI Factory
Nebius received Independence City Council approval for its Chapter 100 industrial incentive plan, clearing the way to build a 1.2‑gigawatt AI factory on a 400‑acre campus in Independence, Missouri. The development will generate roughly 1,200 skilled construction jobs and about...
Quantum Elements Reports Record Logical Qubit Fidelity in Nature Communications Study
Quantum Elements, together with USC, IBM and RWTH Aachen, published a Nature Communications paper showing the highest‑fidelity entangled logical qubits on a 127‑qubit superconducting processor. By merging quantum error detection with a novel logical dynamical decoupling (LDD) technique, the team...
Origin Quantum Computing Achieves High-Accuracy Flux Crosstalk Compensation in Superconducting Qubits
Origin Quantum Computing, in partnership with USTC, unveiled a spin‑echo‑based method that separates quantum and magnetic flux crosstalk in superconducting qubits. The technique, combined with a learning‑driven high‑parallelism measurement scheme, stabilizes frequency‑shift fluctuations to a 20 kHz noise floor. After compensation,...
RVA23 Ends Speculation’s Monopoly in RISC-V CPUs
RVA23 declares the RISC‑V Vector Extension (RVV) a mandatory feature, turning explicit vector parallelism into a baseline capability for all compliant CPUs. By offloading throughput work to deterministic vector units, scalar cores can become simpler, low‑power coordinators without sacrificing performance....
Press Release: Gilat Secures Multi-Million Dollar IFC Hardware Order
Gilat Satellite Networks announced a multi‑million‑dollar order from a leading in‑flight connectivity integrator for its AeroStream Ka‑band wideband amplifiers. The hardware, including 60 W Ka‑band block upconverters, will be delivered over the next 24 months to equip commercial aircraft. Gilat notes...
Video Wednesday
On November 20, 2020, Verb and Johnson & Johnson unveiled OTTAVA, a six‑armed robotic platform designed for minimally invasive procedures. The system combines advanced haptic feedback with AI‑driven motion planning to enhance surgeon precision. Early trials suggest OTTADA can reduce operative time...
Never Sell: MBI Interviews Asianometry
MBI’s “Never Sell” podcast featured a conversation with Asianometry, the popular YouTube educator known for deep dives into semiconductor technology. The episode explored the entire semiconductor value chain, from design through testing, and highlighted the creator’s ability to simplify complex...
Linux Preps IBPB-On-Entry Feature For AMD SEV-SNP Guest VMs
Linux is integrating the IBPB‑on‑Entry feature for AMD SEV‑SNP guest VMs, slated for the upcoming 7.0 kernel and back‑porting to stable releases. The feature, native to AMD EPYC Zen 5 processors, inserts an Indirect Branch Predictor Barrier on VM entry to...
(PR) Razer Introduces BlackShark V3 X HyperSpeed for Xbox and PlayStation - White Edition
Razer announced the White Edition of its BlackShark V3 X HyperSpeed headset for both Xbox and PlayStation consoles. The new colorway retains the lightweight, wireless design and esports‑grade audio performance of the original model. It targets next‑generation console users seeking...
(PR) QNAP Releases TS-H1077AFU 10-Bay, High-Density All-Flash NAS with AMD Ryzen PRO 7000 CPUs
QNAP announced the TS-h1077AFU, a 10‑bay all‑flash NAS built on the ZFS‑based QuTS hero operating system and powered by AMD Ryzen PRO 7000 Series CPUs. The unit ships with ECC‑protected DDR5 memory, delivering enterprise‑grade performance and data integrity in a...
MWC 2026: Huawei New-Gen OceanStor Dorado Converged All-Flash Storage Passes Enterprise Strategy Group Technical Validation
Huawei's New‑Gen OceanStor Dorado Converged All‑Flash Storage received technical validation from Enterprise Strategy Group (ESG). ESG's tests showed the system delivering over 876,000 IOPS with a 32 µs average latency in a high‑concurrency database workload. The architecture supports active‑active failover, tolerates...
Technology Shorts March 2026
Researchers unveiled four breakthrough technologies that could reshape communications and data‑center infrastructure. Columbia University’s "rainbow chip" creates a frequency‑comb, allowing dozens of precise light channels to travel simultaneously through a single fiber, promising massive bandwidth gains for broadband networks. The...
Memory Makers Shift to Hourly Contracts as AI Demand Continues to Climb
Memory manufacturers such as SK hynix, Samsung and Micron have moved to hourly DRAM pricing contracts, forcing OEMs to secure quotes within a single hour. The shift reflects soaring AI‑driven demand that makes prices fluctuate by the hour, prompting faster settlement...
ASUS Raises GeForce RTX 50 Series "Blackwell" Pricing in China, Radeon Pricing Unchanged
ASUS announced price adjustments for its GeForce RTX 50 Series "Blackwell" GPUs in China, reflecting a broader memory component shortage. The flagship RTX 5090 D v2 sees a 500‑yuan (~$72) increase, while mid‑range models like the RTX 5080, 5070 Ti,...
Steam Deck Restocks Resume After Valve Fixes RAM Supply
Valve announced that Steam Deck restocks have resumed after the company resolved a RAM component shortage that had throttled production of higher‑capacity models. The 1TB and 512 GB variants are now appearing at their original MSRP, marking the first significant supply...
AMD Engineer Leverages AI To Help Make A Pure-Python AMD GPU User-Space Driver
AMD’s VP of AI Software, Anush Elangovan, leveraged Claude Code to create a pure‑Python user‑space driver that talks directly to /dev/kfd and /dev/dri/renderD* via ctypes ioctls, bypassing the ROCm/HIP stack. The initial commit adds KFD ioctl bindings, SDMA copy engine...
Open-Source AI Hardware Could Weaken Big Tech’s Grip on AI
Current AI, a $400 million public‑interest partnership, unveiled an open‑source handheld AI device at the India AI Impact Summit. The offline prototype, built with India’s Bhashini translation project, can see, speak, and answer questions in Hindi and English, even identifying candy...
Old ATI R300 Open-Source Driver Sees Another New Fix In 2026
The open‑source ATI R300 Gallium3D driver, originally reverse‑engineered for legacy Radeon hardware, received a critical occlusion‑query fix this week. Independent developer Pavel Ondračkar submitted a patch that avoids the HyperZ breakage of a 2012 attempt and uses a dummy Z‑buffer...
EA Working on Javelin Anticheat Port for Windows-on-Arm
Electronic Arts announced it is hiring a senior engineer to port its Javelin kernel‑level anticheat to the Windows‑on‑Arm (WoA) platform. The move confirms that major publishers are preparing their security stacks for ARM‑based Windows 11 PCs. It coincides with NVIDIA's...
Tejas Networks Unveils Hyper-Scalable DCI Platform at MWC Barcelona
Tejas Networks introduced the TJ1600‑D3, a hyper‑scalable data‑center interconnect (DCI) platform, at MWC 2026 in Barcelona. The WDM optical transport system delivers line rates from 400 Gbps to 1.2 Tbps per wavelength and can scale to 51.2 Tbps of shelf capacity within a compact...
Neither Android nor iOS: DIY Smartphone Runs on ESP32!
Maker LuckyBor has built a functional 4G smartphone using an ESP32‑S3 as its core processor. The device combines a SIMCOM A7682E modem, an OV2640 camera, a 3.5‑inch touchscreen and a 3.5 mm audio jack, delivering call, text and web capabilities despite...
(PR) AAEON Launches Intelli TWL01 Edge, an Industrial Multimedia PC with Dual 4K Display Support
AAEON’s UP brand unveiled the Intelli TWL01 Edge, an industrial‑grade multimedia PC that supports dual 4K HDMI displays and runs on Intel Core N‑series processors. The fanless, compact chassis measures 152 mm × 124.5 mm × 39 mm and offers DIN‑Rail, VESA, and wall‑mount options for rugged deployments. It...
Understanding 10BASE-T1L
10BASE‑T1L, defined in IEEE 802.3cg, delivers 10 Mbps full‑duplex Ethernet over a single twisted‑pair for up to 1 km. The standard lets BACnet/IP run directly on the link, eliminating gateways and allowing seamless integration with IT networks. Existing two‑wire fieldbus runs can be...
Tiny 3D Printed DIY E-Ink Book Reader : 2 Week Battery & USB-C Charging
Paul Lagier unveiled a DIY Tiny E‑Reader that combines a 3D‑printed shell with an energy‑efficient e‑ink display, USB‑C charging, and wireless file transfer. The pocket‑sized device costs roughly $30 to build and delivers up to two weeks of battery life...
Enhancing Cold Chain Visibility
SpotSee introduced WarmMark QR, a battery‑free, QR‑enabled temperature sensor designed for cold‑chain logistics. The device delivers affordable, real‑time, package‑level temperature data, replacing error‑prone visual indicators. By allowing remote data access, it improves recipient adoption and streamlines monitoring for shippers. The...
Engicam Introduces SmarCore iQ6 SMARC Module with Dragonwing IQ-6
Engicam unveiled the SmarCore iQ6 SMARC 2.2 module, powered by Qualcomm’s Dragonwing IQ‑6 series IQ‑615 SoC. The module offers an 8‑core heterogeneous CPU, up to 8 GB LPDDR4X memory, 8 GB eMMC storage, and an integrated Adreno GPU capable of 4K video decoding...
Telekom Relies on Starlink: Satellites to Close the Last Gaps in Mobile Coverage in Germany
Deutsche Telekom announced a partnership with Elon Musk’s Starlink to launch direct‑to‑device satellite mobile service in Germany starting in 2028. The initiative targets the remaining 1.2% of the country without 2G coverage, offering basic voice, data and messaging where terrestrial...
Grinn ReneSOM-V2H Module Runs Renesas RZ/V2H Vision AI Processor
Polish embedded‑systems firm Grinn unveiled the ReneSOM‑V2H, the world’s smallest System‑on‑Module built around Renesas’ RZ/V2H vision AI processor. The 37 × 42.6 mm board combines four Cortex‑A55 cores, two Cortex‑R8 real‑time cores, a Cortex‑M33 MCU and an 8 TOPS DRP‑AI3 accelerator for edge‑AI workloads....
Dependency Chain Stalls
The ninth video in the Performance‑Aware Programming series highlights that a CPU’s ability to extract instruction‑level parallelism is bounded by dependency chains. When later instructions must wait for earlier results, the pipeline stalls, limiting throughput. The post underscores the need...
Phone Screens Got 25% More Efficient Without Losing Pixels
TCL CSOT unveiled its Super Pixel display at MWC 2026, promising up to 25% lower power consumption while delivering sharper images and faster refresh rates. The technology swaps the traditional SPR sub‑pixel rendering for a Real RGB layout, adding only 1.8% more...
NTT DATA Continues Data Center Momentum with Major Commitments From Hyperscale and Enterprise Clients
NTT DATA’s Global Data Centers business secured four capacity commitments totaling nearly 115 MW across campuses in Gainesville, Virginia; Chicago, Illinois; and Sacramento, California. The deals include more than 90 MW for a major hyperscale provider and roughly 20 MW from three enterprise...
COMPUTEX 2026 Brings the Global AI Ecosystem to Taipei
Computex 2026 will run June 2‑5 in Taipei across four venues, featuring 1,500 exhibitors and 6,000 booths, positioning the event as a global AI benchmark. The show’s “AI Together” theme spotlights AI & computing, robotics & mobility, and next‑gen technologies,...
Ayar Labs Closes $500M Series E, Accelerates Volume Production of Co-Packaged Optics
Ayar Labs announced a $500 million Series E round led by Neuberger Berman, bringing total funding to $870 million and valuing the company at $3.75 billion. The capital will be used to scale high‑volume production and testing of its co‑packaged optics (CPO) solution, expand...
Fujitsu to Deploy Underwater Drones and AI for Ocean Digital Twin at Port of Barcelona
Fujitsu Spain and the BCN Port Innovation Foundation have launched a proof‑of‑concept to build an ocean digital twin for the Port of Barcelona. The project combines autonomous underwater drones with AI‑driven analytics to capture high‑resolution seabed imagery, quantify biodiversity, and...
AMD DPTCi Driver Posted For Linux To Better Enhance Ryzen Gaming Handhelds
A request‑for‑comments patch series introduced the AMD Dynamic Power and Thermal Configuration Interface (DPTCi) driver to the Linux kernel, aiming to improve power, performance, and thermal tuning for Ryzen‑based gaming handhelds. The driver exposes seven tuning parameters via a standard...
Zio to Use NEO Battery Cells in Mobile Robots
Zio Robot Co. announced it will replace its Chinese‑sourced batteries with high‑energy lithium‑ion cells from Canada’s NEO Battery Materials for its Mobile Worker (MW) autonomous mobile robots. The partnership aims to boost energy density, high‑rate discharge capability, and overall robot...
Building the Interconnect Foundation: Bump and TSV Planning for Multi-Die Systems
Bump and TSV planning underpins the electrical and mechanical fabric of multi‑die systems, enabling the ultra‑high bandwidth and low latency required for chiplet architectures. As interconnect pitch shrinks, designs can involve millions of contacts, making manual spreadsheet methods infeasible and...
GNOME Mutter 50.rc Released With Better NVIDIA Performance, SDR-Native & Better HDR
The GNOME project released the Mutter 50 release candidate two weeks ahead of the GNOME 50 stable launch, delivering a suite of performance and visual enhancements. Key changes include dramatic latency reductions for the NVIDIA R590 driver, support for SDR‑Native color mode, and...
CHERI: Hardware-Enforced Capability Architecture for Systematic Memory Safety
CHERI (Capability Hardware Enhanced RISC Instructions) introduces a hardware‑enforced capability architecture that replaces raw pointers with bounded, unforgeable references, making out‑of‑bounds memory accesses architecturally impossible. The design adds only about 4‑5% processor area and incurs minimal performance loss, while allowing...
No Lidar, No HD Maps, Six Cameras, One Chip, Autobrains
Autobrains announced a strategic partnership with VinFast to develop an affordable autonomous robo‑car using a vision‑only system. The platform relies on six cameras, a single edge‑compute chip, and an agentic AI architecture that scales via modular skills. It replaces traditional...
Bluefors Introduces Expandable Platform Supporting High Qubit Count Quantum Hardware
Bluefors unveiled its Modular Cryogenic Platform, an expandable vacuum‑chamber system designed to meet the scaling demands of quantum computers. The architecture lets users add modules incrementally, increasing cooling capacity and qubit density without replacing the entire infrastructure. Each module supports...
Illinois Quantum and Microelectronics Park Facilitates Immediate Ecosystem Entry for Future Tenants
The Illinois Quantum and Microelectronics Park (IQMP) launched an On‑Ramp program that places temporary, fully equipped labs in partner facilities to let future tenants start work immediately. Partner sites—mHUB, the UChicago Science Incubator at Hyde Park Labs, and the Discovery Partners Institute—provide...
Apple Introduces Two New Studio Displays
Apple unveiled a refreshed Studio Display lineup, featuring a 27‑inch 5K base model with a 12 MP Center Stage camera, six‑speaker audio, Thunderbolt 5 ports, and up to four‑display daisy‑chain capability. The new Studio Display XDR replaces the Pro Display XDR, adding...
WEBINAR: Two-Part Series on RF Power Amplifier Design
The RF power‑amplifier landscape has shifted from simple scalar load‑pull to sophisticated vector and simulated techniques as mmWave and sub‑THz designs demand tighter margins and wider bandwidths. Keysight’s two‑part masterclass introduces a VNA‑inspired load‑pull method that uses a‑ and b‑waves...
Intel Adapting Linux's LAM In Preparing For ChkTag
Intel engineers are revising the Linux Linear Address Masking (LAM) interface to align with the upcoming ChkTag memory‑tagging extension announced by the x86 Ecosystem Advisory Group. The new patches standardize LAM’s tag width to 4 bits—matching Arm’s MTE and the expected...
David Pogue Tells the Origin Story of the iPhone in Wired’s Excerpt of His Upcoming History of Apple
David Pogue’s forthcoming "Apple: The First 50 Years" offers an in‑depth look at the iPhone’s origin, and Wired has released an exclusive excerpt. The piece spotlights a 2003 collaboration with FingerWorks that produced a large multi‑touch pad, complete with projector‑based...
Intel Preps A Lot Of Xe3 Code For Linux 7.1 Kernel
Intel has submitted its first “drm‑xe‑next” pull request to the DRM‑Next tree, targeting the upcoming Linux 7.1 kernel. The patch set introduces a suite of Xe driver enhancements, including hardware workarounds, SR‑IOV updates, and extensive Xe3(P) support for Crescent Island and...