This week in semiconductors brought notable progress in quantum hardware, chip test flows, and workforce development — all of which will eventually trickle down into the embedded and IoT world.
Quantum chips move toward practical systems
Researchers from the University of Pavia, CEA-Leti, and STMicroelectronics have demonstrated a fully integrated quantum frequency processor on a silicon chip. Built on a silicon-photonics platform, it combines quantum-state generation, frequency mixing, and programmable spectral control on a single die — essentially shrinking a sizeable quantum optics lab onto a chip.
Microsoft has made its Majorana 2-based topological qubits available to DARPA for independent testing at a new quantum research center in Maryland, a key step toward validating real-world robustness of these devices.
IonQ reported successful tests of an end-to-end, real-time quantum error-correction decoder running on a standard off-the-shelf CPU. In benchmarks simulating up to 408 logical qubits and more than 31.5 million operations, the decoder added as little as 0.02% latency, showing that practical control stacks for quantum systems can run on conventional hardware.
In Germany, QUDORA is leading a seven-member consortium invited to submit a full proposal for NFQC-1k, a roughly €122 million project to build a trapped-ion quantum computer with at least 1,000 physical and 50 logical qubits, plus a QPU pilot manufacturing line.
Test data and engineering talent
A featured video from Siemens EDA looks at moving rapidly growing test data volumes in modern SoCs without slowing down production test. Vidya Neerkundar compares approaches like compressed test, hierarchical DFT, and emerging standards, and where each fits best.
On the workforce side, the University of Illinois Urbana-Champaign and Purdue University are launching joint programs to train semiconductor engineers in chip design, manufacturing, equipment, materials, testing, and advanced packaging — aiming to shore up the talent pipeline that future embedded and high-performance systems will depend on.
Source: Semiconductor Engineering
