Automated Optical Inspection


Building good automated models for inspection require more data to be collected, both good and bad. Vijay Thangamariappan, R&D engineer at Advantest, explains how to develop models for automating optical inspection, using a multi-thousand pin socket as an example for how machine learning has helped reduce the return rate due to defects from 2% down to zero. He also explains how to achieve t... » read more

Deep Learning In Industrial Inspection


Deep learning is at the upper end of AI complexity, sifting through more data to achieve more accurate results. Charlie Zhu, vice president of R&D at CyberOptics, talks about how DL can be utilized with inspection to identify defects in chips that are not discernible by traditional computer vision algorithms, classifying multiple objects simultaneously from multiple angles and taking into accou... » read more

Total Critical Area For Optimizing Test Patterns


Increasing complexity at advanced nodes makes it much harder to locate defects and latent defects because there is more surface area to cover and much less space between the various components in a leading-edge chip design. Ron Press, technology enablement director at Siemens Digital Industries Software, talks about why it’s so important to predict where defects are most likely to occur in th... » read more

Growing Challenges With Wafer Bump Inspection


As advanced packaging goes mainstream, ensuring that wafer bumps are consistent has emerged as a critical concern for foundries and OSATs. John Hoffman, computer vision engineering manager at CyberOptics, talks about the shift toward middle-of-line and how that is affecting inspection and metrology, why there is so much concern over co-planarity and alignment, how variation can add up and creat... » read more

Optimizing AI Systems


Inserting AI and machine learning into chips adds a whole new dimension of complexity, and creates a variety of potential problems, including deadlocks, loss of performance, and difficulty in achieving closure on many fronts. Gajinder Panesar, fellow at Siemens EDA, talks with Semiconductor Engineering about what’s changed and how to optimize these new devices and systems by monitoring them f... » read more

Design For Test Data


As design pushes deeper into data-driven architectures, so does test. Geir Eide, director for product management of DFT and Tessent Silicon Lifecycle Solutions at Siemens Digital Industries Software, talks with Semiconductor Engineering about a subtle but significant shift for designing testability into chips so that test data can be used at multiple stages during a device’s lifetime. » read more

Monitoring Performance From Inside A Chip


Deep data, which is generated inside the chip rather than externally, is becoming more critical at each new process node and in advanced packages. Uzi Baruch, chief strategy officer at proteanTecs, talks with Semiconductor Engineering about using that data to identify potential problems before they result in failures in the field, and why it's essential to monitor these devices throughout their... » read more

Silicon Lifecycle Management


How do you track, measure and ensure reliability over the lifetime of a chip, regardless of how or where it is used? Steve Pateras, senior director of marketing for test products at Synopsys, drills down into the impact of hardware-software co-design, over-the-air updates, the expected lifetime of designs, and how the various monitors and sensors are used to track environmental, structural and ... » read more

Using ML In Manufacturing


How to prevent early life failures by applying machine learning to different use cases, and how to interpret models for different tradeoffs on reliability. Jeff David, vice president of AI solutions at PDF Solutions, digs down into how to utilize data to improve reliability. » read more

Ensuring HBM Reliability


Igor Elkanovich, CTO of GUC, and Evelyn Landman, CTO of proteanTecs, talk with Semiconductor Engineering about difficulties that crop up in advanced packaging, what’s redundant and what is not when using high-bandwidth memory, and how continuous in-circuit monitoring can identify potential problems before they happen. » read more

← Older posts Newer posts →