Accelerating nonlinear MEMS simulations with the harmonic balance method
See how harmonic balance is leveraged for solving nonlinear periodic problems in frequency domain for quicker, more precise results without transient analysis.
Simulation Challenges and AI-Accelerated Design in SI/PI
See how simulation drives advances in signal and power integrity, where speed and accuracy remain bottlenecks, and how AI and open-source tools could redefine electronic design.
Introduction to the guest
Stephen Newberry is a signal integrity and power integrity engineer at Chipletz. Stephen’s path into engineering started in the U.S. Coast Guard, where he worked hands-on as an electrician. That experience sparked his curiosity for design and eventually led him to specialize in ensuring that high-speed signals move across chips and boards without corruption. Beyond his day job, he also shares his insights through a blog where he writes about the challenges and advances in this space.
In this episode...
We dive into the role of simulation in SI/PI engineering, why accuracy and speed are such critical bottlenecks, and how emerging technologies—from AI-assisted solvers to low-cost open-source tools—could reshape the way engineers design and test electronic systems. Whether you’re working directly with circuit boards or just curious about how simulation impacts the technology all around us, there’s something here for you.
Steven Newberry is a signal integrity and power integrity engineer at Chipletz whose career began in the US Coast Guard as an electrician, where hands-on troubleshooting sparked his interest in design and ultimately led him into engineering.
Signal integrity ensures high-speed, multi-gigabit signals move across chips and boards without corruption, while power integrity maintains stable voltages that directly affect those signals; together they are essential in any electronic system.
SI/PI is critical in virtually all electronics—from CPUs talking to memory in mobile devices to Ethernet networks in homes and offices—wherever chips must communicate reliably.
Simulation goes hand in hand with measurement, enabling engineers to predict and prevent issues before fabrication; more upfront simulation reduces debugging and troubleshooting later.
Comprehensive simulation of entire boards or packages remains infeasible due to complexity, and setup is slowed by imperfect data exchange formats like ODB++. Simulations can also take days or weeks on complex, high-frequency designs.
Computational speed is the major constraint. Because simulation is iterative, faster turnaround directly shortens design cycles and enables better optimization.
SI/PI is relatively young, with the term only appearing in the late 1980s. Tools continue to advance, incorporating AI optimization, while interface standards like PCIe keep doubling data rates, creating constant new challenges for simulation and manufacturing.
Artificial intelligence is expected to play a transformative role, with startups and incumbents alike working on AI-augmented solvers. Faster, more accurate simulations could lower manufacturing costs, improve device performance, or reduce time-to-market.
Listen to the full episode on YouTube
0:00 Introduction
1:21 What is SI/PI?
2:25 Applications of SI/PI
3:02 Stephen's experience in the US Coast Guard
5:21 Role of simulation in SI/PI
6:12 Accuracy of simulation
6:58 Pros and cons in simulation today
8:21 One thing to change with magic wand
9:00 Advancements in SI/PI
11:30 Near-term future of the industry
14:46 Motivation for knowledge sharing
15:55 One surprising fact about SI/PI
16:57 Future outlook in simulation
18:42 Outro
Learn more about the topics
.png?width=300&name=report%20(1).png)
Report
The state of multiphysics simulation in 2025
We asked 250 engineers and decision-makers about challenges with simulation. See all the results in the full report.
Access the full article →

Blog
Why simulation throughput will decide the next market leader
What is the quiet crisis in engineering software? What does it mean for you? Read all about it here.
Read the full article →

Blog
How Quanscient enables the ultimate foundational physics simulation model
See the 4 features of Quanscient that can enable the next paradigm shift in physics-informed AI.
Read the full article →