EDA
System
IP
SolvNetPlus
Training & Education
In the semiconductor industry, solving one set of problems often leads to another.
We’re seeing this now with the shift from monolithic chips to 2.5D and 3D multi-die designs. These architectures help address major challenges related to performance, bandwidth, and energy efficiency — all of which are critical for AI systems and workloads.
But a new set of problems has emerged: multiphysics.
As logic, memory, and compute are packed more tightly together and stacked in closer proximity, they are increasingly influenced by thermal, electromagnetic, and other physical effects.
Historically, these effects were often treated as secondary concerns that could be isolated, analyzed, and addressed in the latter stages of design. More recently, as advanced node and multi-die designs gained momentum and multiphysics interactions became more pronounced — and harder to predict — engineering teams have built extra margins into their designs to hedge against uncertainty.
But that approach has introduced its own set of problems.
Multiphysics Fusion Technology for Multi-Die Designs Explained
Learn why multiphysics analysis must move earlier in the design flow.
The rising cost of overdesign
One of the industry’s traditional ways of coping with uncertainty has been to overdesign.
When engineers couldn’t fully predict how a chip would behave under real-world conditions, they compensated with extra margin. They added guardbands for timing. They built in safety buffers for power. They made conservative choices to avoid failure later.
At advanced nodes, however, overdesign is becoming too expensive.
Every added margin comes with a cost in power, performance, or area. Every conservative assumption can mean wasted silicon and lower efficiency. Those costs add up quickly.
The International Journal on Science and Technology (IJSAT) estimates the cost of overdesign — primarily driven by excessive design margins (or guardbands) to mitigate process, voltage, and temperature variations — accounts for 20% to 45% of total power penalties and 20% to 35% in wasted silicon area on advanced nodes (sub-5 nm).
These tradeoffs are even more pronounced in AI systems, where performance demands are intense and energy efficiency is critical. Teams cannot afford to leave optimization opportunities on the table simply because they don’t have enough visibility into the physical behavior of the design.
More analysis and correlation are needed. And they must come earlier.
Why multi-die designs raise the stakes
Multiphysics effects don't occur in isolation. In 2.5D and 3D multi-die designs, they interact — and the consequences compound.
- Thermal management: In stacked or tightly integrated designs, heat can become trapped between dies (also called chiplets), causing localized "hotspots" that can degrade or destroy the chip.
- Power integrity: Dropping voltages across through-silicon vias (TSVs) can cause power fluctuations and destabilize clock speeds.
- Electromagnetic interference: When high-speed signals travel through dense structures, interference can cause data corruption and crosstalk.
- Mechanical stress: Stacking different materials with different thermal expansion rates can cause structural warpage, leading to cracking or delamination.
This is why the industry’s “shift left” mindset matters so much. If teams wait too long to evaluate these interactions, they are more likely to face late-stage surprises, repeated fixes, and costly re-spins.
But if they can see those interactions earlier, they can make better architectural and design decisions — before complexity hardens into risk.
A more integrated approach to chip design
At Synopsys, we believe this shift calls for a more integrated approach. One that brings multiphysics awareness and true co-design directly into the workflow, instead of treating it as a downstream activity.
The first Synopsys Multiphysics Fusion solutions, which combine Synopsys’ industry-leading silicon design tools with the gold-standard signoff analysis capabilities from Ansys, reflect that new approach. Now available for production use, the solutions span five key areas: timing signoff, design closure, multi-die designs, analog and mixed-signal (AMS) design, and photonics design.
What connects them is not simply more analysis, but better context and stronger correlation within the workflows engineers already use to design and validate complex systems.
For timing signoff, that means understanding how real-world power and thermal conditions influence timing behavior. For design closure, it means resolving physical issues with greater confidence and fewer iterations. For multi-die design, it means gaining earlier visibility into system-level interactions. For AMS design, it means bringing electromagnetic awareness closer to the point where key decisions are made. And for photonics design, it means eliminating manual design handoffs and accelerating system development.
These are different workflows, but the underlying challenge is the same: understanding how multiple physical domains interact and influence one another.
Co-designing with physics, not around it
The semiconductor industry has long treated physics as something to validate at the end of the design process. That's no longer sustainable.
As multi-die architectures grow more complex and the costs of overdesign become more pronounced, physics must become a core design input. One that shapes tradeoffs, improves correlation, and reduces unnecessary margin from the start.
The teams that make this shift won't just avoid late-stage surprises. They’ll make more confident architectural decisions, reclaim the performance and efficiency that conservative assumptions leave behind, and move faster in a market that rewards speed.
Continue Reading
Breaking the Design Closure Bottleneck with Timing-Aware Multiphysics ECO
Taming Advanced Node Physical Verification – NVIDIA’s Perspective
