Semiconductor design software maker Synopsys Inc. today announced an expanded suite of artificial intelligence tools to aid in the design, verification and testing of advanced computer chips.

With Synopsys.ai, engineers can now benefit from AI assistance at every stage of the chip design process, from system architecture to design and manufacturing, the company said.

Synopsys is a pioneer of electronic design automation, providing the software, hardware and services required by chipmakers to define, plan, design, implement, verify and then manufacture next-generation semiconductors.

Today’s chipmakers are finding themselves under increasing pressure to come up with ever-more powerful chips to accelerate advances in high-performance computing, analytics and AI. AI especially is powering incredible advancements such as human-like chatbots, robots that can perform surgery and self-driving cars. It’s one of the main technologies driving the demand for greater chip complexity, and it can also be harnessed to help deliver those more powerful chips.

Synopsys says its Synopsys.ai EDA suite can help to accelerate chip design with solutions for functional verification and silicon testing, with more capabilities to be added later. In a blog post, Arvind Narayanan, Synopsys EDA Group’s senior director of product line management, explained how AI can help to tackle many of the most time-consuming iterative tasks required to build higher-quality chips.

“AI technologies known as machine learning and reinforcement learning can take on repetitive tasks such as design space exploration, verification coverage and regression analytics, and test program generation,” Narayanan said. “This frees up a substantial amount of time for engineering teams, allowing these experts to focus on value-added tasks such as differentiating their products and quickly creating new features or derivative designs.”

Synopsys.ai has three main components, including DSO.ai for automatic design space exploration, which is the process of improving power, performance and area of chip designs. VSO.ai is used to help verification engineers reach their coverage closure targets more rapidly and find bugs in their designs.

AI is an obvious advantage here, Synopsys says, because the number of design state spaces in which a digital chip design can operate is almost infinite. That means it’s impossible for humans to verify it will always function as intended.

“The regression process could run for days, eating up compute resources through thousands of tests,” Narayanan said. “Often, the ‘last mile’ of closure ends up being very labor intensive, with manual analysis on huge amounts of data limited in yielding actionable insights. VSO.ai revitalizes this process, examining the RTL to infer coverage while also highlighting areas where coverage is needed, saving substantial time and ensuring a high ROI on the tests.”

The last component is TSO.ai, which helps to automate the silicon test process, assessing a new chip design’s defect coverage, pattern count and runtime using automatic test pattern generation tools.

Synopsys EDA Group General Manager Shankar Krishnamoorthy said the challenges of increased complexity, engineering resource constraints and tighter delivery windows were crying out for an AI-driven EDA software stack. “With Synopsys.ai technology, our customers’ ability to search design solution spaces across multiple domains is in hyperdrive,” he said. “They’re finding optimal results far faster as the AI learns run-to-run and it’s transforming their ability to meet and beat tough design and productivity targets.”

The company revealed that the Synopsys.ai suite has already been deployed by nine of the world’s top 10 semiconductor firms, and that those customers have reported seeing a 10-fold improvement in verification productivity and far lower silicon testing costs. That has allowed them to shave weeks off of their product development times, while also reducing their costs.

Analyst Patrick Moorhead of Moor Insights & Strategy said AI has transformed the semiconductor design industry, helping engineers create more advanced chips that humans would never be able to produce unaided. “The horizons AI will open up are hard to imagine, only we know we’ll be able to move faster and do more than we can now,” he said. “Synopsys is now clearly taking the lead in infusing AI throughout the chip development flow, and we should applaud their investment in the industry’s future.”

Image: Synopsys