AI Chip-Layout Tool Has Helped Design Over 100 Chips
Synopsys’ AI solution for semiconductor design hits milestone of 100 commercial chip tapeouts.the company said in a press release Customers such as STMicroelectronics and SK hynix are seeing chips with up to 25% lower total power consumption and significantly reduced die size thanks to Synopsys DSO.ai (Design Space Optimization AI), maximizing productivity 3x better.
To be clear, AI is not taking the job of chip designers. Thanks to our software, Synopsys human Chip designers and hardware engineers are freed from repetitive tasks and, thanks to AI augmentation, can instead focus on innovation.
“By reducing the design and verification cycle and effort, design teams can spend more time innovating on their core ideas,” said Synopsys. I hope that his AI will alleviate the shortage of engineers.
The milestone achieved by Synopsys shows that the use of AI in electronic design automation is fast becoming mainstream. Additionally, this AI could be particularly useful for industrial sectors (such as automakers) that are increasingly looking to get into the chip design business. Synopsys even describes DSO.ai as “the expert engineer in the box.”
So what exactly does Synopsys DSO.ai do? The biggest clue is in the acronym Design Space Optimization — by AI. This tool handles floorplanning for new chips (or iterations thereof). According to Synopsys, DSO.ai is ideal for trending multi-die silicon designs with high volumes of repetitive human planning tasks.
To complete that task, AI software optimizes power, performance, area (PPA) for a given chip design space. Working on PPAs is a proven route to doing better with fewer resources, and has become a very popular target for optimization in recent years due to the lack of key resources due to cryptocurrencies and the pandemic. increase.
Synopsys customers are reaping the benefits of DSO.ai with impressive results. Synopsys claims the customer has more than tripled his productivity, reduced power consumption by up to 15%, significantly reduced the die size of the finished design, and reduced resource usage. It also suggests that the ideal task for AI is to facilitate multiple foundry strategies to reduce the impact of supply chain vulnerabilities and reduce costs.
Synopsys is already looking to expand its use of AI in other areas. Chip design and verification workflow. It’s interesting to see this news as it looks like we’re going to see a breakthrough moment in AI-based chip design, with consumer AI from Google, Microsoft and OpenAI also making headlines.