Magma Announces FineSim Fast Monte Carlo

Most engineers today rely heavily on statistical methods such as traditional Monte Carlo analysis for design reliability, an approach that has limitations making accurate analysis almost impossible. FineSim Fast Monte Carlo uses proprietary dynamic error-controlled algorithms along with sophisticated statistical techniques to provide dramatic improvement in speed and accuracy compared to traditional Monte Carlo statistical analysis. It has shown as much as 100 times better runtime when compared to other commercial methods, with superior accuracy.

“As engineers continue to push their products into more advanced process technologies, the need for improved reliability becomes even more critical. But the increasing number of process parameters and parasitic variability make it virtually impossible to predict accurate statistical yield analysis, reliability and failure analysis,” said Anirudh Devgan, general manager of Magma's Custom Design Business Unit. “FineSim Fast Monte Carlo was developed to address these issues with greater accuracy and predictability.”

The algorithms in FineSim Fast Monte Carlo, coupled with Magma’s native-parallel technology, improve throughput for statistical simulations even further. FineSim Fast Monte Carlo makes statistical analysis practical for many different applications. “Several key customers have tested FineSim Fast Monte Carlo option on their memory, custom digital, and mixed-signal designs, and they are very excited about its performance and accuracy of results,” Devgan said.

Designers can use FineSim Fast Monte Carlo to dramatically improve the speed of statistical analysis, enabling statistical analysis on designs that would previously have been infeasible. Fast Monte Carlo can also improve the accuracy of statistical analysis on a given number of statistical samples, providing users improved overall confidence in the analysis.