GrammaTech, Inc., a leading manufacturer of source-code analysis tools, today announced CodeSonar 3.8. The new version includes dramatically faster code analysis with fewer false positives.

“CodeSonar 3.8 is significantly faster and more precise than its predecessor, making it much easier to analyze projects with millions of lines of code. The speedup was achieved by parallelizing the analysis engine so it takes full advantage of multi-core processors. The improvement is dramatic, especially for large projects. For example, on an 8-core machine, we have seen analysis times reduced by 85 percent,” said Paul Anderson, VP of Engineering at GrammaTech.

Improvements in CodeSonar 3.8 include:
• Faster analysis that takes full advantage of multi-core processors.
• Lower false positive rate due to a combination of new models for C/C++ libraries (e.g., Boost) and improvements to the analysis.
• Workflow enhancements that make it easier for developers to manage defect warnings across multiple project versions and development branches.
• Better checking for concurrency-related errors.
• Addition of award-winning software architecture visualization.

This release is the first version to incorporate CodeSonar’s software architecture visualization, which won the Most Significant New Software Product award at Design West 2012. CodeSonar architecture visualization provides developers with insight into the relationships between elements in software and scales to large projects with millions of lines of source code. Fluid transitions allow developers to visualize massive data sets in real time and to see the program at different levels of abstraction. CodeSonar visualization also includes other program-understanding and navigation features, and supports annotations and sharing of diagrams between team members.

“Improvements to the analysis engine empower developers to pinpoint defects faster and with greater precision,” said Anderson. “Improvements to the user interface make it easy for developers to understand and analyze very large projects, including those developed by complex software-development organizations.”