Spec-Driven Modeling Automation Platform


SDEP (abbreviation of Spec-Driven Extraction Platform) - an innovative modeling platform, providing powerful & rich APIs to build automatic and reusable modeling flows. SDEP enables a new way to shorten the model development turnaround time dramatically. It also provides customers with a system to retain and inherit valuable device modeling expertise and solidify modeling know-hows.
SDEP integrates Primarius'latest technologies in all modeling essential functions covering data analysis, model parameter extraction and model quality checking. It gives user a new powerful weapon & platform to tackle modeling challenges. With the flexible GUI environment and flow control functions, modeling engineers can easily establish a flow for model extraction procedures, routine modeling works, new ideas, etc. These flows on top of SDEP can be naturally automated, reused, and customized.
SDEP can also be used as a modeling expert system to understand model, accumulate modeling know-hows, and train modeling engineers.


Key Advantage

Reduce impact of employee turnover
Deeper understanding of model behaviors
Better balance strategies between fitting & QA
Easier to try-out new modeling strategies & ideas
No programming skill needed to build up a complicated flow
A new way & system to solidify & accumulate modeling know-hows
Better parameter selection strategies with parameter auto-filtering function
Easier to build up a flow for modeling procedures or routine works to shorten model development TAT


  • High resolution trend control
  • Support irregular device map
  • Support data analysis & validation
  • Rule-based device & bias selection
  • Fitting & QA spec co-optimizations
  • On-the-fly model QA with rich checking rules
  • Flow variable, control & debugging features
  • Flow configuration with editable API properties
  • Easy model format (global, local & bin) conversions
  • Flexible GUI environment & powerful data presentation functions
  • Customizable optimization options with local & global algorithms
  • Parameter sensitivity analysis, smoothing, boundary & effective value control
  • Complete modeling functions covering IV/CV/Temperature best-fitting, re-targeting, corner and LDE