Cloud native EDA tools & pre-optimized hardware platforms
Designing complex, mixed-signal IPs is extremely challenging. Traditionally, timing is verified using measurements extracted from dynamic simulations. However, as this is vector-dependent, it relies heavily on designer expertise to select the correct set of critical paths. Often, this approach does not cover all the necessary timing checks across all operational modes and process corners. Failing to check the fastest and slowest paths in the design can lead to silicon failures. This webinar discusses a timing signoff methodology that uses transistor-level static timing analysis to augment dynamic simulation. This methodology performs validation for all timing checks (I/O timing, internal timing) including signal integrity effects (crosstalk delay and noise) and parametric on-chip variation (POCV). It discusses how static timing can be used to quickly create block-level timing models (.lib) so that AMS IP blocks can be used by a digital implementation flow.