The timing foundations gave you the graph and the two inequalities. This section turns them into working practice: the propagation algorithm (with a toy engine you can run), a breakdown of where path delay goes, a way to rank failing paths, and a line-by-line reading of path reports, which is where most daily timing work happens.
Lesson | After it you can... |
|---|---|
Compute slack at every node. | |
Explain why STA scales linearly. | |
Find shared-segment fixes. | |
Read and extend a working engine. | |
Split a path into cells and wires. | |
Predict upsizing effects. | |
Explain estimate-to-routed differences. | |
Assign stage budgets. | |
Explain grouping. | |
Read per-group summaries. | |
Rank work by impact. | |
Track progress honestly. | |
Recompute a report's slack. | |
Read latency, uncertainty and credit. | |
Read increments and a hold report. | |
Avoid the eight classic mistakes. |
Suggested order: top to bottom. Run every Python block and change a number; the engine in the last lesson of the first topic is the thing to modify. Previous sections: Timing Paths and Checks and Constraints and SDC.
The header (path type, group, view), the slack line, and then the clock sections to see skew, uncertainty and credit before the data path increments.