The tool that optimises timing during implementation is rarely the one that signs the chip off. The two use different view lists, different parasitics, and sometimes different delay calculators. A path that passes in one can fail in the other. This topic covers the differences, how to set a margin from measured correlation, the usual causes of mismatch, and what must be identical when work is handed from one tool to the other.
Lesson | After it you can... |
|---|---|
List what differs and count the endpoints that escape. | |
Choose an implementation margin from data. | |
Find which delay component explains a gap. | |
Compare two setups and list every difference. |
The two use different views, parasitics and calculators, so their delays differ; an endpoint close to zero slack can fall on either side.