How do you compare timing between two floorplan alternatives?
From PDVerse PnR Interview Handbook · pdVerse Mentor Guide
Short Answer
Use the same netlist, SDC, scenarios, libraries, pre-CTS clock assumptions, coarse placement, and estimation method. Compare the same named paths and physical distances.
Technical Explanation
A slack difference is meaningful only when the analysis context is controlled. Inspect register-to-register and I/O paths, macro pin sides, detours, congestion, feedthroughs, hierarchy crossings, and buffering opportunity.
Formula Or Decision Rule
Decision rule: attribute the delta to a physical cause; do not claim signoff closure.
What To Check
• Warning sign: One alternative reports better WNS but used another scenario or excluded a path. • Inspect: choose one affected region, macro, row, pin, path, or net and trace the physical cause. • Correct: Align contexts, rerun estimate_timing, and compare full path reports and congestion together.
Command Checks & Actions
• estimate_timing: performs early virtual timing estimation
• report_timing -path_type full_clock_expanded: reports timing paths in the active analysis context
Run only the commands needed for this question. Save the report with the floorplan version and analysis context.Healthy, Suspicious & Hard-stop Results
• Healthy: Use the same netlist, SDC, scenarios, libraries, pre-CTS clock assumptions, coarse placement, and estimation method. Compare the same named paths and physical distances. • Suspicious: One alternative reports better WNS but used another scenario or excluded a path. • Hard stop: required legality or physical feasibility is missing or unexplained. Align contexts, rerun estimate_timing, and compare full path reports and congestion together.
Common Mistake
Do not hide the symptom with an arbitrary utilisation, halo, channel, blockage, pin move, or die-size change. One alternative reports better WNS but used another scenario or excluded a path.
What The Interviewer Is Testing
The interviewer is testing whether you can connect “How do you compare timing between two floorplan alternatives?” to measurable physical evidence and an owned correction.
Follow-up Question & Model Response
Follow-up probe: What would make you change your decision? Model response: “I would change it if the same controlled rerun shows that one alternative reports better WNS but used another scenario or excluded a path. I would use the named objects and reports to decide whether to iterate, waive, or return the issue.”
Practical Example
Symptom: One alternative reports better WNS but used another scenario or excluded a path. Corrective path: Align contexts, rerun estimate_timing, and compare full path reports and congestion together.
Continue practising