Prove that MIS, done wrong (full factor with no window check), can be either safe - but-costly or actually wrong - and state which.
From PDVerse PrimeTime STA Interview Guide · pdVerse Mentor Guide
Ten-second Interview Answer
Applying the full MIS factor with no arrival-window check is the optimistic extreme for hold: MIS shortens min delay, so an unconditional speed-up can make data arrive earlier than reality and mask a genuine hold violation.
Complete Technical Explanation
Start from what MIS actually moves. Multiple-input switching applies a speed-up factor (less than 1) to a gate's delay when several inputs can switch together, and it is scoped to min delay - the hold quantity. The arrival-window filter gates that speed-up on whether the other inputs can really coincide with the input under analysis. Disable the filter and you apply the full speed-up factor unconditionally, which is the same as asserting that the other inputs always switch simultaneously. For hold, a larger min delay is the safe direction, and the window filter can reduce the speed-up where the inputs do not overlap, giving a larger and therefore safer min delay. Full factor with no window check gives the smallest min delay of all, so it is the optimistic and less safe extreme, and it can hide a real hold violation. That is worth stating plainly even when the option is described elsewhere as the more pessimistic or faster setting - that phrasing is about the setup-side or general delay perturbation and about runtime, not about the hold quantity MIS protects. The right reading is that disabling the overlap gating trades accuracy for runtime, and you must verify it does not optimistically clear real hold violations. The interview habit to voice: always tie the words pessimistic and optimistic to a specific check and the specific delay being moved. MIS moves min delay, so reason about hold.
In Pd / Signoff
You meet this when tuning MIS settings for hold signoff runtime on a large post-route run. Before accepting the faster, no-window-check configuration, re-run the hold report with the arrival-window filter active and compare min-delay slacks on the tightest endpoints. If any endpoint moves from passing to failing, the faster setting was clearing violations optimistically and you keep the overlap gating.
Common Trap
Reading a claim that an option is more pessimistic and applying it to whichever check happens to be in front of you. The full-factor MIS setting is more optimistic for hold specifically, because MIS reduces min delay - so quoting the general pessimism and then signing off hold on it is exactly the trap.
Expect Next
For a given gate, how would you demonstrate from the delay numbers alone that the arrival-window filter made the min delay larger rather than smaller?
Topic And Primetime Commands
Delay Calculation / Interconnect & Ceff. No specific PrimeTime command is required for this conceptual answer.
Keep building interview depth
Continue practising STA
Continue learning free
Get a practical low-power chapter
Receive the existing “Low Power and Multivoltage Fundamentals” PDF chapter and its download link by email.
Continue practising