BeginnerQuestion 30 of 40Source PDF page 27

What is the difference between an ideal clock and a propagated clock?

From PDVerse STA Mentor Guide · pdVerse Mentor Guide

Short Answer

An ideal clock is modeled as reaching every flop instantly, or with just a fixed, uniform latency number, before any real clock tree/network has even been built. A propagated clock is timed through the actual, built clock network, capturing real insertion delay and real skew between different flops.

Example

Early in the design flow (RTL, synthesis, placement stages), STA runs with an ideal clock and estimated wires. Later, once clock tree synthesis (CTS) has built the real clock distribution network, STA can use propagated, real clocks.

Why It Matters

Ideal clocks are a convenient simplification for early-stage analysis when the physical clock network doesn't exist yet. But real clock trees have wire delay and branching, so different flops receive the clock signal at slightly different times (skew) and after some delay (insertion delay). Propagated clock analysis is what reflects the truth the chip will actually see.

Mentor Note — Common Mistake

Don't rely on ideal-clock results as your final signoff answer. Once the clock tree exists, you need propagated clock timing to see real skew and insertion delay effects, which ideal modeling completely hides.

Visual explanationIdeal clock versus propagated clock
Ideal clock versus propagated clockAn ideal clock has no modeled tree delay, while a propagated clock accumulates buffer and interconnect delay to the sink.Ideal clockFFZero modeled tree delayPropagated clockBUFBUFFFActual buffer + wire insertion delay

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.