Part 12 answers: How do we spend verification effort wisely?
Every bug has a cost to find and a cost to miss, and both change over the project. Early, bugs are cheap to find and fix; late, each new bug costs more to find and a missed one costs far more. Treating verification as an investment changes what a team does next.
This part covers the cost curve, bug-discovery curves and what their shape means, diminishing returns and when to change technique, triage and severity, fix quality, estimating how many bugs remain, and where infrastructure investment pays back.
Read a bug-discovery curve and say what it predicts.
Decide when a technique has reached diminishing returns.
Estimate the number of remaining bugs and its uncertainty.
Answer each in your head before opening it. If two or more feel shaky, revisit Part 11: Post-Silicon Validation first.
Verification samples a vast space; some bugs need conditions (long runs, software, electrical effects) no pre-silicon platform reproduced.
A validation plan, observability (trace, counters, scan dump), tests and software ready to run, and a triage process.
When the bug can be avoided reliably at acceptable cost and the risk of a missed case is low; otherwise respin.
# | Lesson | Level | Status |
|---|---|---|---|
12.1 | W · Working | Ready | |
12.2 | S · Senior | Ready | |
12.3 | S · Senior | Ready | |
12.4 | W · Working | Ready | |
12.5 | S · Senior | Ready | |
12.6 | P · Principal | Ready | |
12.7 | P · Principal | Ready | |
— | all levels | Ready |
Every lesson in this part is ready. Work through them in order: each one sets up the problem the next one solves.
This course teaches the decisions behind verification. These lessons teach the tools and syntax that carry them out, so this part links to them instead of repeating them.