Name the engine that proves the lab's invariant at 16 bits within its budget.
BDD reachability computes one image per iteration until the fixpoint, which comes one iteration after the last state is first reached. So the number of iterations depends on how far the farthest reachable state lies from the initial ones, not on the number of state bits. IC3 and PDR compute no images: they look for an inductive invariant directly.
The 3-bit LFSR of the Workbench's Reach examples visits its 7 nonzero codes one after another: one new state per iteration, and the fixpoint at iteration 7.
The mutual-exclusion example has 16 reachable states, but several are first reached at the same iteration: its fixpoint is at iteration 6.
So the LFSR needs as many iterations as it has states, while the protocol needs far fewer.
The shape of the reachable states, not their number alone, sets how many images reachability computes.
The lab's budgets: in Explore reachable states, Run to the fixpoint computes at most 256 images each time it is pressed, and the proof runner gives a proof 60 seconds.
At 16 bits, which engine proves the invariant within one use of its budget?
abc pdr
Explore reachable states
Both
Neither
Choose one engine, both or neither.
One engine, both or neither.
Count the 16-bit counter's reachable states, and how many of them each iteration adds, as the example does for the LFSR.
Compare that number of iterations with the 256 images that one press of Run to the fixpoint computes.
Run abc pdr at 16 bits in the lab, and read its verdict and its time.
Kata 8.2 of Binary Decision Diagrams with CUDD on formal.org. The site checks the answer when you press Check answer; sign in to keep your progress.