recreation design – Reasoning about quest and story deadlocks and so forth

Are there frameworks for reasoning about questions resembling: Given my quest and degree design, is it assured that for any mixture of participant decisions there isn’t a method for the participant to lock themselves out of finishing the sport?

For instance, assume a quest like this:

• The Participant can betray NPC A and obtain merchandise b from NPC B as a reward, or
• The Participant can betray NPC B and obtain merchandise a from NPC A as a reward.
• The Participant can commerce both merchandise a or b for a key. That is the one approach to acquire key.
• Solely the key opens a brand new zone. That is the one approach to enter zone.
• The participant can kill NPC A. Upon loss of life, NPC A doesn’t drop merchandise a.

With this setup, the participant can lock themselves out of finishing the sport in the event that they kill NPC A and betray NPC B. On this case, they will neither get merchandise a nor b as a reward, due to this fact by no means acquire key, and by no means enter zone.

Now take a look at these two further guidelines:

1. If killed, NPC A drops merchandise a.
2. The one approach to kill NPC A is thru an merchandise that may solely be obtained from zone.

Including both of those choices / constraints ensures that the participant can by no means lock themselves out of finishing the sport.

For those who wanted some time to learn by these guidelines, then it signifies that it’s no less than a bit tough to cause about them, though that is solely a really small instance. For any life like variety of choices, constraints, NPCs, gadgets, and quest branches, likelihood is excessive that we overlook a possible path through which the participant results in probably the most irritating scenario: being unable to finish the sport, and doubtlessly discovering this out solely very late.

How do folks cause about such issues? I can think about encoding some helpful info into my quest descriptors, gadgets, and so forth., turning all of it into some form of big state machine and computationally validating that there aren’t any useless ends. Or I can flip all of the choices and constraints into Prolog clauses and test for satisfyability. Each appears doable however costly to construct only for one recreation.

How do studios do that? Are there current frameworks? Limitless QA and hoping for the very best? Leaving it to beta testers?