As you may or may not already know, the Fallout creator Tim Cain has a popular channel on YouTube where he shares everything from anecdotes about his career to specific insights into game design and other aspects of game development.
For those who grew up playing some of the legendary RPGs he made at Interplay, Troika, and Obsidian, it's truly a treasure trove of insights and information, made even greater by the regularity and concise nature of his uploads. Typically, each video is dedicated to a single topic, with the idea being to give viewers an insight into the development and design process behind making games in easily digestible chunks, which usually clock between 10-15 minutes long.
In yesterday's video, for example, Cain decided to delve into a topic, which we're betting will be relatable for a lot of players: why escort missions in RPGs are rarely ever done well, and why some of the solutions people come up with to "fix" them are flawed. This video actually builds off another one he uploaded last month, where he previously called escort missions an RPG feature "nobody asked for".
In that video, he outlined the frustration that many players will have experienced of having to constantly backtrack to find an NPC they'd been asked to accompany, the impossibility of regulating their speed to keep level with the character, and the disappointment of failing missions due to straying too far away from the target.
He also said he found it "hard to believe that the designer who made it had never seen an escort quest and wasn't annoyed" and that the only reason he felt they were included was either the developer had "never played an RPG in their entire life, or they somehow thought this escort quest would be fun. And they'd be wrong".
In response to that, people ended up inundating his channel with comments, all suggesting the same solution — make the NPC a temporary companion. So Cain has put out another video poking some holes in that suggestion, and showing why fixing that fix is a little more complicated than most people expect.
In this recent video, Cain began once again by acknowledging the poor reputation of escort quests, reiterating that “I’ve never seen a review asking for more escort quests" before going on to dissect this "fix" most people had been suggesting to him.
He stated that by letting NPCs become companions, it opens up a potential can of worms for developers that most of these comments hadn't considered, particularly surrounding how the NPCs will act (can they attack, give items, etc like a regular companion), how they will potentially impact the game's companion limit (can they exceed the number of companions or not), and how long they stay in your party as a companion.
"Let’s say you say when she (the escort NPC) joins the group as a new companion, she can be in there forever," said Cain. "It doesn’t matter how long it takes you to get to the destination. Well, guess what, problem one — this is no longer an escort quest. This is just a companion acquisition quest. All you’ve done is you’ve added a companion with an exit condition." This then becomes another issue the developer will have to solve.
How does Cain know this? Well, this is a solution he actually attempted 28 years ago on the original Fallout, with the NPC Tandi from Shady Sands, and, in his words, ended up further exacerbating some of the issues with the AI in that game.
Cain recalled, "When you rescue Tandi from the Raiders, she said, ‘Oh, can you please take me home to Shady Sands?’ But for a lot of players, if you’ve done all the quests in Shady Sands, you have no reason to return there, and you quickly discover that if you never return there, you have a free new companion, Tandi. So what happened in QA is a lot of people came back to me in QA and said, ‘I never take Tandi back to Shady Sands. She’s just an extra companion.'
"So this solution you came up with was one we came up with 28 years ago, and we really didn’t solve anything. Instead, we ended up with more companions following the player than we ever planned for, and that caused other issues. Google any reviews of Fallout, and there's a lot of issues with companion AI, and every new companion you got just made those issues worse."
Playing devil's advocate, he then suggested, "Some of you may go, ‘Oh no, I’ll do it the other way. I won’t have them stay in your party forever. If you agree to escort someone, there’s now a timer’. Well, congratulations you’ve just added a timed quest to your game — one of the most single-hated quest types in all of RPGdom, and you just did it. You fixed an escort quest, which is disliked by a lot of people by converting it into a timed quest, which is hated by almost everybody."
The point, we believe Cain is trying to get across here isn't just that escort missions are mostly bad, but that game design and game development is often a series of choices, and those choices often reverberate throughout a production. What may seem like a "good solution" at first glance could introduce a whole bunch of other things you have to think about further down the line, and it's a sign of a good designer to be cognizant of these issues, or at least open to them, and to figure out what fits best for your game and whether a feature is really worth all the hassle.
At the end of the video, he stated, "You’re going to have to pick the solution you hate the least, or that you think will annoy your player base the least, or you think fits best with your design pillars, including quest pillars and exploration pillars...
"Unless you just don’t do escort quests. The same as you don't have to put in timed quests, you don't have to put in escort quests. You can just always leave them out. So, unless you do them well, maybe you shouldn't do them at all."