RADIANT QUESTIONS: IS THERE A RIGHT WAY TO DO THEM?

I can't give a full explanation right now, but I dont feel it's a good idea
While I can obviously just ignore that aspect in a AAA game like FO4, it really killed immersion for me. It's not a bad idea by any means, it just killed it for me for that game. I didnt feel like it fit. I think it can be done right though.
It needs more variety and world sense to be engaging.

For example, I didn't like that a group of mutants/raiders was terrorizing a settlement that was like half the entire map away. It just didn't make sense. I'd rather have unexplained waves of X attack X place instead of trying to make sense of why X group from X place would attack X settlement, if that makes sense.

Yeah, I think they work fine with limitations. For example, in an MMORPG or a mobile game, you might have quests like this, but the player can only do one per day. In a more traditional single-player game for PC or console, you might make the quests' reward be a resource that is only used for upgrading the blacksmith, so once the blacksmith is fully upgraded, the player is done with them. Or you might make the randomized quests cost a resource to perform, so players need to continue doing other activities to get enough resources.

They are definitely filler content, which isn't the best, but is often unavoidable, especially if you want to make a game of unlimited length. Random battles are also filler content. Caves leading from one location to another are filler content. Ingredients for crafting systems are filler content. Filler content gets a bad rap as a way to artificially make a game take longer, but a lot of games need a certain amount of filler content for pacing purposes; it lends more importance to the parts of the game that aren't filler, and lets the player build up their readiness. It lets you create an arc with a finale instead of just a single event. Randomizing the filler content is obviously inferior to hand-crafting every bit of it, but the only way to keep hand-crafting content for as long as the player wants to keep doing the same thing is to run a tabletop game.

By the way, the term "radiant quests" is specific to Fallout 4, but this idea is far older than Fallout 4. The more generic term is "procedurally generated quests" or simply "randomized quests."

I found a thread on Reddit about Fallout 4's randomized quests, and I'll also post the top comment from that thread here, because I think it's important to keep in mind:

author=Favourite
The worst thing about radiant quests it that if you're not familiar with the concept (as I wasn't when I first played), you don't realise what's happening at first, and feel cheated once you do figure it out.

Example: The first time you happen upon the BoS, early in the game, you can be handed a few radiant quests. I didn't realise these were just infinitely repeatable until I'd gone through them three times. I thought I'd at least be building up affinity with the faction, or trust with the stranger-wary quest giver, or eventually happen on something unexpected, but no, nothing changes. You just go through the same dialogue over and over until you realise the game has just deceived you into looting the same places over and over to derail you from completing actual content.

This is a completely solvable problem, though, if you're simply transparent with the player in explaining how your game works. Bethesda often doesn't want to explain anything about how to play their games, because they're convinced that "it breaks immersion" to explain that the world is made up of gameplay systems, but I'm pretty sure turning off the game out of frustration at those systems breaks immersion even more.

bahleted

@Liberty: I mean, they gave XP and gold and loot. As much as any story quest. The only thing they didn't give is a sense of satisfaction and completion upon finishing them.

This is the big problem, really. It's not the lack of story, because these kinds of quests feel unsatisfying even in games with no story. It's the lack of completion, like Zephyr says. When you beat them you haven't accomplished a goal. If they're done particularly poorly, then you haven't even made progress towards a goal. The same goal is still there when you finish, and it's still 0% completed because the mission has started over with new objectives. I agree with everything Zephyr said.

I would conjecture that the solution would be a balance of intricacy in the formula and variety in the quest types. Maybe even a variance in the source. For instance say you can go to a Pub to get work, but it randomizes what people are offering what quests per day/completed quest/rest etc. Put more depth into the story, and contrast these quests with more intimately created ones, both within the same game. So you have the pub option when you want more money for that equipment, or a level up, or to develop a skill tree, but you can move forward with story quests when you want to also.

I sometimes think that planning and developing a system like radiant questions takes more time and creativity than just typing out 100+ different things of text. And in games, text doesn't take up that much space, so it's not like a system that creates generic text will save memory, it'll just mean you spent your time writing code words instead of writing dialogue words.

The problem I always seem to run into with these types of quests is that the algorithm for them always seems to be painfully obvious after doing just four or five of them. They seem to immediately feel like filler, and you can only arrange the bricks that put them together in so many ways before players see the man behind the curtain.

What you can try doing is make them more complex and more rewarding as the player completes more of them. Separate them into tiers, and give a very good incentive to do continue doing them that they can't get through any other method.