TITLE RAN OUT OF CHARACTERS
was supposed to say "stay fresh" not FREE
My idea is in two parts: Part one is the idea, part two is the tool to make it possible.
We all know that once a quest is discovered, a "walk through" and checklist will get posted online so players can follow along and do the quest procedurally. Heck, there are wiki pages, encyclopedias, etcera all dedvoted to such informaiton sharing. The problem is that takes away some of the spirit of discovery and those also accelerate the rate of player advancement.
Idea 1) Have bigger quests, important quests, class quests, epic quests, etcera change. Don't let the same quest just be exactly the same. So when one person goes to a person and gets a quest task it might be different than the person before them or (at least) a month between.
The changes could be small like a changing turn-in item or a turn-in locaiton or and turn-in NPC or a phrase to trigger something or even a mob to kill.
Not only would this make players still figure out some of the quest themselves, and maintain some sense of discovery and wonder and it will also slow down the pace of everyone finishing it and racing to the next content.
Finally this would make the quests feel more personalized because we all aren't repeating the exact same things and it's more convincing that the quest giver is giving us tasks just for us.
Obviously the big tasks like epic fights, important drops etc have to be the same....I just mean the "filler" steps.
Idea 2) Developers create tools that help them create the worlds. Like placing vegetation, rocks, etc. And there is some "AI" that goes into patterns of placing things. So, creating a tool can help tremendously.
So a tool that is a quest manager, like a terrain populator tool.
A tool that looks at a quest and applies some AI to it, so maybe a dark elf isn't assigned the killing of a dark elf but someone else. Or maybe it randomly assigns some monsters to kill or maybe it even connects to a different NPC for a turn-in....maybe based on some input parameters and sometimes just randomly. So yes we all have to turn in a drop to the halfling NPC but maybe it could be any one of a hundred halfling NPC's and we actually have to follow the quest to figure out which to do.
I really like the idea of "smarter" quests, mobs, raids, etc. Random is good and it makes perfect sense. But I'd like whatever is created to be created for it's own purpose, not with the idea of trying to avoid people getting spoilers online. It needs to be organic and make sense in the world. People are going to get spoilers, nothing we can do about that except choose not to do it ourselves. Having a thousand other people complete a quest on spoilers shouldn't take away the fun I have doing it without spoilers.
But a big fat yes to your ideas.
Basically Mad Libs but for quests with weighted relationship tables. Each quest has a core sequence of steps but some/many/all nouns can be table driven. When you get the quests it fills in the flag slot. Tie in some form of proximity spawn on successful perception check and you could have real world quest clues or items spawn forcing players to read and think.
The trouble with randomization is the degree to which it has to be randomized to avoid blatant pattern recognition, and that's before entering into the realm of datamining.
Using Sea of Thieves as an example, both the Tall Tales and 'riddle' maps utilize degrees of randomization in their objectives. I can pick up the same quest as someone in another crew and outside of a couple of key POIs, we may be sent on very different routes. However even though the objectives are randomly chosen, they are chosen from a set list of possibilities, which means that after so much time players understand what all those possible combinations are. E.g. if a riddle map can vary between 'light a lantern at the broken rowboat the southeast', 'play a tune at the remains of the fisherman on the coast', or 'read this map at the painting of the mermaid', you eventually still get a guide that says your map can be one of these three things, and here is where to find each.
This is also ignoring the issue of parity between objectives; using SoT again as an example, I once started a tall tale alongside some friends and by the end of it, I was able to find my final objective in a few minutes while six of us scoured our friend's island for an hour and a half before being able to solve theirs.
Now for a sandbox game where the emphasis is more on replayability, this isn't as big of a deal because it's just keeping a semi-static content type fresh by changing it up each time, not unlike randomizing the location or targets of SWG missions. But when we're talking about one-and-done quests for the sake of obtaining a specific item or chunk of exp, I'm not sure such randomization would be worth the implementation cost. At the end of the day the players that want the mystery will keep the mystery, and those that don't will have a plethora of resources to drive them to those answers. See: Animal Crossing purists vs. Time-Travellers.
Wrote this up in 2013 regarding a similar idea for Shroud. (not the unique skill idea, the unique path part)
Great idea fazool, I hope they do something like it, eventually, but there's been no indication this is a public design goal in the past 6 years.
I like the energy and the idea of this but I think.... In my opinion..... I will have to pass... I have always found it nice to find groups who have the same quests and you can run around and do them together... and as far as people looking a quest up online... I say so be it...if they need help or just trying to power through it... its up to them...
Again I do like the idea but just not for me...
I agree with Nytman,
The idea discourages people from working together to do the same quest. I don't see them implementing anything like this. Not to mention any new ideas would have to start from scratch and we are years beyond the point of having time for that.
I assume this is pure fantasy blue sky hyperbole from the OP. If not then the following surely applies.
There is no AI.
There are no 'developers' with super powers writing super code to magically create amazing worlds.
There are basic tools, mechanics, languages being used to create the game and the worlds. The same as every studio, every code jockey for every game are using.
Nobody has 'magical AI developer' in their resume from a previous job. Nor does VR have access to a quantum computer.
Some studios do write their own engines. But they are still based on an underlying language. Most commonly C++. Then maybe C#, assembly, Python, Java, etc. VR's engine in Unity. It is a known entity with known functionalities. Experienced programmers with the right skill sets can extend Unity to do lots fo fun and creative things, but not dynamically write it's own code. That's some serious Skynet stuff.
Quests, if we every get them, will be procedural and hard scripted. Most probably with a reliable, proven language simple enough for low level programmers to use.
Frankly it's hard enough for programmers to code in all the basic permutations of what a players qualifications to even do a quest are. Much less create dozens, hundreds of random variations.
In ~1997 LUA was being used by 989 Studios to script quests for EQ1. Hey quess what. LUA is still being used by games today.
Unless VR has employed the likes of Linus Torvalds, and they are working madly on some new underlying fundamental magic game language, VR will be using time-tested, common tools and scripting. The ones their employees were hired for knowing.
Nytman said:I like the energy and the idea of this but I think.... In my opinion..... I will have to pass... I have always found it nice to find groups who have the same quests and you can run around and do them together... and as far as people looking a quest up online... I say so be it...if they need help or just trying to power through it... its up to them...
Again I do like the idea but just not for me...
Good point, I didn't think of that. I also really enjoy doing quests with friends (or random people). Hmmm
It's worth noting it's possible to do all of what the OP described, today, in Unity (and many other engines, environments, and languages).
It's also been possible in games in general since (at least) Rogue. It doesn't require anything particularly new or taxing for any PC or Server.
You can even write your own in less than a day, via this tutorial, for free, in Python. (version 3 here)
Also, the "problem" of whether or not to help someone on a quest you don't have isn't an issue with this or any other questing system, that's a human nature issue. :)
At least if you know up front that all quests between milestones will be personalized, you can add in carrots to strongly encourage the emergent behavior of helping others.
But really, it's not like anyone else in your guild (or on your server, if you were really generous) directly benefited from helping someone with their Epic. They helped either because of an expectation of help in the future, or actual altruism.
Yeah as vjek said there is not much that the fazool suggested that could not be easily done with current tech.
You start by making say 20 different templated sub tasks. Each subtask has its own variable goals that draw from local based relationship tables. How you choose to complete the sub-task can then alter the relationship table of the next sub task in the quest.
You don’t need full AI or machine learning to have adaptive or semi random quests. You just need to design with that goal in mind from the beginning then it is literally just data entry that modifies the text, action flags and reaction scripts. I imagine one could actually make a fairly robust and immersive decision table to make 10,000s of quest permutations faster than it would take to write and hard script even 24 epic quests.
Now the feel of these quests would be much more an episodic short adventure rather than a deep lore or long class quest but they could fill in the gap that other games stuffed quest hubs into. As far as having other people help you I would design the quests to be only possible to start with a full group and set a 3 hour time limit on them (to set the expectation of how long the group will need to be together and how many steps the quest can have).
This is actually a system that could easily be added on after the base adventure portion of the game is established. The perception system actually already exists in game so the primary mechanism to facilitate the triggers already exists.
Quests or even informative conversation paths do not need to be *randomly* different. They can vary based on attributes (an intelligent character may get more information but a strong character might get a quest to move boulders that the smart one would not be offered), class, crafting skill, harvesting skill, faction, deity, goodness or evilness or the like. Twenty years ago I played a few single player RPGs that did things like that.
dorotea said:Quests or even informative conversation paths do not need to be *randomly* different. They can vary based on attributes (an intelligent character may get more information but a strong character might get a quest to move boulders that the smart one would not be offered), class, crafting skill, harvesting skill, faction, deity, goodness or evilness or the like. Twenty years ago I played a few single player RPGs that did things like that.
Those are good examples of possible relationship flags that could influence the quest/conversation generation.
It could be a mini-game to see which combinations of flags could increase your chance to get a specific quest that had a highly desired reward, if succesful.
They addressed some of this back in one of their perception streams at least in terms of quests. They showed off how a quest could branch out depending on the choices made in the quest, and if the player had met certain conditions (and that those conditions could exceed just being X level or X level of perception). It is highly doubtful that any two members of the same class, same race, or same race/class will have the same adventure depending on what true/false conditions are implemented within the perception system.
With that being said I highly doubt that the main storyline will have much deviation if that is a route they decide to take. Personally I hope they don't do an intensive main storyline because the world should be the best way of communicating what is going on and what has happened in the past. I also hope they don't go the route of the solo epic hero trope. I really don't want to hear how special my character vs everyone else because they were blessed. Instead, I want to hear about how multiple people are continuing to develop whatever special power/characteristic is in question and how scholars aren't really sure why this is the case but there are various rumors circulating and only time will tell. The solo hero trope really has no place in an MMO in my opinion because an MMO should come from the standpoint of WE mentality vs I mentality.
Also on the note of guides Joppa stated that they were aware of these issues and that hopefully there would be so many unknown/hidden conditions built into the perception system that it would be near impossible to fish out all the information and put it into an extremely accurate and detailed guide.
I like the idea, but the problem is as AstralEcho stated, blatant pattern recognition is an issue. If you add a randomization method to the say a quest that requires you to collect 'X' amount of 'X' items then you need both the code which takes time and also you need ALOT of variety in items, moreso than maybe what is worth doing this approach. For this to be a meaningful change it needs to be a large change applicable to majority of quests that would benefit from such a design, otherwise it will come off as tacked on and a cheap gimick. But if done properly would require enough variety in items as to avoid blatant pattern recognition otherwise you run into the issue in the last sentence AND is potentially more detrimental if implemented poorly (AKA they add in a randomness that chooses between 5 differing item types) due to the fact that now there is a pattern recognition that is probably more easier to solve BY looking up a WIKI, which ends up defeating the point of this system in the first place.
philo said:I agree with Nytman,
The idea discourages people from working together to do the same quest. I don't see them implementing anything like this. Not to mention any new ideas would have to start from scratch and we are years beyond the point of having time for that.
If "quests" are small tasks to go kill X of Y, collect N of M, etc., then sure, collecting a group to tear through a quest hub's content is a good thing. But those aren't quests in any historical sense of the word. They're just tasks, work-for-hire. They're also super tedious and boring but people do them anyway because 90% of your XP gains come from turning them in.
This has a simple solution: Don't offer XP for turning in a quest/task. The XP should come from what you did along the way: killing stuff, discovering stuff, etc. The turn-in might offer some other kind of reward such as faction or a few coins.
Now, if PRotF has these little tasks then sure, gather a group to complete them if you like. But please, let's also do away with "quest hubs" that guide you across the world in a set pattern. These tasks can be pseudo-random and still doable by a group if the tasks are made shareable. The XP and monetary rewards would be divided amongst the group, though (faction gains should probably be the same for everyone).
Quests -- real quests -- should be more involved than just wandering a few yards to kill some nuisance mob and return. They don't all have to be EQ epic weapon level complex but there should be multiple steps. Those steps can be randomized, too, and still be doable by a group since the group gets XP from *doing* the quest, not from turning it in.
If we think outside the box a bit we can easily walk away from the themepark not-quests and quest hubs and still have a great social experience.
fazool said:TITLE RAN OUT OF CHARACTERS
The changes could be small like a changing turn-in item or a turn-in locaiton or and turn-in NPC or a phrase to trigger something or even a mob to kill.
Idea 2) Developers create tools that help them create the worlds. Like placing vegetation, rocks, etc. And there is some "AI" that goes into patterns of placing things. So, creating a tool can help tremendously.
So a tool that is a quest manager, like a terrain populator tool.
A tool that looks at a quest and applies some AI to it, so maybe a dark elf isn't assigned the killing of a dark elf but someone else.
Idea 1 seems a bit frustrating and pointless.. My NCP was there to turn in but now he is on the other side of town, why? It's still the same quest with the same tenants. This will be easily dissected online or with online tools too, if a turn in PC isn't at X he is at X... so it really wouldn't defeat the purpose of making it different so online people can't make guides.
Idea 2 like a random dungeon generator that has been in zillions of games such as Path of Exile and Diablo? Meh, could work - they already have faction-based quests I'm sure. I don't see how this is a new or revolutionary idea. I don't know what could change to make MMORPG a unique experience for each player. There are just too many people doing the same thing and with the same goals. They could implement 1 time quests or randomly generated quests that one person can grab... I guess, but the main quests and story line will still be linear, do X to get X.
I'm not sure how these ideas are groundbreaking though.
