A method to approach open world dungeon resetting.
Hartwell
Member, Leader of Men, Kickstarter, Alpha One
This is related to the question thread Dungeon Spawn Rate. Where as that thread is a question on spawn rates, this one is more of a discussion on a potential approach that the game could take with dungeons.
Traditional approaches to open world content isn't very well suited for open world dungeons. Usually MMORPG games will have a world boss spawns and do no world dungeons at all. Why bother create an obstacle that players could just camp or log out at to check for a respawn? This kind of leaves Ashes of Creation in an experimental position. Another game that I feel was rather experimental with open world content is the game Guild Wars 2, particularly Guild Wars 2: Heart of Thorns. They had several zones throughout the world added in that expansion. Each one had their own unique approach to world bosses. One of those zones was like a traditional instanced dungeon, but open world in the context of how the game treated open world content.
Dragon's Stand is a zone where all players are rounded up in a single camp, waiting for their opportunity to strike the stronghold of an elder dragon. When that opportunity comes, because enough time has passed and enough players are present to trigger the opening event, the next step begins. Three lanes open up. Each lane requires enough players to participate in, so that the defenses of the final boss are weakened. Once weakened, all players in the zone then meet up at the final boss to defeat them. After the boss is defeated, players have a little bit of time to gather all of their rewards, talk with NPCs and get out of the zone before it needs to close. People cannot spawn camp this boss and players must work together from start to finish to reach their goals. All obstacles, including the sieged gates that were blown up, get reset for the next time players wished to encounter it.
This is basically how a normal dungeon and raid instances work in many MMORPGs today, but it allows for all players to join in. Players start at the entrance, end at the final boss, and eventually the dungeon gets reset. Would something along the lines of this approach work for AoC if it had a longer reset timer?
Traditional approaches to open world content isn't very well suited for open world dungeons. Usually MMORPG games will have a world boss spawns and do no world dungeons at all. Why bother create an obstacle that players could just camp or log out at to check for a respawn? This kind of leaves Ashes of Creation in an experimental position. Another game that I feel was rather experimental with open world content is the game Guild Wars 2, particularly Guild Wars 2: Heart of Thorns. They had several zones throughout the world added in that expansion. Each one had their own unique approach to world bosses. One of those zones was like a traditional instanced dungeon, but open world in the context of how the game treated open world content.
Dragon's Stand is a zone where all players are rounded up in a single camp, waiting for their opportunity to strike the stronghold of an elder dragon. When that opportunity comes, because enough time has passed and enough players are present to trigger the opening event, the next step begins. Three lanes open up. Each lane requires enough players to participate in, so that the defenses of the final boss are weakened. Once weakened, all players in the zone then meet up at the final boss to defeat them. After the boss is defeated, players have a little bit of time to gather all of their rewards, talk with NPCs and get out of the zone before it needs to close. People cannot spawn camp this boss and players must work together from start to finish to reach their goals. All obstacles, including the sieged gates that were blown up, get reset for the next time players wished to encounter it.
This is basically how a normal dungeon and raid instances work in many MMORPGs today, but it allows for all players to join in. Players start at the entrance, end at the final boss, and eventually the dungeon gets reset. Would something along the lines of this approach work for AoC if it had a longer reset timer?
1
Comments
1. If you defeat the last boss in the dungeon, you then have 1min to loot it. After that 1min you get teleported to the beginning of the dungeon. (This is to prevent groups from spawn camping bosses)
2. If there are multiple boss monster in a dungeon, then increase their respawn timer with the amount of groups inside of the dungeon.
On top of that, in GW2 you can fast travel pretty much anywhere you like. Once you visit a Waypoint you can travel to it at any time unless it is contested. This means even if you are on the other side of the map when a meta event starts you can still get there in time to participate. In the case of Dragon's Stand, the waypoints have another function of controlling the flow of gameplay. As noted in the OP, when the meta event starts, players that enter the zone are put at the beginning of the map and have to work their way through it. There are many Waypoints in the zone but apart from the one at the beginning of the map, these are all locked off when the meta event starts. As the players progress through the map, they unlock more Waypoints allowing joining players to jump in and quickly get to where the action is rather than having to run through from the beginning.
Since Ashes won't have any kind of fast travel it will be hard for players to participate in events like this.
Spawn camping (which seems to be the issue people want to prevent) is only really an issue when bosses are needed for quests, or if they drop specific desired items.
The first of those is simple - just don't make quests that require boss mobs. There are other things that can be done.
The best way to deal with the second issue seems to me as if it would be to not have bosses with unique loot, but rather have all bosses in the dungeon share the same loot table that consists of materials for crafting gear.
If all bosses have the same loot table, there is no value in waiting for one specific boss to respawn. You are better off moving on to the next boss. The thing with open dungeons as opposed to instanced dungeons is that they don't really need to have a "final" boss - although I would love to see a single boss instance at the end of each wing of a dungeon, seems as good a way as any to make use of the limited instancing that they have given themselves room to use.
The question now becomes one of making sure the dungeon is large enough to hold 3 or 4 full groups who are all going from one boss to the next.
That's what I'd do, at least.
While the example from GW2 does sound interesting, I don't really see it being feasable with the dynamic content style Ashes should have. It sounds to me like it could make a really good event though.
I'm not sure about this approach. I suppose it would depend upon boss difficulty. If all bosses dropped the same loot, but some were harder than others, people could cherry pick the easy ones and dance around the rest. I do think a global loot table is the answer for the overpowered items that cause groups to skip bosses just to get straight to the good stuff. I'm not certain if the best solution is to make all bosses share the same loot table for everything, but I do think you've got a point. That would be a more liberal approach towards dungeons.
To be fair, this is because those games also make good use of instances (both single boss instances usually at the end of open dungeons and instanced dungeons). Without exception, the hard content is in an instance of some form.
This remains my hope for Ashes - open dungeons with a fairly large initial area, opening up to multiple wings, each wing large enough to support two full groups simultaneously, and each wing having access to a single boss instance with a 20 hour lockout.
With this, they could put the entirety of the dungeon on the same loot table, give the boss in the instances better loot, and have players leaving the instance end up out in the open world, a good way away from the dungeon entrance (possibly randomized like respawn locations will be to prevent camping players here).
My theory is that this would see most groups that head to the dungeon running through one wing, spending a bit of time farming bosses as they respawn, then after a while they will go for the real prize, kill the biss in the instance and then have no real need to go back to that zone for the day (though could if they wanted).
I know this isn't the only system that would do this, and I know there would he anlot of people that would dislike this kind of thing, but to me it makes sense with the restrictions and requirements specific to Ashes.
Withings it there could be let's say 10 room where could randomly rotate 5 bosses and 4 army and 1 empty room.
And let's add that the pass that need to be open could be clues to what archetype are very helpful for the boss. Like the fastest way to a room is throw 4 pass that determine half of you group.
I expect the dungeons will be a somewhat smaller affair, able to be tackled with a single party of 8 for the most part. And I expect that all of the suggestions in this thread could work to some degree.
I think the major issue is making the dungeons scalable to the number of attending players. At non-peak hours, a less popular dungeon might never see more than 1 party of players at a time. But that same dungeon has to be able to hold 10+ parties in case it becomes extremely popular. (Popularity could vary based on rewards, time of day, community/node migration, level restrictions, random events, etc.) With full instancing that's not a problem, but that's not what Intrepid wants for the most part.
Personally, I think that problem should only be partially solved. By which I mean, I think dungeons should support multiple parties, but only to a point. It's fine if some dungeons are overcrowded and there is some competition for entry into the dungeon or for kill credit/loot. To me, that makes sense for an immersive open-world design, even if it's frustrating at times. More importantly, that competition/scarcity forces the community to spread out and find other dungeons to run, rather than grouping around just the most convenient/grind-able ones.
Unfortunately, as much as I'd like to add my own proposed solution for the multi-party problem, I can't think of anything specific that hasn't been suggested already. Just sharing my thoughts on the problem...
Each non-instanced zone (dungeon or overland) has a player cap assigned to it, and any players attempting to zone in to that zone if the cap is reached or surpassed will simply be placed in a parallel version of the zone.
The developers may design a dungeon and think that it is a good fit for 4 active groups of 8, but as they are happy for a bit of competition to be had, they could set the player number of this zone to 40 (5 groups of 8). Should that zone then have 40 people in it, and another group (or even just another player) wants to enter that zone, then they will be in the second version of that zone.
If you are in that zone, and invite another player in to your group that is in a different version of that same zone, then it is a simple action (right clicking in the group window in EQ2) to select to be moved to the same version of the zone as another party member. This is why the target number of players for the zone can sometimes be surpassed.
This is by no means the only way to deal with this specific issue - but as Intrepid do have a good number of ex-SoE/Daybreak employees, I wouldn't be surprised if a slightly improved version of this is the option they take.
If the transition between instances is mostly seamless, it can still meet the goals of being immersive and open-world-like. But is it true enough to the sandbox vision? From a purist perspective, you should never be able to create new resources or zones by spawning extra instances. There should only be one true version of the world (per server).
I honestly don't know if AoC should go that far or not. I know at least that I could enjoy the game either way.
Purity is always trumped by pragmatism.
Nothing in an MMO is more important than making sure players have content to run.
With the way Ashes content is being delivered, there will likely be times early on in the games life where the servers are full, but there is only 1 dungeon active on a given server due to node status.
If you have 5% of the servers online characters wanting to run a dungeon, that is 500 players in one dungeon.
That will lead to amazingly negative first month write ups, which in turn will put Intrepid in damage control mode right from the start.
The way I suggested isn't the only way to do it, I'm quite sure (phasing is another option, or a combination ofthe two). All I know is, something with a similar end result will be done.
Ah, yeah I hadn't considered how much worse the problem is likely to be during the first couple of weeks. So instancing will probably be necessary for a while at least. Or rather sharding (like from the WoW Classic launch), where whole zones are duplicated. They just need to sync the nodes between each shard and it should be fine. And if they have that, then maybe they won't need any dungeon-specific instancing.
I'm not 100%sure how this works in WoW (why would I go back to play a game that is less good than it originally was, and was originally subpar anyways?), but from what I've been told, it is basically Blizzard copying what EQ2 did 16 years ago. I know Blizzard limits it to 3, where EQ2 had no limit - if the zone was designed to hold 120 players and 1200 players were in it, there would be 10 versions of that zone. What I don't know is the transition between different versions/instances/shards (they all mean the same thing) of the same zone.
It's worth pointing out that it would be perfectly possible to implement it without the need for a zone transition (or loading screen), and so from a player perspective it should go totally unnoticed. The only time players would need to be aware it is there is in the edge case where you invite someone to your group that is already in the zone.
What ever they come up with wouldn't just need to be for the first few weeks, either. It would need to persist.
There is no guarantee that some alliance hell bent on destruction won't find a way to destroy nodes faster than players on that server can rebuild them - so a server may well end up with a dearth of dungeons years after launch.
For the love of all that is good, no. The last thing we need is for game play to devolve into a routine mini-game of players being conditioned to passively wait for announcements, then descend upon bosses in droves.
I'd much rather people simply discover rare monsters through active exploration and combat them by actively organising themselves and others.
I guess 7days before it occurs to allow those who like to participating, to move at the location of the event.
https://ashesofcreation.wiki/Bosses
That notification only applies to legendary world bosses (and other things that affect nodes, generally speaking), not to all world bosses.
We have no way of knowing what makes a world boss legendary, but if they are sending notifications, I think it's safe to say they would be exceedingly rare events.
I'll guess that legendary one will be those who attack metropolis.
I can't imagine with the world mechanism other bosses.
Regional boss that destroy all nodes of the region if not stop?
It would be harder havingthe base population of an area completely changing than it would be to have one boss mob change.
Not all overland bosses need to go on a destructive rampage, attacking nodes and such. Those kinds of things will be rare.
Most bosses will probably simply be found within the bounds of the area assigned to their group. If a particular node state says that a specific area is populated by lions, there is no reason a boss tier (group or raid) lion couldn't be included in that.
It might be much harder to accomplish this, but dungeons have some pretty significant built in weaknesses to the experience simply by being static places. Open world or instanced makes no real difference, once you go there once you've pretty much finished it off.
I think that cycling dungeons would help keep things interesting, but I think one thing I remembered hearing was something about node development coordinating with dungeon unlocking and their development. At least that was the plan.
I could imagine a necropolis taking place in a lvl 0 node that send armies to any near node 3+ until the boss is slayed.
But to me those boss need to be impactful or it won't fit AOC style.
The only way that could work would be using procedurally generated dungeons which comes with its own set of problems to deal with.
What we do know is that not every boss needs to have an impact on the world. If every boss did have an impact, then either we will constantly just be defending our node and the game would suck, or there would be nothing but trash mobs out in the world, and the game would suck.
There needs to be bosses out in the world generally minding their own business so that we can come along and ruin their day, and take their stuff.
I have to agree that boss in raid that will be influenced by all 5 metropolis will probably be considered as Legendary.
Have it reset once a month or every couple weeks, and shuffle it around.
The main issue with this type of content is how much more limited you are in your ability to tell a story, and how much more work you need to go to in order to achieve that more limited ability to tell a story than you are with hand crafted content.
A more Ashes specific issue is that since the game is essentially going to have procedurally generated populations, if that were added on top of procedurally generated dungeon layouts, the chances of conflicts or issues arising is multiplied.
Procedurally generated content as your main dungeon content type is an ok system if you need a smaller amount of highly repetitive content that has no intricate storytelling within the content itself, which is why it works well enough in ARPG's that are traditionally less about storytelling and world building than an MMORPG is.
What I would love to see in an MMO one day though (and Steven may be the only person on the planet in the position to make it happen, whom also has the testicular fortitude to give it a go) is a zone where procedurally generated content is a mechanic specifically and intentionally used to mimic disorientation.
A simple example of this could be the lore based version of Skyrim's Labyrinthian (not the version that was in the game).
According to lore, that zone should have been a twisting, winding maze where getting lost is as much of a risk to life as anything else that could be found there. The way the game was designed though, the zone was presented as a linear dungeon.
But what would be cool is if a zone with that backstory existed in Ashes, and in order to mimic the sense of being completely lost, the zone was procedurally generated each time a group or player entered it. It would work best in this specific situation if it was an instanced single group zone - but it could make a memorable content piece of done properly.
Another example of a similar thing with a different execution would be a wide open area like a desert (or an underwater area, tundra, prairie). Instead of actually creating a zone, the zone exists on the server as a massive pile of "chunks". The chunks are placed when and only when a character needs them, and are placed randomly from a list of possible chunks. When a specific chunk is no longer needed, it is removed from the zone so that when a character does get close enough to needing it, the server generates a new random chunk to place there.
This is an area that could well be left as an open world area so you can run in to other players (reminds me of a Playstation game whose name escapes me right now). However, if you travel far enough from a feature such as a ruins or oasis that you came across, and turn back to try and find it again, you won't be able to as it won't be there any more.
This would have the effect of players essentially exploring a massive area, knowing that anything they come across they are likely the first people to have ever seen. But because the area is supposed to be uncharted (unchartable even), the fact that you can't ever find a thing you came across previously makes perfect sense in the context of things.